1 <?xml version="1.0" encoding="ISO-8859-1"?>
2 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
3 "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
6 ##########################################################################
9 Be sure that <h1> tags are in the leftmost column so that the awk
10 postprocessor can elide the Contents section - HTMLDOC will insert
12 ##########################################################################
15 <html xmlns="http://www.w3.org/1999/xhtml">
17 <title>The Fetchmail FAQ</title>
18 <meta name="description"
19 content="Frequently asked questions about fetchmail."/>
20 <meta name="keywords" content="fetchmail, POP, POP2, POP3, IMAP, remote mail"/>
23 <table width="100%" cellpadding="0" summary="Canned page footer">
25 <td width="30%">Back to <a href="index.html">Fetchmail Home
27 <td width="30%" align="right">$Date$</td>
32 <h1 id="FAQ">Frequently Asked Questions About Fetchmail</h1>
34 <p>Before reporting any bug, please read <a href="#G3">G3</a> for
35 advice on how to include diagnostic information that will get your
36 bug fixed as quickly as possible.</p>
38 <p>Note that this FAQ is occasionally updated from the SVN repository
39 and speaks in the past tense ("since") about a fetchmail release that is
40 not yet available. Please try a release candidate for that version in
41 case you need the new option.</p>
43 <p>If you have a question or answer you think ought to be added to
44 this FAQ list, file it to one of the trackers at <a
45 href="http://developer.berlios.de/projects/fetchmail/">our BerliOS
46 project site</a> or post to one of the fetchmail mailing lists (see
49 <h1 id="Contents">Contents</h1>
51 <a href="#Contentdetail">Detailed Contents</a><br/>
52 <a href="#C_G">G. General problems</a><br/>
53 <a href="#C_B">B. Build-time problems</a><br/>
54 <a href="#C_F">F. Fetchmail configuration file grammar questions</a><br/>
55 <a href="#C_C">C. Configuration questions</a><br/>
56 <a href="#C_T">T. How to make fetchmail play nice with various MTAs</a><br/>
57 <a href="#C_S">S. How to make fetchmail work with various servers</a><br/>
58 <a href="#C_I">I. How to fetchmail work with specific ISPs</a><br/>
59 <a href="#C_K">K. How to set up well-known security and authentication</a><br/>
60 <a href="#C_R">R. Runtime fatal errors</a><br/>
61 <a href="#C_H">H. Hangs and lockups</a><br/>
62 <a href="#C_D">D. Disappearing mail</a><br/>
63 <a href="#C_M">M. Multidrop-mode problems</a><br/>
64 <a href="#C_X">X. Mangled mail</a><br/>
65 <a href="#C_O">O. Other problems</a><br/>
67 <h1 id="Contentdetail">Detailed Contents</h1>
69 <h2 id="C_G">General problems</h2>
71 <a href="#G1">G1. What is fetchmail and why should I bother?</a><br/>
72 <a href="#G2">G2. Where do I find the latest FAQ and fetchmail sources?</a><br/>
73 <a href="#G3">G3. I think I've found a bug. Will you fix it?</a><br/>
74 <a href="#G4">G4. I have this idea for a neat feature. Will you add it?</a><br/>
75 <a href="#G5">G5. I want to make fetchmail behave like Outlook Express.</a><br/>
76 <a href="#G6">G6. Is there a mailing list for exchanging tips?</a><br/>
77 <a href="#G7">G7. So, what's this I hear about a fetchmail paper?</a><br/>
78 <a href="#G8">G8. What is the best server to use with fetchmail?</a><br/>
79 <a href="#G9">G9. What is the best mail program to use with fetchmail?</a><br/>
80 <a href="#G10">G10. How can I avoid sending my password en clair?</a><br/>
81 <a href="#G11">G11. Is any special configuration needed to use a dynamic IP address?</a><br/>
82 <a href="#G12">G12. Is any special configuration needed to use firewalls?</a><br/>
83 <a href="#G13">G13. Is any special configuration needed to <em>send</em> mail?</a><br/>
84 <a href="#G14">G14. Is fetchmail Y2K-compliant?</a><br/>
85 <a href="#G15">G15. Is there a way in fetchmail to support disconnected IMAP mode?</a><br/>
86 <a href="#G16">G16. How will fetchmail perform under heavy loads?</a><br/>
89 <h2 id="C_B">Build-time problems</h2>
91 <a href="#B1"><strike>B1. Make coughs and dies when building on FreeBSD.</strike></a><br/>
92 <a href="#B2">B2. Lex bombs out while building the fetchmail lexer.</a><br/>
93 <a href="#B3">B3. I get link failures when I try to build fetchmail.</a><br/>
94 <a href="#B4">B4. I get build failures in the intl directory.</a><br/>
96 <h2 id="C_F">Fetchmail configuration file grammar questions</h2>
98 <a href="#F1">F1. Why does my old .fetchmailrc no longer work?</a><br/>
99 <a href="#F2">F2. The .fetchmailrc parser won't accept my all-numeric user name.</a><br/>
100 <a href="#F3">F3. The .fetchmailrc parser won't accept my host or username beginning with 'no'.</a><br/>
101 <a href="#F4">F4. I'm getting a 'parse error' message I don't understand.</a><br/>
103 <h2 id="C_C">Configuration questions</h2>
105 <a href="#C1">C1. Why do I need a .fetchmailrc when running as root
106 on my own machine?</a><br/>
107 <a href="#C2">C2. How can I arrange for a fetchmail daemon to get
108 killed when I log out?</a><br/>
109 <a href="#C3">C3. How do I know what interface and address to use
110 with --interface?</a><br/>
111 <a href="#C4">C4. How can I set up support for sendmail's anti-spam
113 <a href="#C5">C5. How can I poll some of my mailboxes more/less
114 often than others?</a><br/>
115 <a href="#C6">C6. Fetchmail works OK started up manually, but not
116 from an init script.</a><br/>
117 <a href="#C7">C7. How can I forward mail to another
121 <h2 id="C_T">How to make fetchmail play nice with various MTAs</h2>
123 <a href="#T1">T1. How can I use fetchmail with sendmail?</a><br/>
124 <a href="#T2">T2. How can I use fetchmail with qmail?</a><br/>
125 <a href="#T3">T3. How can I use fetchmail with exim?</a><br/>
126 <a href="#T4">T4. How can I use fetchmail with smail?</a><br/>
127 <a href="#T5">T5. How can I use fetchmail with SCO's MMDF?</a><br/>
128 <a href="#T6">T6. How can I use fetchmail with Lotus Notes?</a><br/>
129 <a href="#T7">T7. How can I use fetchmail with Courier IMAP?</a><br/>
130 <a href="#T8">T8. How can I use fetchmail with vbmailshield?</a><br/>
132 <h2 id="C_S">How to make fetchmail work with various servers</h2>
134 <a href="#S1"><strike>S1. How can I use fetchmail with qpopper?</strike></a><br/>
135 <a href="#S2">S2. How can I use fetchmail with Microsoft Exchange?</a><br/>
136 <a href="#S3">S3. How can I use fetchmail with HP OpenMail?</a><br/>
137 <a href="#S4">S4. How can I use fetchmail with Novell GroupWise?</a><br/>
138 <a href="#S5">S5. How can I use fetchmail with InterChange?</a><br/>
139 <a href="#S6">S6. How can I use fetchmail with MailMax?</a><br/>
140 <a href="#S7">S7. How can I use fetchmail with FTGate?</a><br/>
142 <h2 id="C_I">How to fetchmail work with specific ISPs</h2>
144 <a href="#I1">I1. How can I use fetchmail with Compuserve RPA?</a><br/>
145 <a href="#I2">I2. How can I use fetchmail with Demon Internet's SDPS?</a><br/>
146 <a href="#I3">I3. How can I use fetchmail with usa.net's servers?</a><br/>
147 <a href="#I4">I4. How can I use fetchmail with geocities POP3 servers?</a><br/>
148 <a href="#I5">I5. How can I use fetchmail with Hotmail or Lycos Webmail?</a><br/>
149 <a href="#I6">I6. How can I use fetchmail with MSN?</a><br/>
150 <a href="#I7">I7. How can I use fetchmail with SpryNet?</a><br/>
151 <a href="#I8">I8. How can I use fetchmail with comcast.net or other
152 Maillennium servers?</a><br/>
153 <a href="#I9">I9. How can I use fetchmail with GMail/Google Mail?</a><br/>
155 <h2 id="C_K">How to set up well-known security and authentication
158 <a href="#K1">K1. How can I use fetchmail with SOCKS?</a><br/>
159 <a href="#K2">K2. How can I use fetchmail with IPv6 and IPsec?</a><br/>
160 <a href="#K3">K3. How can I get fetchmail to work with ssh?</a><br/>
161 <a href="#K4">K4. What do I have to do to use the IMAP-GSS protocol?</a><br/>
162 <a href="#K5">K5. How can I use fetchmail with SSL?</a><br/>
163 <a href="#K6">K6. How can I tell fetchmail not to try TLS if the server
164 advertises it? Why does fetchmail use SSL even though not configured?</a><br/>
166 <h2 id="C_R">Runtime fatal errors</h2>
168 <a href="#R1">R1. Fetchmail isn't working, and -v shows 'SMTP
169 connect failed' messages.</a><br/>
170 <a href="#R2">R2. When I try to configure an MDA, fetchmail doesn't
172 <a href="#R3">R3. Fetchmail dumps core when given an invalid rc
174 <a href="#R4"><strike>R4. Fetchmail dumps core in -V mode, but operates
175 normally otherwise.</strike></a><br/>
176 <a href="#R5">R5. Running fetchmail in daemon mode doesn't
178 <a href="#R6">R6. Fetchmail randomly dies with socket errors.</a><br/>
179 <a href="#R7">R7. Fetchmail running as root stopped working after
180 an OS upgrade</a><br/>
181 <a href="#R8">R8. Fetchmail is timing out after fetching certain
182 messages but before deleting them</a><br/>
183 <a href="#R9">R9. Fetchmail is timing out during message fetches</a><br/>
184 <a href="#R10"><strike>R10. Fetchmail is dying with SIGPIPE.</strike></a><br/>
185 <a href="#R11">R11. My server is hanging or emitting errors on CAPA.</a><br/>
186 <a href="#R12">R12. Fetchmail isn't working and reports getaddrinfo
188 <a href="#R13">R13. What does "Interrupted system call" mean?</a>
190 <h2 id="C_H">Hangs and lockups</h2>
192 <a href="#H1">H1. Fetchmail hangs when used with pppd.</a><br/>
193 <a href="#H2">H2. Fetchmail hangs during the MAIL FROM
195 <a href="#H3">H3. Fetchmail hangs while fetching mail.</a><br/>
198 <h2 id="C_D">Disappearing mail</h2>
200 <a href="#D1">D1. I think I've set up fetchmail correctly, but I'm
201 not getting any mail.</a><br/>
202 <a href="#D2">D2. All my mail seems to disappear after a dropped
204 <a href="#D3">D3. Mail that was being fetched when I interrupted my
205 fetchmail seems to have been vanished.</a><br/>
208 <h2 id="C_M">Multidrop-mode problems</h2>
210 <a href="#M1">M1. I've declared local names, but all my multidrop
211 mail is going to root anyway.</a><br/>
212 <a href="#M2">M2. I can't seem to get fetchmail to route to a local
213 domain properly.</a><br/>
214 <a href="#M3">M3. I tried to run a mailing list using multidrop,
215 and I have a mail loop!</a><br/>
216 <a href="#M4"><strike>M4. My multidrop fetchmail seems to be having DNS
217 problems.</strike></a><br/>
218 <a href="#M5">M5. I'm seeing long DNS delays before each message is
220 <a href="#M6">M6. How do I get multidrop mode to work with
222 <a href="#M7">M7. Multidrop mode isn't parsing envelope addresses
223 from my Received headers as it should.</a><br/>
224 <a href="#M8">M8. Users are getting multiple copies of
228 <h2 id="C_X">Mangled mail</h2>
230 <a href="#X1">X1. Spurious blank lines are appearing in the headers
231 of fetched mail.</a><br/>
232 <a href="#X2">X2. My mail client can't see a Subject
234 <a href="#X3">X3. Messages containing "From" at start of line are
235 being split.</a><br/>
236 <a href="#X4">X4. My mail is being mangled in a new and different
238 <a href="#X5"><strike>X5. Using POP3, retrievals seems to be fetching too
239 much!</strike></a><br/>
240 <a href="#X6">X6. My mail attachments are being dropped or
242 <a href="#X7">X7. Some mail attachments are hanging
244 <a href="#X8">X8. A spurious ) is being appended to my
246 <a href="#X9">X9. Missing "Content-Transfer-Encoding" header
247 with Domino IMAP</a><br/>
248 <h2 id="C_O">Other problems</h2>
250 <a href="#O1">O1. The --logfile option doesn't work if the logfile
251 doesn't exist.</a><br/>
252 <a href="#O2">O2. Every time I get a POP or IMAP message the header
253 is dumped to all my terminal sessions.</a><br/>
254 <a href="#O3">O3. Does fetchmail reread its rc file every poll
256 <a href="#O4">O4. Why do deleted messages show up again when I take
257 a line hit while downloading?</a><br/>
258 <a href="#O5">O5. Why is fetched mail being logged with my name,
259 not the real From address?</a><br/>
260 <a href="#O6">O6. I'm seeing long sendmail delays or hangs near the
261 start of each poll cycle.</a><br/>
262 <a href="#O7">O7. Why doesn't fetchmail deliver mail in date-sorted
264 <a href="#O8">O8. I'm using pppd. Why isn't my monitor option
266 <a href="#O9">O9. Why does fetchmail keep retrieving the same
267 messages over and over?</a><br/>
268 <a href="#O10"><strike>O10. Why is the received date on all my messages the
269 same?</strike></a><br/>
270 <a href="#O11">O11. I keep getting messages that say "Repoll
271 immediately" in my logs.</a><br/>
272 <a href="#O12">O12. Fetchmail no longer expunges mail on a 451 SMTP response.</a><br/>
273 <a href="#O13">O13. I want timestamp information in my fetchmail logs.</a><br/>
274 <a href="#O14">O14. Fetchmail no longer deletes oversized mails with
276 <a href="#O15">O15. Fetchmail always retains the first message in the
278 <a href="#O16">O16. Why is the Fetchmail FAQ only available in
279 ISO-216 A4 format? How do I get the FAQ in Letter
283 <h1 id="G">General problems</h1>
284 <h2><a id="G1" name="G1">G1. What is fetchmail and why should I
287 <p>Fetchmail is a one-stop solution to the remote mail retrieval
288 problem for Unix machines, quite useful to anyone with an
289 intermittent or dynamic-IP connection to a remote mailserver, SLIP or
290 PPP dialup, or leased line when SMTP isn't desired. Fetchmail can
291 collect mail using any variant of POP or IMAP and forwards to a the
292 local SMTP (via TCP socket) or LMTP (via TCP or Unix socket) listener or
293 into an MDA program, enabling all the normal
294 forwarding/filtering/aliasing mechanisms that would apply to local mail
295 or mail arriving via a full-time TCP/IP connection.</p>
297 <p>Fetchmail is not a toy or a coder's learning exercise, but an
298 industrial-strength tool capable of transparently handling every
299 retrieval demand from those of a simple single-user ISP connection
300 up to mail retrieval and rerouting for an entire client domain.
301 Fetchmail is easy to configure, unobtrusive in operation, powerful,
302 feature-rich, and well documented.</p>
304 <p>Fetchmail is <a href="http://www.opensource.org/">Open Source</a>
305 Software. The openness of the sources enables you to review and
306 customize the code, and contribute your changes.</p>
308 <p>A former fetchmail maintainer once claimed that Open Source software
309 were the strongest quality assurance, but the current maintainers do not
310 believe that open source alone is a criterion for quality – <a
311 href="fetchmail-SA-2005-01.txt">the remotely exploitable POP3
312 vulnerability (CVE-2005-2335)</a> lingered undiscovered in
313 fetchmail's code for years, which is a hint that open source code does
314 not audit itself.</p>
316 <p>Fetchmail is licensed under the <a
317 href="http://www.gnu.org/copyleft/gpl.html">GNU General Public
320 <p>If you found this FAQ in the distribution, see the README for
321 fetchmail's full feature list.</p>
323 <h2><a id="G2" name="G2">G2. Where do I find the latest FAQ and
324 fetchmail sources?</a></h2>
326 <p>The latest HTML FAQ is available alongside the latest fetchmail
327 sources at the fetchmail home page: <a
328 href="http://www.fetchmail.info/">http://www.fetchmail.info/</a>.
329 You can also usually find both in the <a
330 href="http://www.ibiblio.org/pub/Linux/system/mail/pop/!INDEX.short.html">
331 POP mail tools directory on iBiblio</a>.</p>
333 <p>A text dump of this FAQ is included in the fetchmail
334 distribution. Because it freezes at distribution release time, it
335 may not be completely current.</p>
337 <h2><a id="G3" name="G3">G3. I think I've found a bug. Will you fix
340 <p>The first thing you should to is to upgrade to the newest version of
341 fetchmail, and then see if the problem reproduces. So you'll probably
342 save us both time if you upgrade and test with <a href="#G2">the latest
343 version</a> <em>before</em> sending in a bug report.</p>
345 <p>Bugs will be fixed, provided you include enough diagnostic information
346 for me to go on. Send bugs to <a
347 href="mailto:fetchmail-users@lists.berlios.de">fetchmail-users</a>.
348 When reporting bugs, please include the following:</p>
351 <li>Your operating system.</li>
353 <li>Your compiler version, if you built from source; otherwise, the
354 name and origin of the RPM or other binary package you
357 <li>A copy of your POP or IMAP server's greeting line.</li>
359 <li>The name and version of the SMTP listener or MDA you are
362 <li>Any command-line options you used.</li>
364 <li>The output of fetchmail -V called with whatever other
365 command-line options you used.</li>
367 <li><strong>The output of <kbd>fetchmail --nodetach -vvv --nosyslog</kbd> with whatever other command-line options you use routinely.</strong></li>
370 <p>If you have FTP access to your remote mail account, and you have
371 any suspicion that the bug was triggered by a particular message,
372 please include a copy of the message that triggered the bug.</p>
374 <p>If your bug is something that used to work but stopped working
375 when you upgraded, then you can help pin the bug down by trying <a
376 href="http://download.berlios.de/fetchmail/">intermediate versions
377 of fetchmail</a> until you identify the revision that broke your
378 feature. The smart way to do this is by binary search on the
379 version sequence. First, try the version halfway between your last
380 good one and the current one. If it works, the failure was
381 introduced in the upper half of the sequence; if it doesn't, the
382 failure was introduced in the lower half. Now bisect that half in
383 the same way. In a very few tries, you should be able to identify
384 the exact adjacent pair of versions between which your bug was
385 introduced – and with information like that, I can usually come up
386 with a fix very quickly.</p>
388 <p>Another useful thing you can do, if you're using POP3, is to
389 test for IMAP4 support on your mailserver using the autoprobe
390 function of fetchmailconf. If you have IMAP4, and fetchmailconf
391 doesn't tell you it's broken, switch immediately. POP3 is a weak,
392 poorly-designed protocol with chronic problems, and the later
393 versions after RFC1725 actually get worse rather than better.
394 Changing over to IMAP4 may well make your problem go away – and if
395 your ISP doesn't have IMAP4 support, bug them to supply it.</p>
397 <p>It is helpful if you include your .fetchmailrc file, but not
398 necessary unless your symptom seems to involve an error in
399 configuration parsing. If you do send in your .fetchmailrc, mask
400 the passwords first!</p>
402 <p>If fetchmail seems to run and fetch mail, but the headers look
403 mangled (that is, headers are missing or blank lines are inserted
404 in the headers) then read the FAQ items in section <a
405 href="#X1">X</a> before submitting a bug report. Pay special
406 attention to the item on <a href="#generic_mangling">diagnosing
407 mail mangling</a>. There are lots of ways for other programs in the
408 mail chain to screw up that look like fetchmail's fault, but you
409 may be able to fix these by tweaking your configuration.</p>
411 <p>A transcript of the failed session with "--nosyslog --nodetach -vvv"
412 (yes, that's <em>three</em> -v options, enabling debug mode) will almost
413 always be useful. It is very important that the transcript include your
414 POP/IMAP server's greeting line, so I can identify it in case of server
415 problems. This transcript will not reveal your passwords, which are
416 specially masked out precisely so transcripts can be passed around.</p>
418 <p>If you upgraded your fetchmail and something broke, you should
419 include session transcripts with "--nosyslog --nodetach -vvv" of both
420 the working and failing versions. Very often, the source of the problem
421 can instantly identified by looking at the differences in protocol
424 <p>If the bug involves a core dump or hang, a gdb stack trace is
425 good to have. (Bear in mind that you can attach gdb to a running
426 but hung process by giving the process ID as a second argument.)
427 You will need to reconfigure with:</p>
430 CFLAGS=-g LDFLAGS=" " ./configure
433 <p>Then rebuild in order to generate a version that can be
434 traced with a debugger such as gdb, dbx or idb.</p>
436 <p>Best of all is a mail file which, when fetched, will reproduce
437 the bug under the latest (current) version.</p>
439 <p>Any bug I can reproduce will usually get fixed quite quickly.
440 Bugs I can't reproduce are a crapshoot. If the solution isn't obvious
441 when I first look, it may evade me for a long time (or to put it another
442 way, fetchmail is well enough tested that the easy bugs have long since
443 been found). So if you want your bug fixed rapidly, it is not just
444 sufficient but <em>necessary</em> that you give me a way to
445 easily reproduce it.</p>
447 <h2><a id="G4" name="G4">G4. I have this idea for a neat feature.
448 Will you add it?</a></h2>
450 <p>If it's reasonable for fetchmail and cannot be solved with reasonable
451 effort outside of fetchmail, perhaps.</p>
453 <p>You can do spam filtering better with procmail or maildrop on
454 the server side and (if you're the server sysadmin) sendmail.cf
455 domain exclusions. If you really want fetchmail to do it from the
456 client side, use a <code>preconnect</code> command to call
457 <a href='http://mailfilter.sourceforge.net/'>mailfilter</a>.</p>
459 <p>You can do other policy things better with the
460 <code>mda</code> option and script wrappers around fetchmail. If
461 it's a prime-time-vs.-non-prime-time issue, ask yourself whether a
462 wrapper script called from crontab would do the job.</p>
464 <p>fetchmail's first job is transport though, and it should do this
465 well. If a feature would cause fetchmail to deteriorate in other
466 respects, the feature will probably not be added.</p>
468 <p>For reasons fetchmail doesn't have other commonly-requested
469 features (such as password encryption, or multiple concurrent polls
470 from the same instance of fetchmail) see <a
471 href="esrs-design-notes.html">ESR's design
472 notes</a>. Note that this document is partially obsoleted by the
473 <a href="design-notes.html">updated design notes.</a></p>
475 <h2><a id="G5" name="G5">G5. I want to make fetchmail behave like
476 Outlook Express.</a></h2>
478 <p>The second-most-requested feature for fetchmail, after
479 content-based filtering, is the ability to have it remove messages
480 from a maildrop after N days, typically to be used with the
481 <code>keep</code> option as a sort of poor man's newsgroup
482 facility. Microsoft's Outlook Express supports this.</p>
484 <p>This feature is not yet implemented. It may be at a future date,
485 spare time of developers permitting.</p>
487 <h2><a id="G6" name="G6">G6. Is there a mailing list for exchanging
490 <p>There is a fetchmail-users list
491 <fetchmail-users@lists.berlios.de>
492 for bug reports and people who want to discuss configuration issues of
493 fetchmail. It's a Mailman list, see <a
494 href="http://lists.berlios.de/mailman/listinfo/fetchmail-users">http://lists.berlios.de/mailman/listinfo/fetchmail-users</a>.</p>
495 <p>There is a fetchmail-devel list
496 <fetchmail-devel@lists.berlios.de> for people who want to discuss
497 fixes and improvements in fetchmail and help co-develop it. It's a
498 Mailman list, which you can sign up for at <a
499 href="http://lists.berlios.de/mailman/listinfo/fetchmail-devel">http://lists.berlios.de/mailman/listinfo/fetchmail-devel</a>.
500 There is also an announcements-only list,
501 <fetchmail-announce@lists.berlios.de>, which you can sign up for at <a
502 href="http://lists.berlios.de/mailman/listinfo/fetchmail-announce">http://lists.berlios.de/mailman/listinfo/fetchmail-announce</a>.</p>
504 <h2><a id="G7" name="G7">G7. So, what's this I hear about a
505 fetchmail paper?</a></h2>
507 <p>Eric S. Raymond also considered fetchmail development a sociological
508 experiment, an extended test to see if my theory about the critical
509 features of the Linux development model is correct.</p>
511 <p>He considers the experiment a success. He wrote a paper about it titled <a
512 href="http://www.catb.org/~esr/writings/cathedral.html">The
513 Cathedral and the Bazaar</a> which was first presented at Linux
514 Kongress '97 in Bavaria and very well received there. It was also
515 given at Atlanta Linux Expo, Linux Pro '97 in Warsaw, and the first
516 Perl Conference, at UniForum '98, and was the basis of an invited
517 presentation at Usenix '98. The folks at Netscape told ESR it helped
519 href="http://wp.netscape.com/newsref/pr/newsrelease558.html">give
520 away the source for Netscape Communicator</a>.</p>
522 <p>If you're reading a non-HTML dump of this FAQ, you can find the
523 paper on the Web with a search for that title.</p>
525 <h2><a id="G8" name="G8">G8. What is the best server to use with
528 <p>Fetchmail will work with any POP, IMAP, ETRN, or ODMR server
529 that conforms to the relevant standards/RFCs (and even some outright
530 broken ones like <a href="#S2">Microsoft Exchange</a> and <a
531 href="#S6">Novell GroupWise</a>). This doesn't mean it works equally
532 well with all, however. POP2 servers, and POP3 servers without UIDL,
533 limit fetchmail's capabilities in various ways described on the manual
536 <p>Most modern Unixes (and effectively all Linux/*BSD systems) come
537 with POP3 support preconfigured (but beware of the horribly broken
538 POP3 server mentioned in <a href="#D2">D2</a>). An increasing
539 minority also feature IMAP (you can detect IMAP support by using the
540 'Probe for supported protocols' function in the fetchmailconf
541 utility - unfortunately it does not detect SSL-wrapped variants).</p>
543 <p>If you have the option, we recommend using or installing an
544 IMAP4rev1 or UIDL- and TOP-capable POP3 server. IMAP enables some
545 significant performance optimizations.</p>
547 <p>Don't be fooled by NT/Exchange propaganda. M$ Exchange is just
548 plain broken (see item <a href="#S2">S2</a>) and NT cannot handle
549 the sustained load of a high-volume remote mail server. Even
550 Microsoft itself knows better than to try this; their own Hotmail
551 service runs over Solaris! For extended discussion, see John
552 Kirch's excellent <a href="http://unix-vs-nt.org/kirch/">white
553 paper</a> on Unix vs. NT performance.</p>
555 <p>A decent POP3/IMAP server that has recently become popular is <a
556 href="http://dovecot.org/">Dovecot</a>.</p>
558 <p>Avoid <a href="http://home.pages.de/~mandree/qmail-bugs.html">qmail,
559 it's broken and unmaintained.</a></p>
561 <h2><a id="G9" name="G9">G9. What is the best mail program to use
562 with fetchmail?</a></h2>
564 <p>Fetchmail will work with all popular <a href="#T1">mail
565 transport programs</a>. It also doesn't care which user agent you
566 use, and user agents are as a rule almost equally indifferent to
567 how mail is delivered into your system mailbox. So any of the
568 popular Unix mail agents – <a
569 href="http://www.instinct.org/elm/">elm</a>, <a
570 href="http://www.washington.edu/pine/">pine</a>, <a
571 href="http://www.cs.indiana.edu/docproject/mail/mh.html">mh</a>, or
572 <a href="http://www.mutt.org">mutt</a> – will work fine with
575 <p>All this having been said, I can't resist putting in a discreet
576 plug for <a href="http://www.mutt.org">mutt</a>. Mutt's interface
577 is only a little different from that of its now-moribund ancestor
578 elm, but its flexibility and excellent handling of MIME and PGP put it
579 in a class by itself. You won't need its built-in POP3 support, though.
582 <h2><a id="G10" name="G10">G10. How can I avoid sending my password
585 <p>Depending on what your mail server you are talking to, this
586 ranges from trivial to impossible. It may even be next to
589 <p>In general there is little point in trying to secure your fetchmail
590 transaction unless you trust the security of the server host you are
591 retrieving mail from. Your vulnerability is more likely to be an
592 insecure local network on the server end (e.g. to somebody with a
593 TCP/IP packet sniffer intercepting Ethernet traffic between the modem
594 concentrator or DSL POP you dial in to and the mailserver host).</p>
596 <p>Having realized this, you need to ask whether password
597 encryption alone will really address your security exposure. If you
598 think you might be snooped between server and client, it's better
599 to use end-to-end encryption such as GnuPG (see below) on your whole
600 mail stream so none of it can be read. One of the advantages of
601 fetchmail over conventional SMTP-push delivery is that you may be able
602 to arrange encryption by using ssh(1); see <a href="#K3">K3</a>.</p>
604 <p>Note that ssh is not a complete privacy solution either, as your
605 mail could have been snooped in transit to your POP server from
606 wherever it originated. For best security, agree with your
607 correspondents to use a tool such as <a
608 href="http://www.gnupg.org/">GnuPG</a> (Gnu Privacy Guard) or PGP
609 (Pretty Good Privacy).</p>
611 <p>If ssh/sshd isn't available, or you find it too complicated for
612 you to set up, password encryption will at least keep a malicious
613 cracker from deleting your mail, and require him to either tap your
614 connection continuously or crack root on the server in order to
617 <p>You can deduce what encryptions your mail server has available
618 by looking at the server greeting line (and, for IMAP, the response
619 to a CAPABILITY query). Do a <code>fetchmail -v</code> to see
620 these, or telnet direct to the server port (110 for POP3, 143 for
623 <p>If your mailserver is using IMAP 2000, it'll have CRAM-MD5
624 support built in. Fetchmail autodetects this; you can skip the rest
627 <p>The POP3 facility you are most likely to have available is APOP.
628 This is a POP3 feature supported by many servers (fetchmailconf's
629 autoprobe facility will detect it and tell you if you have it). If
630 you see something in the greeting line that looks like an
631 angle-bracket-enclosed Internet address with a numeric left-hand
632 part, that's an APOP challenge (it will vary each time you log in).
633 For some hosts, you need to register a secret on the host (using
634 <code>popauth(8)</code> or some program like that). Specify the
635 secret as your password in your .fetchmailrc; it will be used to
636 encrypt the current challenge, and the encrypted form will be sent
637 back the the server for verification. Note that APOP is no longer
638 considered secure since March 2007.</p>
640 <p>Alternatively, you may have Kerberos available. This may require
641 you to set up some magic files in your home directory on your
642 client machine, but means you can omit specifying any password at
645 <p>Fetchmail supports two different Kerberos schemes. One is a POP3
646 variant called KPOP; consult the documentation of your mail server
647 to see if you have it (one clue is the string "krb-IV" in the
648 greeting line on port 110). The other is an IMAP and POP3 facility
649 described by RFC1731 and RFC1734. You can tell if this one is
650 present by looking for AUTH=KERBEROS_V4 in the CAPABILITY
653 <p>If you are fetching mail from a CompuServe POP3 account, you can
654 use their RPA authentication. See <a href="#I1">I1</a> for details.
655 If you are fetching mail from
656 Microsoft Exchange using IMAP, you will be able to use NTLM.</p>
658 <p>Your POP3 server may have the RFC1938 OTP capability to use
659 one-time passwords (if it doesn't, you can get OTP patches for the
660 2.2 version of the Qualcomm popper from <a href="#cmetz">Craig
661 Metz</a>). To check this, look for the string "otp-" in the
662 greeting line. If you see it, and your fetchmail was built with
663 OPIE support compiled in (see the distribution INSTALL file),
664 fetchmail will detect it also. When using OTP, you will specify a
665 password but it will not be sent en clair.</p>
667 <p>You can get both POP3 and IMAP OTP patches from <a id="cmetz"
668 name="cmetz">Craig Metz</a> at <a
669 href="http://www.inner.net/opie">http://www.inner.net/opie</a>.</p>
671 <p>These patches use a SASL authentication method named "X-OTP"
672 because there is not currently a standard way to do this; fetchmail
673 also uses this method, so the two will interoperate happily. They
674 better, because this is how Craig gets his mail ;-)</p>
676 <p>Finally, you can use <a href="#K5">SSL</a> for complete
677 end-to-end encryption if you have an SSL-enabled mailserver.</p>
679 <h2><a id="G11" name="G11">G11. Is any special configuration needed
680 to use a dynamic IP address?</a></h2>
682 <p>Yes. In order to avoid giving indigestion to certain picky MTAs
683 (notably <a href="#T3">exim</a>), fetchmail always makes the RCPT TO
684 address it feeds the MTA a fully qualified one with a hostname
685 part. Normally it does this by appending @ and "localhost", but
686 when you are using Kerberos or ETRN mode it will append @ and your
687 machine's fully-qualified domain name (FQDN).</p>
689 <p>Appending the FQDN can create problems when fetchmail is running
690 in daemon mode and outlasts the dynamic IP address assignment your
691 client machine had when it started up.</p>
693 <p>Since the new IP address (looked up at RCPT TO interpretation
694 time) doesn't match the original, the most benign possible result
695 is that your MTA thinks it's seeing a relaying attempt and refuses.
696 More frequently, fetchmail will try to connect to a nonexistent
697 host address and time out. Worst case, you could up forwarding your
698 mail to the wrong machine!</p>
700 <p>Use the <code>smtpaddress</code> option to force the appended
701 hostname to one with a (fixed) IP address of 127.0.0.1 in your
702 <code>/etc/hosts</code>. (The name 'localhost' will usually work;
703 or you can use the IP address itself.)</p>
705 <p>Only one fetchmail option interacts directly with your IP
706 address, '<code>interface</code>'. This option can be used to set
707 the gateway device and restrict the IP address range fetchmail will
708 use. Such a restriction is sometimes useful for security reasons,
709 especially on multihomed sites. See <a href="#C3">C3</a>.</p>
711 <p>I recommend against trying to set up the <code>interface</code>
712 option when initially developing your poll configuration – it's
713 never necessary to do this just to get a link working. Get the link
714 working first, observe the actual address range you see on
715 connections, and add an <code>interface</code> option (if you need
718 <p>You can't use ETRN if you have a dynamic IP address (your ISP
719 changes your IP address occasionally, possibly with every connect).
720 You need to have your own registered domain and a definite IP
721 address registered for that domain. The server needs to be
722 configured to accept mail for your domain but then queue it to
723 forward to your machine. ETRN just tells to server to flush its
724 queue for your domain. Fetchmail doesn't actually get the mail in
727 <p>You can use On-Demand Mail Relay (ODMR) with a dynamic IP
728 address; that's what it was designed for, and it provides
729 capabilities very similar to ETRN. Unfortunately ODMR servers are
730 still not yet widely deployed, as of 2006.</p>
732 <p>If you're using a dynamic-IP configuration, one other
733 (non-fetchmail) problem you may run into with outgoing mail is that
734 some sites will bounce your email because the hostname you're giving
735 them isn't real (and doesn't match what they get doing a reverse
736 DNS on your dynamically-assigned IP address). If this happens, you
737 need to hack your sendmail so it masquerades as your host.
744 <p>in your <code>sendmail.cf</code> will work, or you can set</p>
747 MASQUERADE_AS(smarthost.here)
750 <p>in the m4 configuration and do a reconfigure. (In both cases,
751 replace <code>smarthost.here</code> with the actual name of your
752 mailhost.) See the <a
753 href="http://www.lege.com/sendmail-FAQ.txt">sendmail FAQ</a> for
756 <h2><a id="G12" name="G12">G12. Is any special configuration needed
757 to use firewalls?</a></h2>
759 <p>No. You can use fetchmail with SOCKS, the standard tool for
760 indirecting TCP/IP through a firewall. You can find out about
761 SOCKS, and download the SOCKS software including server and client
762 code, at the <a href="http://www.socks.nec.com/">SOCKS distribution
765 <p>The specific recipe for using fetchmail with a firewall is at <a
766 href="#K1">K1</a></p>
768 <h2><a id="G13" name="G13">G13. Is any special configuration needed
769 to <em>send</em> mail?</a></h2>
771 <p>A user asks: but how do we send mail out to the POP3 server? Do
772 I need to implement another tool or will fetchmail do this too?</p>
774 <p>Fetchmail only handles the receiving side. The sendmail or other
775 preinstalled MTA on your client machine will handle sending mail
776 automatically; it will ship mail that is submitted while the
777 connection is active, and put mail that is submitted while the
778 connection is inactive into the outgoing queue.</p>
780 <p>Normally, sendmail is also run periodically (every 15 minutes on
781 most Linux systems) in a mode that tries to ship all the mail in
782 the outgoing queue. If you have set up something like pppd to
783 automatically dial out when your kernel is called to open a TCP/IP
784 connection, this will ensure that the mail gets out.</p>
786 <h2><a id="G14" name="G14">G14. Is fetchmail
787 Y2K-compliant?</a></h2>
789 <p>Fetchmail is fully Y2K-compliant.</p>
791 <p>Fetchmail could theoretically have problems when the 32-bit
792 time_t counters roll over in 2038, but I doubt it. Timestamps
793 aren't used for anything but log entry generation. Anyway, if you
794 aren't running on a 64-bit machine by then, you'll deserve to
797 <h2><a id="G15" name="G15">G15. Is there a way in fetchmail to
798 support disconnected IMAP mode?</a></h2>
800 <p>No. Fetchmail is a mail transport agent, best understood as a
801 protocol gateway between POP3/IMAP servers and SMTP. Disconnected
802 operation requires an elaborate interactive client. It's a very
803 different problem.</p>
805 <h2><a id="G16" name="G16">G16. How will fetchmail perform under
806 heavy loads?</a></h2>
808 <p>Fetchmail streams message bodies line-by-line; the most core it
809 ever requires per message is enough memory to hold the RFC822
810 header, and that storage is freed when body processing begins. It
811 is, accordingly, quite economical in its use of memory. It will store
812 the UID or UIDL data in core however, which can become considerable if
813 you are keeping lots of messages on the server.</p>
815 <p>After startup time, a fetchmail running in daemon mode stats its
816 configuration file once per poll cycle to see whether it has
817 changed and should be rescanned. Other than that, a fetchmail in
818 normal operation doesn't touch the disk at all; that job is left up
819 to the MTA or MDA the fetchmail talks to.</p>
821 <p>Fetchmail's performance is usually bottlenecked by latency on
822 the POP server or (less often) on the TCP/IP link to the server.
823 This is not a problem readily solved by tuning fetchmail, or even
824 by buying more TCP/IP capacity (which tends to improve bandwidth
825 but not necessarily latency).</p>
828 <h1>Build-time problems</h1>
829 <h2><a id="B1" name="B1"><strike>B1. Make coughs and dies when building on
830 FreeBSD.</strike></a></h2>
832 <p style="font-style:italic;">As of release 6.3.0, fetchmail's
833 Makefile[.in] should work flawlessly with BSD's portable make used on
834 FreeBSD. With older releases, use GNU make (usually installed as
835 <code>gmake</code>; otherwise try <kbd>pkg_add -r gmake</kbd>).</p>
837 <h2><a id="B2" name="B2">B2. Lex bombs out while building the
838 fetchmail lexer.</a></h2>
840 <p>fetchmail 6.3.0 and newer ship with the lexer and parser in .c
841 formats, so you do not need to use lex unless you hacked the .l or .y
844 <p>fetchmail's lexer has been developed with GNU flex and uses some of
845 its specialties, so the lexer cannot be compiled with the lex tools
846 shipped by some UNIX vendors (HP, SGI, Sun).</p>
848 <h2><a id="B3" name="B3">B3. I get link failures when I try to
849 build fetchmail.</a></h2>
851 <p>If you get errors resembling these:</p>
854 mxget.o(.text+0x35): undefined referenceto '__res_search'
855 mxget.o(.text+0x99): undefined reference to '__dn_skipname'
856 mxget.o(.text+0x11c): undefined reference to '__dn_expand'
857 mxget.o(.text+0x187): undefined reference to '__dn_expand'
858 make: *** [fetchmail] Error 1
861 <p>then you must add "-lresolv" to the LOADLIBS line in your
862 Makefile once you have installed the 'bind' package.</p>
864 <p>If you get link errors involving <tt>dcgettext</tt>, like
868 rcfile_y.o: In function 'yyparse':
869 rcfile_y.o(.text+0x3aa): undefined reference to 'dcgettext__'
870 rcfile_y.o(.text+0x4f2): undefined reference to 'dcgettext__'
871 rcfile_y.o(.text+0x5ee): undefined reference to 'dcgettext__'
872 rcfile_y.o: In function 'yyerror':
873 rcfile_y.o(.text+0xc7c): undefined reference to 'dcgettext__'
874 rcfile_y.o(.text+0xcc8): undefined reference to 'dcgettext__'
875 rcfile_y.o(.text+0xdf9): more undefined references to 'dcgettext__' follow
878 <p>install an up to date version of GNU gettext, reconfigure and rebuild
879 fetchmail. If that does not help, reconfigure with '--disable-nls' added
880 to the "./configure" command and rebuild.</p>
882 <h2><a id="B4" name="B4">B4. I get build failures in the intl
885 <p>Reconfigure with <tt>--disable-nls</tt> and recompile.</p>
888 <h1>Fetchmail configuration file grammar questions</h1>
889 <h2><a id="F1" name="F1">F1. Why does my old .fetchmailrc file no
890 longer work?</a></h2>
892 <h3>If your file predates 6.3.0</h3>
894 <p>The <tt>netsec</tt> option was discontinued and needs to be
897 <h3>If your file predates 5.8.9</h3>
899 <p>If you were using ETRN mode, change your <tt>smtphost</tt>
900 option to a <tt>fetchdomains</tt> option.</p>
902 <h3>If your file predates 5.8.3</h3>
904 <p>The <tt>'via localhost'</tt> special case for use with ssh tunnelling is
905 gone. Use the <tt>%h</tt> feature of <tt>plugin</tt> instead.</p>
907 <h3>If your file predates 5.6.8</h3>
909 <p>In 5.6.8, the <tt>preauth</tt> keyword and option were changed
910 back to <tt>auth</tt>. The <tt>preauth</tt> synonym will still be
911 supported through a few more point releases.</p>
913 <h3>If your file predates 5.6.5</h3>
915 <p>The <tt>imap-gss</tt>, <tt>imap-k4</tt>, and <tt>imap-login</tt>
916 protocol types are gone. This is a result of a major re-factoring
917 of the authentication machinery; fetchmail can now use Kerberos V4
918 and GSSAPI not just with IMAP but with POP3 servers that have
919 RFC1734 support for the AUTH command.</p>
921 <p>When trying to identify you to an IMAP or POP mailserver,
922 fetchmail now first tries methods that don't require a password
923 (GSSAPI, KERBEROS_IV); then it looks for methods that mask your
924 password (CRAM-MD5, X-OTP); and only if it the server doesn't
925 support any of those will it ship your password en clair.</p>
927 <p>Setting the <tt>preauth</tt> option to any value other than
928 'password' will prevent from looking for a password in your
929 <tt>.netrc</tt> file or querying for it at startup time.</p>
931 <h3>If your file predates 5.1.0</h3>
933 <p>In 5.1.0, the <tt>auth</tt> keyword and option were changed to
934 <tt>preauth</tt>.</p>
936 <h3>If your file predates 4.5.5</h3>
938 <p>If the <code>dns</code> option is on (the default), you may need
939 to make sure that any hostname you specify (for mail hosts or for
940 an SMTP target) is a canonical fully-qualified hostname). In order
941 to avoid DNS overhead and complications, fetchmail no longer tries
942 to derive the fetchmail client machine's canonical DNS name at
945 <h3>If your file predates 4.0.6:</h3>
947 <p>Just after the '<code>via</code>' option was introduced, I
948 realized that the interactions between the '<code>via</code>',
949 '<code>aka</code>', and '<code>localdomains</code>' options were
950 out of control. Their behavior had become complex and confusing, so
951 much so that I was no longer sure I understood it myself. Users
952 were being unpleasantly surprised.</p>
954 <p>Rather than add more options or crock the code, I re-thought it.
955 The redesign simplified the code and made the options more
956 orthogonal, but may have broken some complex multidrop
959 <p>Any multidrop configurations that depended on the name just
960 after the '<code>poll</code>' or '<code>skip</code>' keyword being
961 still interpreted as a DNS name for address-matching purposes, even
962 in the presence of a '<code>via</code>' option, will break.</p>
964 <p>It is theoretically possible that other unusual configurations
965 (such as those using a non-FQDN poll name to generate Kerberos IV
966 tickets) might also break; the old behavior was sufficiently murky
967 that we can't be sure. If you think this has happened to you,
968 contact the maintainer.</p>
970 <h3>If your file predates 3.9.5:</h3>
972 <p>The '<code>remote</code>' keyword has been changed to
973 '<code>folder</code>'. If you try to use the old keyword, the
974 parser will utter a warning.</p>
976 <h3>If your file predates 3.9:</h3>
978 <p>It could be because you're using a .fetchmailrc that's written
979 in the old popclient syntax without an explicit
980 '<code>username</code>' keyword leading the first user entry
981 attached to a server entry.</p>
983 <p>This error can be triggered by having a user option such as
984 '<code>keep</code>' or '<code>fetchall</code>' before the first
985 explicit username. For example, if you write</p>
988 poll openmail protocol pop3
989 keep user "Hal DeVore" there is hdevore here
992 <p>the '<code>keep</code>' option will generate an entire user
993 entry with the default username (the name of fetchmail's invoking
996 <p>The popclient compatibility syntax was removed in 4.0. It
997 complicated the configuration file grammar and confused users.</p>
999 <h3>If your file predates 2.8:</h3>
1001 <p>The '<code>interface</code>', '<code>monitor</code>' and
1002 '<code>batchlimit</code>' options changed after 2.8.</p>
1004 <p>They used to be global options with '<code>set</code>' syntax
1005 like the batchlimit and logfile options. Now they're per-server
1006 options, like '<code>protocol</code>'.</p>
1008 <p>If you had something like</p>
1011 set interface = "sl0/10.0.2.15"
1014 <p>in your .fetchmailrc file, simply delete that line and insert
1015 'interface sl0/10.0.2.15' in the server options part of your
1016 'defaults' declaration.</p>
1018 <p>Do similarly for any '<code>monitor</code>' or
1019 '<code>batchlimit</code>' options.</p>
1021 <h2><a id="F2" name="F2">F2. The .fetchmailrc parser won't accept
1022 my all-numeric user name.</a></h2>
1024 <p>Either upgrade to a post-5.0.5 fetchmail or put string quotes
1027 <p>The configuration file parser in older fetchmail versions
1028 treated any all-numeric token as a number, which confused it when
1029 it was expecting a name. String quoting forces the token's
1032 <p>The lexical analyzer in 5.0.6 and beyond is smarter and assumes
1033 any token following "username" or "password" is a string.</p>
1035 <h2><a id="F3" name="F3">F3. The .fetchmailrc parser won't accept
1036 my host or username beginning with 'no'.</a></h2>
1038 <p>See <a href="#F2">F2</a>. You're caught in an unfortunate crack
1039 between the newer-style syntax for negated options ('no keep', 'no
1040 rewrite' etc.) and the older style run-on syntax ('nokeep',
1041 'norewrite' etc.).</p>
1043 <p>Upgrade to a 5.0.6 or later fetchmail, or put string quotes
1044 around your token.</p>
1046 <h2><a id="F4" name="F4">F4. I'm getting a 'parse error' message I
1047 don't understand.</a></h2>
1049 <p>The most common cause of mysterious parse errors is putting a
1050 server option after a user option. Check the manual page; you'll
1051 probably find that by moving one or more options closer to the
1052 'poll' keyword you can eliminate the problem.</p>
1054 <p>Yes, I know these ordering restrictions are hard to understand.
1055 Unfortunately, they're necessary in order to allow the 'defaults'
1056 feature to work.</p>
1059 <h1>Configuration questions</h1>
1060 <h2><a id="C1" name="C1">C1. Why do I need a .fetchmailrc when
1061 running as root on my own machine?</a></h2>
1063 <p>Ian T. Zimmerman <itz@rahul.net> asked:</p>
1065 <p>On the machine where I'm the only real user, I run fetchmail as
1066 root from a cron job, like this:</p>
1069 fetchmail -u "itz" -p POP3 -s bolero.rahul.net
1072 <p>This used to work as is (with no .fetchmailrc file in root's
1073 home directory) with the last version I had (1.7 or 1.8, I don't
1074 remember). But with 2.0, it RECPs all mail to the local root user,
1075 unless I create a .fetchmailrc in root's home directory
1079 skip bolero.rahul.net proto POP3
1083 <p>It won't work if the second line is just "<code>user
1084 itz</code>". This is silly.</p>
1086 <p>It seems fetchmail decides to RECP the 'default local user'
1087 (i.e. the uid running fetchmail) unless there are local aliases,
1088 and the 'default' aliases (itz->itz) don't count. They
1093 <p>No they shouldn't. I thought about this for a while, and I don't
1094 much like the conclusion I reached, but it's unavoidable. The
1095 problem is that fetchmail has no way to know, in general, that a
1096 local user 'itz' actually exists.</p>
1098 <p>"Ah!" you say, "Why doesn't it check the password file to see if
1099 the remote name matches a local one?" Well, there are two
1102 <p>One: it's not always possible. Suppose you have an SMTP host
1103 declared that's not the machine fetchmail is running on? You
1106 <p>Two: How do you know server itz and SMTP-host itz are the same
1107 person? They might not be, and fetchmail shouldn't assume they are
1108 unless local-itz can explicitly produce credentials to prove it
1109 (that is, the server-itz password in local-itz's .fetchmailrc
1112 <p>Once you start running down possible failure modes and thinking
1113 about ways to tinker with the mapping rules, you'll quickly find
1114 that all the alternatives to the present default are worse or
1115 unacceptably more complicated or both.</p>
1117 <h2><a id="C2" name="C2">C2. How can I arrange for a fetchmail
1118 daemon to get killed when I log out?</a></h2>
1120 <p>The easiest way to dispatch fetchmail on logout (which will work
1121 reliably only if you have just one login going at any time) is to
1122 arrange for the command 'fetchmail -q' to be called on logout.
1123 Under bash, you can arrange this by putting 'fetchmail -q' in the
1124 file '~/.bash_logout'. Most csh variants execute '~/.logout' on
1125 logout. For other shells, consult your shell manual page.</p>
1127 <p>Automatic startup/shutdown of fetchmail is a little harder to
1128 arrange if you may have multiple login sessions going. In the
1129 contrib subdirectory of the fetchmail distribution there is some
1130 shell code you can add to your .bash_login and .bash_logout
1131 profiles that will accomplish this. Thank James Laferriere
1132 <babydr@nwrain.net> for it.</p>
1134 <p>Some people start up and shut down fetchmail using the ppp-up
1135 and ppp-down scripts of pppd.</p>
1137 <h2><a id="C3" name="C3">C3. How do I know what interface and
1138 address to use with --interface?</a></h2>
1140 <p>This depends a lot on your local networking configuration (and
1141 right now you can't use it at all except under Linux and the newer
1142 BSDs). However, here are some important rules of thumb that can
1143 help. If they don't work, ask your local sysop or your Internet
1146 <p>First, you may not need to use --interface at all. If your
1147 machine only ever does SLIP or PPP to one provider, it's almost
1148 certainly by a point to point modem connection to your provider's
1149 local subnet that's pretty secure against snooping (unless someone
1150 can tap your phone or the provider's local subnet!). Under these
1151 circumstances, specifying an interface address is fairly
1154 <p>What the option is really for is sites that use more than one
1155 provider. Under these circumstances, typically one of your provider
1156 IP addresses is your mailserver (reachable fairly securely via the
1157 modem and provider's subnet) but the others might ship your packets
1158 (including your password) over unknown portions of the general
1159 Internet that could be vulnerable to snooping. What you'll use
1160 --interface for is to make sure your password only goes over the
1161 one secure link.</p>
1163 <p>To determine the device:</p>
1166 <li>If you're using a SLIP link, the correct device is probably
1169 <li>If you're using a PPP link, the correct device is probably
1172 <li>If you're using a direct connection over a local network such
1173 as an ethernet, use the command 'netstat -r' to look at your
1174 routing table. Try to match your mailserver name to a destination
1175 entry; if you don't see it in the first column, use the 'default'
1176 entry. The device name will be in the rightmost column.</li>
1179 <p>To determine the address and netmask:</p>
1182 <li>If you're talking to slirp, the correct address is probably
1183 10.0.2.15, with no netmask specified. (It's possible to configure
1184 slirp to present other addresses, but that's the default.)</li>
1186 <li>If you have a static IP address, run 'ifconfig <device>',
1187 where <device> is whichever one you've determined. Use the IP
1188 address given after "inet addr:". That is the IP address for your
1189 end of the link, and is what you need. You won't need to specify a
1192 <li>If you have a dynamic IP address, your connection IP will vary
1193 randomly over some given range (that is, some number of the least
1194 significant bits change from connection to connection). You need to
1195 declare an address with the variable bits zero and a complementary
1196 netmask that sets the range.</li>
1199 <p>To illustrate the rule for dynamic IP addresses, let's suppose
1200 you're hooked up via SLIP and your IP provider tells you that the
1201 dynamic address pool is 255 addresses ranging from 205.164.136.1 to
1202 205.164.136.255. Then</p>
1205 interface "sl0/205.164.136.0/255.255.255.0"
1208 <p>would work. To range over any value of the last two octets
1209 (65536 addresses) you would use</p>
1212 interface "sl0/205.164.0.0/255.255.0.0"
1215 <h2><a id="C4" name="C4">C4. How can I set up support for
1216 sendmail's anti-spam features?</a></h2>
1218 <p>This answer covers versions of sendmail from 8.9.3-20 (the
1219 version installed in Red Hat 6.2) upwards. If you have an older
1220 version, upgrade to sendmail 8.9.</p>
1222 <p>Stock sendmails can now do anti-spam exclusions based on a
1223 database of filter rules. The human-readable form of the database
1224 is at <tt>/etc/mail/access</tt>. The database itself is at
1225 <tt>/etc/mail/access.db</tt>.</p>
1227 <p>The table itself uses email addresses, domain names, and network
1228 numbers as keys. For example,</p>
1231 spammer@aol.com REJECT
1232 cyberspammer.com REJECT
1236 <p>would refuse mail from spammer@aol.com, any user from
1237 cyberspammer.com (or any host within the cyberspammer.com domain),
1238 and any host on the 192.168.212.* network. (This feature can be
1239 used to do other things as well; see the <a
1240 href="http://www.sendmail.org/m4/anti_spam.html">sendmail
1241 documentation</a> for details)</p>
1243 <p>To actually set up the database, run</p>
1246 makemap hash deny <deny
1249 <p>in /etc/mail.</p>
1251 <p>To test, send a message to your mailing address from that host
1252 and then pop off the message with fetchmail, using the -v argument.
1253 You can monitor the SMTP transaction, and when the FROM address is
1254 parsed, if sendmail sees that it is an address in spamlist,
1255 fetchmail will flush and delete it.</p>
1257 <p>Under no circumstances put your <strong>mailhost</strong> or
1258 <strong>any host you accept mail from</strong> using fetchmail into
1259 your reject file. You <strong>will</strong> lose mail if you do
1262 <h2><a id="C5" name="C5">C5. How can I poll some of my mailboxes
1263 more/less often than others?</a></h2>
1265 <p>Use the <cite>interval</cite> keyword on the ones that should be
1266 checked less often. For example, if you do a poll every 5 minutes,
1267 and want to poll some mailboxes every 5 minutes and some every 30
1268 minutes, use something like this:</p>
1271 poll mainsite.example.com proto pop3 user ....
1272 poll secondary.example.com proto pop3 interval 6 user ...
1275 <p>Then secondary.example.com will be polled every 6th time that
1276 mainsite.example.com is polled, which with a polling interval of
1277 every 5 minutes means that secondary.example.com will be polled
1278 every 30 minutes.</p>
1280 <h2><a id="C6" name="C6">Fetchmail works OK started up manually,
1281 but not from an init script.</a></h2>
1283 <p>Often, startup scripts have a different environment than an
1284 interactive login shell. For instance, $HOME might point to "/root"
1285 when you are logged in as root, but it might be either unset, or
1286 set to "/" when the startup scripts are running. That means
1287 fetchmail at startup can't find the .fetchmailrc.</p>
1289 <p>Pick a location (such as /etc/fetchmailrc) and use fetchmail's
1290 -f option to point fetchmail at it. That should solve the
1293 <h2><a id="C7" name="C7">C7. How can I forward mail to another
1296 <p>To forward mail to a host other than the one you are running
1297 fetchmail on, use the <code>smtphost</code> or
1298 <code>smtpname</code> option. See the manual page for details.</p>
1301 <h1>How to make fetchmail play nice with various MTAs</h1>
1302 <h2><a id="T1" name="T1">T1. How can I use fetchmail with
1305 <p>For most sendmails, no special configuration is required. Eric
1306 Allman tells me that if <code>FEATURE(always_add_domain)</code> is
1307 included in sendmail's configuration, you can leave the
1308 <code>rewrite</code> option off.</p>
1310 <p>If your sendmail complains "sendmail does not relay", make
1311 sure your sendmail.cf file says <code>Cwlocalhost</code> so that
1312 sendmail recognizes 'localhost' as a name of its host.</p>
1314 <p>If you're mailing from another machine on your local network,
1315 also ensure that its IP address is listed in ip_allow or name in
1316 name_allow (usually in /etc/mail/)</p>
1318 <p>If you find that your sendmail doesn't like the address
1319 'FETCHMAIL-DAEMON@localhost' (which is used in the bouncemail that
1320 fetchmail generates), you may have to set
1321 <code>FEATURE(accept_unqualified_senders)</code>.</p>
1323 <p>Günther Leber reports that Digital Unix sendmails won't
1324 work with fetchmail. The symptom is an error message "<code>553
1325 Local configuration error, hostname not recognized as
1326 local</code>". The problem is that fetchmail normally feeds
1327 sendmail with the client machine's host address in the MAIL FROM
1328 line. These sendmails think this means they're seeing the result of
1329 a mail loop and suppress the mail. You may be able to work around
1330 this by running in <code>--invisible</code> mode.</p>
1332 <p>If you want to support multidrop mode, and you can get access to
1333 your mailserver's sendmail.cf file, it's a good idea to add this
1337 H?l?Delivered-To: $h
1340 <p>This will cause the mailserver's sendmail to reliably write the
1341 appropriate envelope address into each message before fetchmail
1342 sees it, and tell fetchmail which header it is.  With this
1343 change, multidrop mode should work reliably even when the Received
1344 header omits the envelope address (which will typically be the case
1345 when the message has multiple recipients).  However it will
1346 still not distinguish the recipients, your only advantage is that
1347 no bounce will be sent if a message is BCC addressed to multiple
1348 users at your site.  To fix even that problem, you might want
1349 to try the following hack, which is however untested and quite
1353 H?J?Delivered-To: $u
1355 Mmdrop, P=/usr/bin/procmail, F=lsDFMqSPfhnu9J,
1356 S=EnvFromSMTP/HdrFromSMTP, R=EnvToSMTP/HdrToSMTP,
1357 T=DNS/RFC822/X-Unix,
1358 A=procmail -Y -a $u -d $h
1361 <p>For both hacks, you have to declare '<code>envelope
1362 "Delivered-To:"</code>' on the fetchmail side, to put the virtual
1363 domain (e.g. 'domain.com') with RELAY permission into your access
1364 file and to add a line reading '<code>domain.com
1365 local:local-pop-user</code>' for the first and '<code>domain.com
1366 mdrop:local-pop-user</code>' for the second hack to your
1369 <p>You will notice that if the mail already has a Delivered-To
1370 header, sendmail will not add another.  Further, editing
1371 sendmail.cf directly is not very comfortable.  Solutions for
1372 both problems can be found in Peter 'Rattacresh' Backes' 'hybrid'
1373 patch against sendmail.  Have a look at it, you can find it in
1374 the contrib subdirectory.</p>
1376 <p>Feel free to try Martijn Lievaart's detailed recipe in the
1377 contrib subdirectory of the fetchmail source distribution, it
1378 attempts to realize multidrop mailboxes with an external
1381 <p>If for some reason you are invoking sendmail via the
1382 <tt>mda</tt> option (rather than delivering to port 25 via smtp),
1383 don't forget to include the -i switch. Otherwise you will
1384 occasionally get mysterious delivery failures with a SIGPIPE as the
1385 sendmail instance dies. The problem is messages with a single dot
1386 at start of a text line.</p>
1388 <h2><a id="T2" name="T2">T2. How can I use fetchmail with
1391 <h3>qmail as your local SMTP server</h3>
1393 <p>Avoid <a href="http://home.pages.de/~mandree/qmail-bugs.html">qmail,
1394 it's broken and unmaintained.</a></p>
1396 <p>Turn on the <code>forcecr</code> option; qmail's listener mode
1397 doesn't like header or message lines terminated with bare
1399 (This information contributed by Robert de Bath
1400 <robert@mayday.cix.co.uk>.)</p>
1402 <h3>qmail as your ISP's POP3 server</h3>
1404 <p>Note that qmail's POP3 server, as of version 1.03 and netqmail 1.05,
1405 miscalculates the message sizes, so you may see size-related fetchmail
1408 <p>If a mailhost is using the qmail package, then it is usually possible
1409 to set up one fetchmail link to reliably collect the mail for an entire
1412 <p>One of the basic features of qmail is the 'Delivered-To:'
1413 message header. Whenever qmail delivers a message to a local
1414 mailbox it puts the username and hostname of the envelope recipient
1415 on this line. One major reason for this is to prevent mail
1416 loops, the other is to transport envelope information which is essential
1417 for multidrop (domain-in-a-mailbox) schemes.</p>
1419 <p>To set up qmail to batch mail for a disconnected site, the
1420 ISP-mailhost will have normally put that site in its 'virtualhosts'
1421 control file so it will add a prefix to all mail addresses for this
1422 site. This results in mail sent to
1423 'username@userhost.userdom.example.com' having a 'Delivered-To:' line
1427 Delivered-To: mbox-userstr-username@userhost.userdom.example.com
1430 <p>A single host maildrop will be slightly simpler:</p>
1433 Delivered-To: mbox-userstr-username@userhost.example.com
1436 <p>The ISP can make the 'mbox-userstr-' prefix anything they choose
1437 but a string matching the user host name is likely.</p>
1439 <p>To use this line you must:</p>
1442 <li>Ensure the option '<code>envelope "Delivered-To"</code>' is in the fetchmail
1445 <li>Ensure the option '<code>qvirtual "mbox-userstr-"</code>' is
1446 in the fetchmail config file, in order to remove this prefix from the
1447 username. (added by Luca Olivetti)</li>
1449 <li>Ensure you have a <code>localdomains</code> option containing
1450 '<code>userdom.example.com</code>' or '<code>userhost.userdom.example.com</code>'
1454 <h2><a id="T3" name="T3">T3. How can I use fetchmail with
1457 <p>If you have <code>rewrite</code> on:</p>
1459 <p>There is an RFC1123 requirement that MAIL FROM and RCPT TO
1460 addresses you pass to it have to be canonical (e.g. with a fully
1461 qualified hostname part). Therefore fetchmail tries to pass fully
1462 qualified RCPT TO addresses. But exim does not by default accept
1463 'localhost' as a fully qualified domain. This can be fixed.</p>
1465 <p>In exim.conf, add 'localhost' to your local_domains declaration
1466 if it's not already present. For example, the author's site at
1467 thyrsus.com would have a line reading:</p>
1470 local_domains = thyrsus.com:localhost
1473 <p>If you have <code>rewrite</code> off:</p>
1475 <p>MAIL FROM is a potential problem if the MTAs upstream from your
1476 fetchmail don't necessarily pass canonicalized From and Return-Path
1477 addresses, and fetchmail's <code>rewrite</code> option is off. The
1478 specific case where this has come up involves bounce messages
1479 generated by sendmail on your mailer host, which have the
1480 (un-canonicalized) origin address MAILER-DAEMON.</p>
1482 <p>The right way to fix this is to enable the <code>rewrite</code>
1483 option and have fetchmail canonicalize From and Return-Path
1484 addresses with the mailserver hostname before exim sees them. This
1485 option is enabled by default, so it won't be off unless you turned
1488 <p>If you must run with <code>rewrite</code> off, there is a switch
1489 in exim's configuration files that allows it to accept domainless
1490 MAIL FROM addresses; you will have to flip it by putting the
1494 sender_unqualified_hosts = localhost
1497 <p>in the main section of the exim configuration file. Note that
1498 this will result in such messages having an incorrect domain name
1499 attached to their return address (your SMTP listener's hostname
1500 rather than that of the remote mail server).</p>
1502 <h2><a id="T4" name="T4">T4. How can I use fetchmail with
1505 <p>Smail 3.2 is very nearly plug-compatible with sendmail, and may
1506 work fine out of the box.</p>
1508 <p>We have one report that when processing multiple messages from a
1509 single fetchmail session, smail sometimes delivers them in an order
1510 other than received-date order. This can be annoying because it
1511 scrambles conversational threads. This is not fetchmail's problem,
1512 it is an smail 'feature' and has been reported to the maintainers
1515 <p>Very recent smail versions require an
1516 <code>-smtp_hello_verify</code> option in the smail config file.
1517 This overrides smail's check to see that the HELO address is
1518 actually that of the client machine, which is never going to be the
1519 case when fetchmail is in the picture. According to RFC1123 an SMTP
1520 listener <em>must</em> allow this mismatch, so smail's new behavior
1521 (introduced sometime between 3.2.0.90 and 3.2.0.95) is a bug.</p>
1523 <p>You may also need to say
1524 <code>-smtp_hello_broken_allow=127.0.0.1</code> in order for smail
1525 to accept the "localhost" that fetchmail normally appends to
1526 recipient addresses.</p>
1528 <h2><a id="T5" name="T5">T5. How can I use fetchmail with SCO's
1531 <p>MMDF itself is difficult to configure, but it turns out that
1532 connecting fetchmail to MMDF's SMTP channel isn't that hard. You
1534 href="http://aplawrence.com/Unixart/uucptofetch.html">MMDF
1535 recipe</a> that describes replacing a UUCP link with fetchmail
1538 <h2><a id="T6" name="T6">T6. How can I use fetchmail with Lotus
1541 <p>The Lotus Notes SMTP gateway tries to deduce when it should
1542 convert \n to \r\n, but its rules are not the intuitive and
1543 correct-for-RFC822 ones. Use 'forcecr'.</p>
1545 <h2><a id="T7" name="T7">T7. How can I use fetchmail with Courier
1548 <p>The courier mta doesn't like RCPT addresses that look like
1549 <code>someone@localhost</code>. Work around this with an
1550 <code>smtphost</code> or <code>smtpaddress</code>.</p>
1552 <h2><a name="T8">T8. How can I use fetchmail with vbmailshield?</a></h2>
1554 <p>vbmailshield's SMTP interpreter is broken. It doesn't understand RSET.</p>
1556 <p>As a workaround, you can set batchlimit to 1 so RSET is never used.</p>
1559 <h1>How to make fetchmail work with various servers</h1>
1560 <h2><a id="S1" name="S1"><strike>S1. How can I use fetchmail with
1561 qpopper?</strike></a></h2>
1563 <p><em>The information that used to be here was obsolete and dropped.</em></p>
1565 <h2><a id="S2" name="S2">S2. How can I use fetchmail with Microsoft
1568 <p>It's been reliably reported that Exchange 2000's POP3 support is
1569 so broken that it's unusable. One symptom is that messages without
1570 a terminating newline get the POP3 message termination dot emitted
1571 -- you guessed it -- right after the last character of the message,
1572 with no terminating newline added. This will hang fetchmail or any
1573 other RFC-compliant server. IMAP is alleged to work OK, though.</p>
1575 <p>Older versions of Exchange are semi-usable. They randomly drop
1576 attachments on the floor, though. Microsoft acknowledges this
1577 as a known bug and apparently has no plans to fix it.</p>
1579 <p>Fetchmail using IMAP supports the proprietary NTLM mode used
1580 with M$ Exchange servers. To enable this, configure fetchmail with
1581 the --enable-NTLM option and recompile it. Specify a user option
1582 value that looks like 'user@domain': the part to the left of the @
1583 will be passed as the username and the part to the right as the
1586 <p>M$ Exchange violates the POP3 and IMAP RFCs. Its LIST command
1587 does not reveal the real sizes of mail in the pop mailbox, but the
1588 sizes of the compressed versions in the exchange mail database
1589 (thanks to Arjan De Vet and Guido Van Rooij for alerting us to this
1592 <p>Fetchmail works with M$ Exchange, despite this brain damage. Two
1593 features are compromised. One is that the --limit option will not
1594 work right (it will check against compressed and not actual sizes).
1595 The other is that a too-small SIZE argument may be passed to your
1596 ESMTP listener, assuming you're using one (this should not be a
1597 problem unless the actual size of the message is above the
1598 listener's configured length limit).</p>
1600 <p>Somewhat belatedly, I've learned that there's supposed to be a
1601 registry bit that can fix this breakage:</p>
1604 HKEY_LOCAL_MACHINE\System\CurrentControlSet\Services\MsExchangeIs\Parameters
1605 System\Pop3 Compatibility
1608 <p>This is a bitmask that controls the variations from the standard
1609 protocol. The bits defined are:</p>
1612 <dt>0x00000001:</dt>
1614 <dd>Report exact message sizes for the LIST command</dd>
1616 <dt>0x00000002:</dt>
1618 <dd>Allow arbitrary linear whitespace between commands and
1621 <dt>0x00000004:</dt>
1623 <dd>Enable the LAST command</dd>
1625 <dt>0x00000008:</dt>
1627 <dd>Allow an empty PASS command (needed for users with blank
1628 passwords, but illegal in the protocol)</dd>
1630 <dt>0x00000010:</dt>
1632 <dd>Relax the length restrictions for arguments to commands
1633 (protocol requires 40, but some user names may be longer than
1636 <dt>0x00000020:</dt>
1638 <dd>Allow spaces in the argument to the USER command.</dd>
1641 <p>There's another one that may be useful to know about:</p>
1644 KEY_LOCAL_MACHINE\System\CurrentControlSet\Services\MsExchangeIs\Parameters
1645 System\Pop3 Performance
1649 <dt>0x00000001:</dt>
1651 <dd>Render messages to a temporary stream instead of sending
1652 directly from the database (should always be on)</dd>
1654 <dt>0x00000002: Flag unrenderable messages (instead of just failing
1655 commands) (should only be on if you are seeing the problems
1656 reported in KB Q168109)</dt>
1658 <dt>0x00000004:</dt>
1660 <dd>Return from the QUIT command before all messages have been
1664 <p>The Microsoft pod-person who revealed this information to me
1665 admitted that he couldn't find it anywhere in their public
1668 <p>Another specific problem we have seen with Exchange servers has
1669 as its symptom a response to LOGIN that says "NO Ambiguous Alias".
1670 Grant Edwards writes:</p>
1672 <p>This means that Exchange Server is too f*&#ing stupid to
1673 figure out which mailbox belongs to you. Instead of actually
1674 keeping track of which inbox belongs to which user, it uses some
1675 half-witted, guess-o-matic heuristic to try to guess your mailbox
1676 name from your username.</p>
1678 <p>In your case it doesn't work because your username maps to more
1679 than one mailbox. For some people it doesn't work because their
1680 username maps to zero mailboxes. This is yet another inept, lame,
1681 almost criminally negligent design decision from our friends in
1684 <p>You've got several options:</p>
1687 <li>Get your administrator to configure the server so that
1688 usernames and mailbox names are the same.</li>
1690 <li>Get your administrator to add an alias that maps your username
1691 explicitly to your mailbox name.</li>
1694 <p>But, the best option involves finding a server that runs better
1697 <h2><a id="S3" name="S3">S3. How can I use fetchmail with HP
1700 <p>No special configuration is required, but OpenMail versions
1701 prior to 6.0 have an annoying bug similar to the big one in <a
1702 href="#S2">Microsoft Exchange</a>. The message sizes it gives in
1703 the LIST are rounded to the nearest 1024 bytes. It also has a nasty
1704 habit of discarding headers it doesn't recognize, such as X- and
1705 Resent- headers.</p>
1707 <p>As with M$ Exchange, the only real fix for these problems is to
1708 get a POP (or preferably IMAP) server that isn't brain-dead.
1709 OpenMail's project manager claims these bugs have been fixed in
1712 <p>We've had a more recent report (December 2001) that the TOP
1713 command fails, returning only one line regardless of its argument,
1714 on something identifying itself as "OpenMail POP3 interface".</p>
1716 <h2><a id="S4" name="S4">S4. How can I use fetchmail with Novell GroupWise?</a></h2>
1718 <p>The Novell GroupWise IMAP server would be better named
1719 GroupFoolish; it is (according to the designer of IMAP) unusably
1720 broken. Among other things, it doesn't include a required content
1721 length in its BODY[TEXT] response.</p>
1723 <p>Fetchmail works around this problem, but we strongly recommend
1724 voting with your dollars for a server that isn't brain-dead.</p>
1726 <h2><a id="S5" name="S5">S5. How can I use fetchmail with
1727 InterChange?</a></h2>
1729 <p>You can't. At least not if you want to be able to see
1730 attachments. InterChange has a bug similar to the MailMax server (<a
1731 href="#S6">see below</a>):
1732 it reports the message length with attachments but doesn't download
1733 them on TOP or RETR.</p>
1735 <p>On Jan 9 2001, the people at InfiniteMail sent me mail informing
1736 me that their new 3.61.08 release of InterChange fixes this
1737 problem. I don't have any reports one way or the other yet.</p>
1739 <h2><a id="S6" name="S6">S6. How can I use fetchmail with MailMax?</a></h2>
1741 <p>You can't. At least not if you want to be able to see
1742 attachments. MailMax has a bug; it reports the message length with
1743 attachments but doesn't download them on TOP or RETR.</p>
1745 <p>Also, we're told that TOP sometimes fails to retrieve the entire
1746 message even when enough lines have been specified. The MailMax
1747 developers have acknowledged this bug as of 4 May 2000, but there
1748 is no fix yet. If you must use this server, force RETR with the
1749 <tt>fetchall</tt> option.</p>
1751 <h2><a id="S7" name="S7">S7. How can I use fetchmail with FTGate?</a></h2>
1753 <p>The FTGate V2 server (and possibly older versions as well) has a
1754 weird bug. It answers OK twice to a TOP request! Use the
1755 <code>fetchall</code> option to force use of RETR and work around
1759 <h1>How to fetchmail work with specific ISPs</h1>
1760 <h2><a id="I1" name="I1">I1. How can I use fetchmail with CompuServe RPA?</a></h2>
1762 <p>First, make sure your fetchmail has the RPA support compiled in.
1763 Stock fetchmail binaries (such as you might get from an RPM) don't.
1764 You can check this by looking at the output of <code>fetchmail
1765 -V</code>; if you see the string "+RPA" after the version ID you're
1766 good to go, otherwise you'll have to build your own from sources
1767 (see the INSTALL file in the source distribution for
1770 <p>Give your CompuServe pass-phrase in lower case as your password.
1771 Add '@compuserve.com' to your user ID so that it looks like 'user
1772 <UserID>@compuserve.com', where <UserID> can be either
1773 your numerical userID or your E-mail nickname. An RPA-enabled
1774 fetchmail will automatically check for csi.com in the POP server's
1775 greeting line. If that's found, and your user ID ends with
1776 '@compuserve.com', it will query the server to see if it is
1777 RPA-capable, and if so do an RPA transaction rather than a
1778 plain-text password handshake.</p>
1780 <p><strong>Warning:</strong> the debug (-v -v) output of fetchmail
1781 will show your pass-phrase in Unicode!</p>
1783 <p>These two .fetchmailrc entries show the difference between an
1784 RPA and non-RPA configuration:</p>
1787 # This version will use RPA
1788 poll csi.com via "pop.site1.csi.com" with proto POP3 and options no dns
1789 user "CSERVE_USER@compuserve.com" there with password "CSERVE_PASSWORD"
1790 is LOCAL_USER here options fetchall stripcr
1792 # This version will not use RPA
1793 poll non-rpa.csi.com via "pop.site1.csi.com" with proto POP3 and options no dns
1794 user "CSERVE_USER" there with password "CSERVE_POP3_PASSWORD"
1795 is LOCAL_USER here options fetchall stripcr
1798 <h2><a id="I2" name="I2">I2. How can I use fetchmail with Demon
1799 Internet's SDPS?</a></h2>
1801 <h3>Single-drop mode</h3>
1803 <p>You can get fetchmail to download the email for just one user
1804 from Demon Internet's POP3 server by giving it a username
1805 consisting of your Demon user name followed by your account name,
1806 with an at-sign between them.</p>
1808 <p>For example, to download email for the user
1809 <philh@vision25.demon.co.uk>, you could use the following
1810 .fetchmailrc file:</p>
1813 set postmaster "philh"
1814 poll pop3.demon.co.uk with protocol POP3:
1815 user "philh@vision25" is philh
1818 <h3>Multi-drop mode</h3>
1820 <p>Demon Internet's SDPS service is an implementation of POP3. All
1821 messages have a Received: header added when they enter the
1822 maildrop, like this:</p>
1825 Received: from punt-1.mail.demon.net by mailstore for fred@xyz.demon.co.uk
1826 id 899963657:10:27896:0; Thu, 09 Jul 98 05:54:17 GMT
1829 <p>To enable multi-drop mode you need to tell fetchmail that
1830 'mailstore' is the name of the host which accepted the mail, and
1831 let it know the hostname part(s) of your E-mail address. The
1832 following example assumes that your hostname is xyz.demon.co.uk,
1833 and that you have also bought "mail forwarding" for the domain
1834 my-company.co.uk (in which case your MTA must also be configured to
1835 accept mail sent to user@my-company.co.uk)</p>
1838 poll pop3.demon.co.uk proto pop3 aka mailstore no dns:
1839 localdomains xyz.demon.co.uk my-company.co.uk
1843 <p>Note that Demon may delete mail on the server which is more than
1844 30 days old; see their <a
1845 href="http://www.demon.net/helpdesk/producthelp/mail/sdps-tech.html/">POP3
1846 page</a> for details.</p>
1848 <h3>The SDPS extension</h3>
1850 <p>There's a different way to do multidrop. It's not necessary on
1851 Demon Internet, since fetchmail can parse Received addresses, but
1852 the person who implemented this didn't know that. It may be useful
1853 if Demon Internet ever changes mail transports.</p>
1855 <p>SDPS includes a non-standard extension for retrieving the
1856 envelope of a message (*ENV), which fetchmail optionally supports
1857 if compiled with the --enable-SDPS option. If you have it, the
1858 first line of the fetchmail -V response will include the string
1861 <p>Once you have SDPS compiled in, fetchmail in POP3 mode will
1862 automatically detect when it's talking to a Demon Internet host in
1863 multidrop mode, and use the *ENV extension to get an envelope To
1866 <p>The autodetection works by looking at the hostname in the POP3
1867 greeting line; if you're accessing Demon Internet through a proxy
1868 it may fail. To force SDPS mode, pick "sdps" as your protocol.</p>
1870 <h2><a id="I3" name="I3">I3. How can I use fetchmail with usa.net's
1873 <p>Enable '<code>fetchall</code>'. A user reports that the 2.2
1874 version of USA.NET's POP server reports that you must use the
1875 '<code>fetchall</code>' option to make sure that all of the mail is
1876 retrieved, otherwise some may be left on the server. This is almost
1877 certainly a server bug.</p>
1879 <p>The usa.net servers (at least in their 2.2 version, June 1998)
1880 don't handle the TOP command properly, either. Regardless of the
1881 argument you give it, they retrieve only about 10 lines of the
1882 message. Fetchmail normally uses TOP for message retrieval in order
1883 to avoid marking messages seen, but '<code>fetchall</code>' forces
1884 it to use RETR instead.</p>
1886 <p>Also, we're told USA.NET adds a ton of hops to your messages.
1887 You may need to raise the MaxHopCount parameter in your sendmail.cf
1888 to avoid having fetched mail rejected.</p>
1890 <h2><a id="I4" name="I4">I4. How can I use fetchmail with geocities
1891 POP3 servers?</a></h2>
1893 <p>Nathan Cutler reports that the the mail.geocities.com POP3
1894 servers fail to include the first Received line of the message in
1895 the send to fetchmail. This can solve problems if your MUA
1896 interprets Received continuations as body lines and doesn't parse
1897 any of the following headers.</p>
1899 <p>Workaround is to use "mda" keyword or "--mda" switch:</p>
1902 mda "sed -e '1s/^\t/Received: /' | formail | /usr/bin/procmail -d <user>"
1905 <p>Replace \t with exactly one tabulation character.</p>
1907 <p>You should also consider using "fetchall" option because
1908 Geocities' servers sometimes think that the first 45 messages have
1909 already been read.</p>
1911 <h2><a id="I5" name="I5">I5. How can I use fetchmail with Hotmail or Lycos Webmail?</a></h2>
1913 <p>You can't directly. But you can use fetchmail with hotmail or lycos
1914 webmail with the help of the <a
1915 href='http://people.freenet.de/courierdave/'>HotWayDaemon</a>
1916 daemon. You don't even need to install hotwayd as a daemon in
1917 <samp>inetd.conf</samp> but can use it as a plugin. Your
1918 configuration should look like this:</p>
1921 poll localhost protocol pop3 tracepolls
1922 plugin "/usr/local/sbin/hotwayd -l 0 -p yourproxy:yourproxyport"
1923 username "youremail@hotmail.com" password "yourpassword"
1927 <p>As a second option you may consider using <a
1928 href="http://linux.cudeso.be/linuxdoc/gotmail.php">gotmail</a>.</p>
1930 <h2><a id="I6" name="I6">I6. How can I use fetchmail with MSN?</a></h2>
1932 <p>You can't. MSN uses something that looks like POP3, except the
1933 authentication part is nonstandard. And of course they don't
1934 document it, so nobody but their Windows clients can speak it.</p>
1936 <p>This is a customer lock-in tactic; we recommend boycotting MSN
1937 as the only appropriate response.</p>
1939 <p>As of 5.0.8, we have support for the client side of NTLM
1940 authentication. It's possible this may enable fetchmail to talk to
1941 MSN; if so, somebody should report it so this FAQ can be
1944 <h2><a id="I7" name="I7">I7. How can I use fetchmail with SpryNet?</a></h2>
1946 <p>The SpryNet POP3 servers mark a message queried with TOP as
1947 seen. This means that if your connection drops in mid-message, it
1948 may end up invisibly stuck on your mail spool. Use the
1949 <code>fetchall</code> flag to ensure that it's recovered on the
1952 <h2><a id="I8" name="I8">I8. How can I use fetchmail with comcast.net or
1953 other Maillennium servers?</a></h2>
1955 <p>Stock fetchmail will work with a
1956 Maillennium POP3/PROXY server... <em>but</em> this server will
1957 truncate "TOP" responses after 64 - 82 kB (we have varying reports),
1958 in violation of Internet Standard #53 aka. RFC-1939 (POP3). Don't
1959 mistake this for a fetchmail bug. (Reported July 2003.) Comcast
1960 documented they haven't understood what this is about in <a
1961 href="http://lists.ccil.org/pipermail/fetchmail-friends/2004-April/008523.html">two
1962 messages from April 2004.</a></p>
1964 <p>Beginning with version 6.3.2, fetchmail will fall back to the RETR
1965 command if the greeting string contains "Maillennium POP3/PROXY server",
1966 and print a warning message. This means however that fetchmail has no
1967 means to prevent the "seen" flag from being set on the server (Note that
1968 officially, POP3 has no notion of seen tracking, but it works for some
1971 <p>Workaround for older versions: use the <tt>fetchall</tt> option.</p>
1973 <h2><a id="I9" name="I9">I9. How can I use fetchmail with GMail/Google Mail?</a></h2>
1975 <p>Google's IMAP servers, as of April 2008, are broken and re-encode
1976 MIME-encoded headers improperly and are not feature-complete yet. The
1977 model how their servers organize mail also deviates in significant ways
1978 from what the POP3 or IMAP protocol 'fathers' conceived. This means all
1979 sorts of strange effects, for instance, your sent mail may show up in
1980 the mail that fetchmail fetches. It's best to avoid fetching mail from
1981 Google until they are using standards-compliant software.</p>
1984 <h1>How to set up well-known security and authentication
1986 <h2><a id="K1" name="K1">K1. How can I use fetchmail with SOCKS?</a></h2>
1988 <p>Giuseppe Guerini added a <kbd>--with-socks</kbd> compile-time option
1989 that supports linking with socks library. If you specify the value of
1990 this option as "yes", the configure script will try to find the Rconnect
1991 library and set the makefile up to link it. You can also specify a
1992 directory containing the Rconnect library.</p>
1994 <p>Alan Schmitt has added a similar <kbd>--with-socks5</kbd> option that may
1995 work better if you have a recent version of the SOCKS library.</p>
1997 <p>In either case, fetchmail has no direct configuration hooks, but you
1998 can specify which socks configuration file the library should read by
1999 means of the <tt>SOCKS_CONF</tt> environment variable. In order to
2000 bypass the SOCKS proxy altogether, you could run (adding your usual
2001 options to the end of this line):</p>
2003 <pre>env SOCKS_CONF=/dev/null fetchmail</pre>
2005 <h2><a id="K2" name="K2">K2. How can I use fetchmail with IPv6 and
2008 <p>To use fetchmail with IPv6, you need a system that supports
2009 IPv6, the "Basic Socket Interface Extensions for IPv6" (RFC 2133).
2012 <p>The NRL IPv6+IPsec software distribution can be obtained from:
2014 href="http://web.mit.edu/network/isakmp/">http://web.mit.edu/network/isakmp/</a></p>
2016 <p>More information on using IPv6 with Linux can be obtained
2021 href="http://www.bieringer.de/linux/IPv6/IPv6-HOWTO/IPv6-HOWTO.html">
2022 http://www.bieringer.de/linux/IPv6/IPv6-HOWTO/IPv6-HOWTO.html</a></li>
2025 <h2><a id="K3" name="K3">K3. How can I get fetchmail to work with
2028 <p>Use the <tt>plugin</tt> option. This is dead simple with
2032 plugin "ssh %h /usr/sbin/imapd"
2035 <p>You may have to use a different absolute pathname, whatever the
2036 location of imapd on your mailserver is. This option tells
2037 fetchmail that instead of opening a connection on the server's port
2038 143 and doing standard IMAP authentication, fetchmail should ssh to
2039 the server and run imapd, using the more secure ssh authentication
2040 (as well as getting ssh's end-to-end encryption). Most IMAP daemons
2041 will detect that they've been called from the command line and
2042 assume the connection is preauthenticated.</p>
2044 <p>POP3 daemons aren't quite as smart. They won't know they are
2045 preauthenticated in this mode, so you'll actually have to ship your
2046 password. It will be under ssh encryption, though, so that
2047 shouldn't be a problem.</p>
2049 <h2><a id="K4" name="K4">K4. What do I have to do to use the
2050 IMAP-GSS protocol?</a></h2>
2052 <p>Fetchmail can use RFC1731 GSSAPI authorization to safely
2053 identify you to your IMAP server, as long as you can share
2054 Kerberos V credentials with your mail host and you have a GSSAPI-capable
2055 IMAP server - those are few.</p>
2057 <p>fetchmail does not compile in support for GSS by
2058 default, since it requires libraries from the Kerberos V
2059 distribution (available via FTP at <a
2060 href="ftp://athena-dist.mit.edu/pub/ATHENA/kerberos">athena-dist.mit.edu</a>).
2061 If you have these, compiling in GSS support is simple: add a
2062 <code>--with-gssapi=[/path/to/krb5/root]</code> option to
2063 configure. For instance, I have all of my Kerberos V libraries
2064 installed under /usr/krb5 so I run <code>configure
2065 --with-gssapi=/usr/krb5</code></p>
2067 <p>Setting up Kerberos V authentication is beyond the scope of this
2068 FAQ (you may find Jim Rome's paper <a
2069 href="http://www.ornl.gov/~jar/HowToKerb.html">How to Kerberize
2070 your site</a> helpful), but you'll at least need to add a
2071 credential for imap/[mailhost] to the keytab of the mail server
2072 (IMAP doesn't just use the host key). Then you'll need to have your
2073 credentials ready on your machine (cf. kinit).</p>
2075 <p>After that things are very simple. Set your protocol to imap-gss
2076 in your .fetchmailrc, and omit the password, since imap-gss doesn't
2077 need one. You can specify a username if you want, but this is only
2078 useful if your mailbox belongs to a username different from your
2079 Kerberos principal.</p>
2081 <p>Now you don't have to worry about your password appearing in
2082 cleartext in your .fetchmailrc, or across the network.</p>
2084 <h2><a id="K5" name="K5">K5. How can I use fetchmail with
2087 <p>You'll need to have the <a
2088 href="http://www.openssl.org/">OpenSSL</a> libraries installed, and they
2089 should at least be version 0.9.6.
2090 Configure with --with-ssl. If you have the OpenSSL libraries
2091 installed in commonly-used default locations, this will
2092 suffice. If you have them installed in a non-default location,
2093 you'll need to specify the OpenSSL installation directory as an argument
2094 to --with-ssl after an equal sign.</p>
2096 <p>Fetchmail binaries built this way support <code>ssl</code>,
2097 <code>sslkey</code>, and <code>sslcert</code> options that control
2098 SSL encryption, and will automatically use <code>tls</code> if the
2099 server offers it. You will need to have an SSL-enabled mailserver to
2100 use these options. See the manual page for details and some words
2101 of care on the limited security provided.</p>
2103 <p>If your open OpenSSL session dies with a message that complains
2104 "PRNG not seeded", update or improve your operating system. This
2105 means that the OpenSSL library on your machine has been unable to
2106 locate a source of random bits from which to seed its random-number
2107 generator; normally these come from the <tt>/dev/urandom</tt>, and
2108 this message probably means your OS doesn't have that device.</p>
2110 <p>An interactive program could seed the random number generator
2111 from keystroke timings or some other form of user input. Because
2112 fetchmail is primarily designed to run forever as a background
2113 daemon, that option is not available in this case.</p>
2115 <p>If you don't have the libraries installed, but do have the
2116 OpenSSL utility toolkit, something like this may work (but will not
2117 authenticate the server):</p>
2120 poll MYSERVER port 993 plugin "openssl s_client -connect %h:%p"
2121 protocol imap username MYUSERNAME password MYPASSWORD
2124 <p>You should note that SSL is only secure against a "man-in-the-middle"
2125 attack if the client is able to verify that the peer's public key is the
2126 correct one, and has not been substituted by an attacker. fetchmail can do
2127 this in one of two ways: by verifying the SSL certificate, or by checking
2128 the fingerprint of the peer's public key.</p>
2130 <p>There are three parts to SSL certificate verification: checking that the
2131 domain name in the certificate matches the hostname you asked to connect to;
2132 checking that the certificate expiry date has not passed; and checking that
2133 the certificate has been signed by a known Certificate Authority (CA). This
2134 last step takes some preparation, as you need to install the root
2135 certificates of all the CA's which you might come across.</p>
2137 <p>The easiest way to do this is using the root CA keys supplied in the
2138 OpenSSL distribution, which means you need to download and unpack the
2139 source tarball from www.openssl.org. Once you have done that:</p>
2142 <li><code>mkdir /etc/ssl/certs</code></li>
2143 <li>in the openssl-x.x.x/certs directory: <code>cp *.pem /etc/ssl/certs/</code></li>
2144 <li>in the openssl-x.x.x/tools directory: edit c_rehash and set
2145 <code>$dir="/etc/ssl"</code></li>
2146 <li>run "perl c_rehash". This generates a number of symlinks within the
2147 /etc/ssl/certs/ directory</li>
2150 <p>Now in .fetchmailrc, set option sslcertpath to point to this
2154 poll pop3.example.com proto pop3 uidl no dns
2155 user foobar@example.com password xyzzy is foobar ssl sslcertpath /etc/ssl/certs
2158 <p>If the server certificate has not been signed by a known CA (e.g. it is a
2159 self-signed certificate), then this certificate validation will always
2162 <p>Certificate verification is always attempted. If it fails, by default a
2163 warning is printed but the connection carries on (which means you are not
2164 protected against attack). If your server's certificate has been properly
2165 set up and verifies correctly, then add the "sslcertck" option to enforce
2166 validation. If your server doesn't have a valid certificate though (e.g. it
2167 has a self-signed certificate) then it will never verify, and the only way
2168 you can protect yourself is by checking the fingerprint.</p>
2170 <p>To check the peer fingerprint: first use fetchmail -v once to connect to
2171 the host, at a time when you are pretty sure that there is no attack in
2172 progress (e.g. you are not traversing any untrusted network to reach the
2173 server). Make a note of the fingerprint shown. Now embed this in your
2174 .fetchmailrc using the sslfingerprint option: e.g.</p>
2177 poll pop3.example.com proto pop3 uidl no dns
2178 user foobar@example.com password xyzzy is foobar
2179 ssl sslfingerprint "67:3E:02:94:D3:5B:C3:16:86:71:37:01:B1:3B:BC:E2"
2182 <p>When you next connect, the public key presented by the server will be
2183 verified against the fingerprint given. If it's different, it may mean that
2184 a man-in-the-middle attack is in progress - or it might just mean that the
2185 server changed its key. It's up to you to determine which has happened.</p>
2187 <h2><a id="K6" name="K6">K6. How can I tell fetchmail not to use TLS
2188 if the server advertises it? Why does fetchmail use SSL even
2189 though not configured?</a></h2>
2191 <p>Some servers advertise STLS (POP3) or STARTTLS (IMAP), and fetchmail
2192 will automatically attempt TLS negotiation if SSL was enabled at compile
2193 time. This can however cause problems if the upstream didn't configure
2194 his certificates properly.</p>
2196 <p>In order to prevent fetchmail from trying TLS (STLS, STARTTLS)
2197 negotiation, add this option:</p>
2199 <pre>sslproto ssl23</pre>
2201 <p>This restricts fetchmail's SSL/TLS protocol choice from the default
2202 "SSLv2, SSLv3, TLSv1" to the two SSL variants, disabling TLSv1. Note
2203 however that this causes the connection to be unencrypted unless an
2204 encrypting "plugin" is used or SSL is requested explicitly.</p>
2207 <h1>Runtime fatal errors</h1>
2208 <h2><a id="R1" name="R1">R1. Fetchmail isn't working, and -v shows
2209 'SMTP connect failed' messages.</a></h2>
2211 <p>Fetchmail itself is probably working, but your SMTP port 25
2212 listener is down or inaccessible.</p>
2214 <p>The first thing to check is if you can telnet to port 25 on your
2215 smtp host (which is normally 'localhost' unless you've specified an
2216 smtp option in your .fetchmailrc or on the command line) and get a
2217 greeting line from the listener. If the SMTP host is inaccessible
2218 or the listener is down, fix that first.</p>
2220 <p>In Red Hat Linux 6.x, SMTP is disabled by default. To fix this,
2221 set "DAEMON=yes" in your /etc/sysconfig/sendmail file, then restart
2222 sendmail by running "/sbin/service sendmail restart".</p>
2224 <p>If the listener seems to be up when you test with telnet, the
2225 most benign and typical problem is that the listener had a
2226 momentary seizure due to resource exhaustion while fetchmail was
2227 polling it -- process table full or some other problem that stopped
2228 the listener process from forking. If your SMTP host is not
2229 'localhost' or something else in /etc/hosts, the fetchmail glitch
2230 could also have been caused by transient nameserver failure.</p>
2232 <p>Try running fetchmail -v again; if it succeeds, you had one of
2233 these kinds of transient glitch. You can ignore these hiccups,
2234 because a future fetchmail run will get the mail through.</p>
2236 <p>If the listener tests up, but you have chronic failures trying
2237 to connect to it anyway, your problem is more serious. One way to
2238 work around chronic SMTP connect problems is to use --mda. But this
2239 only attacks the symptom; you may have a DNS or TCP routing
2240 problem. You should really try to figure out what's going on
2241 underneath before it bites you some other way.</p>
2243 <p>We have one report (from toby@eskimo.com) that you can sometimes
2244 solve such problems by doing an <code>smtp</code> declaration with
2245 an IP address that your routing table maps to something other than
2246 the loopback device (he used ppp0).</p>
2248 <p>We also have a report that this error can be caused by having an
2249 /etc/hosts file that associates your client host name with more
2250 than one IP address.</p>
2252 <p>It's also possible that your DNS configuration isn't looking at
2253 <code>/etc/hosts</code> at all. If you're using libc5, look at
2254 <code>/etc/resolv.conf</code>; it should say something like:</p>
2260 <p>so your <code>/etc/hosts</code> file is checked first. If you're
2261 running GNU libc6, check your <code>/etc/nsswitch.conf</code> file.
2262 Make sure it says something like</p>
2268 <p>again, in order to make sure <code>/etc/hosts</code> is seen
2271 <p>If you have a hostname set for your machine, and this hostname
2272 does not appear in /etc/hosts, you will be able to telnet to port
2273 25 and even send a mail with rcpt to: user@host-not-in-/etc/hosts,
2274 but fetchmail can't seem to get in touch with sendmail, no matter
2275 what you set smtpaddress to.</p>
2277 <p>We had another report from a Linux user of fetchmail 2.1 who
2278 solved his SMTP connection problem by removing the reference to
2279 -lresolv from his link line and relinking. Apparently in some older
2280 Linux distributions the libc bind library version works better.</p>
2282 <p>As of 2.2, the configure script has been hacked so the bind
2283 library is linked only if it is actually needed. So under Linux it
2284 won't be, and this particular cause should go away.</p>
2286 <h2><a id="R2" name="R2">R2. When I try to configure an MDA,
2287 fetchmail doesn't work.</a></h2>
2289 <p>(I hear this one from people who have run into the blank-line
2290 problem in <a href="#X1">X1</a>.)</p>
2292 <p>Try sending yourself test mail and retrieving it using the
2293 command-line options '<code>-k -m cat</code>'. This will dump
2294 exactly what fetchmail retrieves to standard output (plus the
2295 Received line fetchmail itself adds to the headers).</p>
2297 <p>If the dump doesn't match what shows up in your mailbox when you
2298 configure an MDA, your MDA is mangling the message. If it doesn't
2299 match what you sent, then fetchmail or something on the server is
2302 <h2><a id="R3" name="R3">R3. Fetchmail dumps core when given an
2303 invalid rc file.</a></h2>
2305 <p>Note that this bug should no longer occur when using prepackaged
2306 fetchmail versions or installing unmodified original tarballs, since
2307 these ship with a proper parser .c file.</p>
2309 <p>This is usually reported from AIX or Ultrix, but has even been
2310 known to happen on Linuxes without a recent version of
2311 <code>flex</code> installed. The problem appears to be a result of
2312 building with an archaic version of lex.</p>
2314 <p>Workaround: fix the syntax of your .fetchmailrc file.</p>
2316 <p>Fix: build and install the latest version of <a
2317 href="http://flex.sourceforge.net/">flex</a>.</p>
2319 <h2><a id="R4" name="R4"><strike>R4. Fetchmail dumps core in -V mode, but
2320 operates normally otherwise.</strike></a></h2>
2322 <p><em>The information that used to be here referred to bugs in Linux libc5
2323 systems, which are deemed obsolete by now.</em></p>
2325 <h2><a id="R5" name="R5">R5. Running fetchmail in daemon mode
2326 doesn't work.</a><br/>
2329 <p>We have one report from a SunOS 4.1.4 user that trying to run
2330 fetchmail in detached daemon mode doesn't work, but that using the
2331 same options with -N (nodetach) is OK. We have another report of
2332 similar behavior from one Linux user, but many other Linux users
2333 report no problem.</p>
2335 <p>If this happens, you have a specific portability problem with
2336 the code in daemon.c that detaches and backgrounds the daemon
2337 fetchmail. The isolated Linux case has been chased down to a
2338 failure in dup(2) that may reflect a glibc bug.</p>
2340 <p>As a workaround, you can start fetchmail with -N and an
2341 ampersand to background it. A Sun user recommends this:</p>
2344 (fetchmail --nodetach <other params> &)
2347 <p>The extra pair of parens is significant --- it makes sure that
2348 the process detaches from the initial shell (one more shell is
2349 started and dies immediately, detaching fetchmail and making it
2350 child of PID 1). This is important when you start fetchmail
2351 interactively and than quit interactive shell. The line above makes
2352 sure fetchmail lives after that!</p>
2354 <h2><a id="R6" name="R6">R6. Fetchmail randomly dies with socket
2357 <p>Check the MTU value in your PPP interface reported by
2358 <code>/sbin/ifconfig</code>. If it's over 600, change it in your
2359 PPP options file. (<code>/etc/ppp/options</code> on my box). Here
2360 are option values that work:</p>
2367 <p>Another circumstance that can trigger this is if you are polling
2368 a virtual-mail-server name that is round-robin connected to
2369 different actual servers, so you get different IP addresses on
2370 different poll cycles. To work around this, change the poll name
2371 either to the real name of one of the servers in the ring or to a
2372 corresponding IP address.</p>
2374 <h2><a id="R7" name="R7">R7. Fetchmail running as root stopped
2375 working after an OS upgrade</a></h2>
2377 <p>In RH 6.0, the HOME value in the boot-time root environment
2378 changed from /root to / as the result of a change in init. Move
2379 your .fetchmailrc or use a -f option to explicitly point at the
2380 file. (Oddly, a similar problem has been reported from Debian
2383 <h2><a id="R8" name="R8">R8. Fetchmail is timing out after fetching
2384 certain messages but before deleting them</a></h2>
2386 <p>There's a TCP/IP stalling problem under Redhat 6.0 (and possibly
2387 other recent Linuxes) that can cause this symptom. Brian Boutel
2391 <p>TCP timestamps are turned on on my Linux boxes (I assume it's
2392 now the default). This uses 12 extra bytes per segment. When the
2393 tcp connection starts, the other end agrees a MSS of 1460, and then
2394 fragments 1460 byte chunks into 1448 and 12, because is is not
2395 allowing for the timestamp.</p>
2397 <p>Then, for reasons I can't explain, it waits a long time
2398 (typically 2 minutes) after the ack is sent before sending the next
2399 (fragmented) packet. Turning off tcp timestamps avoids the
2400 fragmentation and restores normal behaviour. To do this,
2403 <p>echo 0 > /proc/sys/net/ipv4/tcp_timestamps</p>
2405 <p>I'm still unclear about the details of why this is happening. At
2406 least [now] I am now getting good performance and no queue
2410 <h2><a id="R9" name="R9">R9. Fetchmail is timing out during message
2413 <p>This is probably a general networking issue. Sending a "RETR"
2414 command will cause the server to start sending large amounts of
2415 data, which means large packets. If your networking layer has a
2416 packet-fragmentation problem or improper firewall settings break Path
2417 MTU discovery (when for instance all ICMP traffic is blocked), that's
2418 where you'll see it.</p>
2420 <h2><a id="R10" name="R10"><strike>R10. Fetchmail is dying with
2421 SIGPIPE.</strike></a></h2>
2423 <p><em>Fetchmail 6.3.5 and newer block SIGPIPE, and many older versions have
2424 already handled this signal, so you shouldn't be seeing SIGPIPE
2427 <h2><a id="R11" name="R11">R11. My server is hanging or emitting
2428 errors on CAPA.</a></h2>
2430 <p>Your POP3 server is broken. You can work around this with the
2431 declaration <tt>auth password</tt> in your .fetchmailrc.</p>
2433 <h2><a id="R12" name="R12">R12. Fetchmail isn't working and reports
2434 getaddrinfo errors.</a></h2>
2435 <ol><li>Make sure you haven't mistyped the host name or address, and that
2436 your DNS is working. If you cannot fix DNS, give the numeric host
2437 literal, for instance, 192.168.0.1</li>
2438 <li>Make sure your <code>/etc/services</code> file (or other
2439 services database) contains the necessary service entries. If you
2440 cannot fix the services database, use the --service option and give the
2441 numeric port address. Common port addresses are:<table
2442 summary="Common port addresses for IMAP, POP3 and their SSL
2444 <tr><th>service</th><th>port</th></tr>
2445 <tr><td>IMAP</td><td>143</td></tr>
2446 <tr><td>IMAP+SSL</td><td>993</td></tr>
2447 <tr><td>POP3</td><td>110</td></tr>
2448 <tr><td>POP3+SSL</td><td>995</td></tr>
2451 <h2><a id="R13" name="R13">R13. What does "Interrupted system call"
2454 <p>Non-fatal signals (such as timers set by fetchmail itself) can
2455 interrupt long-running functions and will then be reported as
2456 "Interrupted system call". These can sometimes be timeouts.</p>
2459 <h1>Hangs and lockups</h1>
2460 <h2><a id="H1" name="H1">H1. Fetchmail hangs when used with
2463 <p>Your problem may be with pppd's 'demand' option. We have a
2464 report that fetchmail doesn't play well with it, but works with
2465 pppd if 'demand' is turned off. We have no idea why this is.</p>
2467 <h2><a id="H2" name="H2">H2. Fetchmail hangs during the MAIL FROM
2470 <p>The symptom: 'fetchmail -v' retrieves mail fine, but appears to
2471 hang after sending the MAIL FROM command</p>
2474 SMTP> MAIL FROM: <someone@somewhere>
2477 <p>The hang is actually occuring when sendmail looks up a sender's
2478 address in DNS. The problem isn't in fetchmail but in the
2479 configuration of sendmail. You must enable the 'nodns' and
2480 'nocanonify' features of sendmail.</p>
2482 <p>Here was my fix for RedHat 7.2:</p>
2485 <li># cd /etc/mail</li>
2487 <li># cp sendmail.mc sendmail-mine.mc</li>
2489 <li>Edit sendmail-mine.mc and add lines:
2497 <li>Build a new sendmail.cf
2500 # m4 sendmail-mine.cf > /etc/sendmail.cf
2504 <li>Restart sendmail.</li>
2507 <p>For more details consult the file
2508 /usr/share/sendmail-cf/README.</p>
2510 <h2><a id="H3" name="H3">H3. Fetchmail hangs while fetching
2513 <p>The symption: 'fetchmail -v' retrieves the first few messages,
2514 but hangs returning:</p>
2517 fetchmail: SMTP< 550 5.0.0 Access denied
2518 fetchmail: SMTP> RSET
2519 fetchmail: SMTP< 250 2.0.0 Reset state
2520 .......fetchmail: flushed
2521 fetchmail: POP3> DELE 1
2522 fetchmail: POP3< +OK marked deleted
2525 <p>Check and see if you're allowing sendmail connections through
2528 <p>Adding 'sendmail : 127.0.0.1' to /etc/hosts.allow could solve
2532 <h1>Disappearing mail</h1>
2533 <h2><a id="D1" name="D1">D1. I think I've set up fetchmail
2534 correctly, but I'm not getting any mail.</a></h2>
2536 <p>Maybe you have a .forward or alias set up that you've forgotten
2537 about. You should probably remove it.</p>
2539 <p>Or maybe you're trying to run fetchmail in multidrop mode as
2540 root without a .fetchmailrc file. This doesn't do what you think it
2541 should; see question <a href="#C1">C1</a>.</p>
2543 <p>Or you may not be connecting to the SMTP listener. Run fetchmail
2544 -v and see <a href="#R1">R1</a>.</p>
2546 <p>Or you may have your local user set incorrectly. In the
2550 user 'remoteuser' there with password '*' is 'localuser' here
2553 <p>make sure that 'localuser' does exist and can receive mail.</p>
2555 <h2><a id="D2" name="D2">D2. All my mail seems to disappear after a
2556 dropped connection.</a></h2>
2558 <p>One POP3 daemon used in the Berkeley Unix world that reports
2559 itself as POP3 version 1.004 actually throws the queue away. 1.005
2560 fixed that. If you're running this one, upgrade immediately. (It
2561 also truncates long lines at column 1024.)</p>
2563 <p>Many POP servers, if an interruption occurs, will restore the
2564 whole mail queue after about 10 minutes. Better ones will restore it
2565 right away. If you have an interruption and don't see it right
2566 away, cross your fingers and wait ten minutes before retrying.</p>
2568 <p>Good servers are designed to restore the entire queue, including
2569 messages you have deleted. If you have one of these and it flakes out on
2570 you a lot, try setting a small <code>--fetchlimit</code> value. This
2571 will result in more IP connects to the server, but will mean it actually
2572 executes changes to the queue more often.</p>
2574 <h2><a id="D3" name="D3">D3. Mail that was being fetched when I
2575 interrupted my fetchmail seems to have been vanished.</a></h2>
2577 <p>Fetchmail only sends a delete mail request to the server when
2578 either (a) it gets a positive delivery acknowledgment from the SMTP
2579 listener, or (b) it gets one of the spam-filter errors (see the
2580 description of the <code>antispam></code> option) from the
2581 listener. No interrupt can cause it to lose mail.</p>
2583 <p>However, IMAP2bis has a design problem in that its normal fetch
2584 command marks a message 'seen' as soon as the fetch command to get
2585 it is sent down. If for some reason the message isn't actually
2586 delivered (you take a line hit during the download, or your port 25
2587 listener can't find enough free disk space, or you interrupt the
2588 delivery in mid-message) that 'seen' message can lurk invisibly in
2589 your server mailbox forever.</p>
2591 <p>Workaround: add the '<code>fetchall</code>' keyword to your
2594 <p>Solution: switch to an <a href="http://www.imap.org/">IMAP4</a>
2598 <h1>Multidrop-mode problems</h1>
2599 <h2><a id="M1" name="M1">M1. I've declared local names, but all my
2600 multidrop mail is going to root anyway.</a></h2>
2602 <p>Somehow your fetchmail is never recognizing the hostname part of
2603 recipient names it parses out of Envelope-header lines (or these are
2604 improperly configured) as
2605 matching a name within the designated domains. To check this, run
2606 fetchmail in foreground with -v -v on. You will probably see a lot
2607 of messages with the format "line rejected, %s is not an alias of
2608 the mailserver" or "no address matches; forwarding to %s."</p>
2610 <p>These errors usually indicate some kind of configuration
2613 <p>The easiest workaround is to add a '<code>via</code>' option (if
2614 necessary) and add enough '<code>aka</code>' declarations to cover all
2615 of your mailserver's aliases, then say '<code>no dns</code>'. This will
2616 take DNS out of the picture (though it means mail may be uncollected if
2617 it's sent to an alias of the mailserver that you don't have listed).</p>
2619 <p>Occasionally these errors indicate the sort of header-parsing
2620 problem described in <a href="#M7">M7</a>.</p>
2622 <h2><a id="M2" name="M2">M2. I can't seem to get fetchmail to route
2623 to a local domain properly.</a></h2>
2625 <p>A lot of people want to use fetchmail as a poor man's
2626 internetwork mail gateway, picking up mail accumulated for a whole
2627 domain in a single server mailbox and then routing based on what's
2628 in the To/Cc/Bcc lines.</p>
2630 <p>In general, this is not really a good idea. It would be smarter
2631 to just let the mail sit in the mailserver's queue and use
2632 fetchmail's ETRN or ODMR modes to trigger SMTP sends periodically
2633 (of course, this means you have to poll more frequently than the
2634 mailserver's expiration period). If you can't arrange this, try
2635 setting up a UUCP feed.</p>
2637 <p>If neither of these alternatives is available, multidrop mode
2638 may do (though you <em>are</em> going to get hurt by some mailing
2639 list software; see the caveats under THE USE AND ABUSE OF MULTIDROP
2640 MAILBOXES on the man page, and check what is needed at <a
2641 href="http://home.pages.de/~mandree/mail/multidrop">Matthias
2642 Andree's "Requisites for working multidrop
2643 mailboxes"</a>). If you want to try it, the way to do it is
2644 with the '<code>localdomains</code>' option.</p>
2646 <p>In general, if you use localdomains you need to make sure of two
2649 <p><strong>1. You've actually set up your .fetchmailrc entry to
2650 invoke multidrop mode.</strong></p>
2652 <p>Many people set a '<code>localdomains</code>' list and then
2653 forget that fetchmail wants to see more than one name (or the
2654 wildcard '*') in a '<code>here</code>' list before it will do
2655 multidrop routing.</p>
2657 <p><strong>2. You may have to set 'no envelope'.</strong></p>
2659 <p>Normally, multidrop mode tries to deduce an envelope address
2660 from a message before parsing the To/Cc/Bcc lines (this enables it
2661 to avoid losing to mailing list software that doesn't put a
2662 recipient address in the To lines).</p>
2664 <p>Some ways of accumulating a whole domain's messages in a single
2665 server mailbox mean it all ends up with a single envelope address
2666 that is useless for rerouting purposes. In this particular case, sell
2667 your ISP a clue. If that does not work, you may have to set
2668 '<code>no envelope</code>' to prevent fetchmail from being
2669 bamboozled by this, but a missing envelope makes multidrop routing
2672 <p>Check also answer <a href="#T1">T1</a> on a reliable way to do
2673 multidrop delivery if your ISP (or your mail redirection provider)
2676 <h2><a id="M3" name="M3">M3. I tried to run a mailing list using
2677 multidrop, and I have a mail loop!</a></h2>
2679 <p>This isn't fetchmail's fault. Check your mailing list. If the
2680 list expansion includes yourself or anybody else at your mailserver
2681 (that is, not on the client side) you've created a mail loop. Just
2682 chop the host part off any local addresses in the list.</p>
2684 <p>If you use sendmail, you can check the list expansion with
2685 <code>sendmail -bv</code>.</p>
2687 <h2><a id="M4" name="M4"><strike>M4. My multidrop fetchmail seems to be
2688 having DNS problems.</strike></a></h2>
2690 <p><em>The answer that used to be here no longer applies to
2693 <h2><a id="M5" name="M5">M5. I'm seeing long DNS delays before each
2694 message is processed.</a></h2>
2696 <p>Use the '<code>aka</code>' option to pre-declare as many of your
2697 mailserver's DNS names as you can. When an address's host part
2698 matches an aka name, no DNS lookup needs to be done to check
2701 <p>If you're sure you've pre-declared all of your mailserver's DNS
2702 names, you can use the '<code>no dns</code>' option to prevent
2703 other hostname parts from being looked up at all.</p>
2705 <p>Sometimes delays are unavoidable. Some SMTP listeners try to
2706 call DNS on the From-address hostname as a way of checking that the
2707 address is valid.</p>
2709 <h2><a id="M6" name="M6">M6. How do I get multidrop mode to work
2710 with majordomo?</a></h2>
2712 <p>In order for sendmail to execute the command strings in the
2713 majordomo alias file, it is necessary for sendmail to think that
2714 the mail it receives via SMTP really is destined for a local user
2715 name. A normal virtual-domain setup results in delivery to the
2716 default mailbox, rather than expansion through majordomo.</p>
2718 <p>Michael <michael@bizsystems.com> gave us a recipe for
2719 dealing with this case that pairs a run control file like this:</p>
2722 poll your.pop3.server proto pop3:
2724 localdomains virtual.localdomain1.com virtual.localdomain2.com ...
2725 user yourISPusername is root * here,
2726 password yourISPpassword fetchall
2729 <p>with a hack on your local sendmail.cf like this:</p>
2732 #############################################
2733 # virtual info, local hack for ruleset 98 #
2734 #############################################
2736 # domains to treat as direct mapped local domain
2738 CVvirtual.localdomain1.com virtual.localdomain2.com ...
2739 ---------------------------
2741 -------------------------
2742 # handle virtual users
2744 R$+ <@ $=V . > $: $1 < @ $j . >
2745 R< @ > $+ < @ $=V . > $: $1 < @ $j . >
2746 R< @ > $+ $: $1
2747 R< error : $- $+ > $* $#error $@ $1 $: $2
2748 R< $+ > $+ < @ $+ > $: $>97 $1
2751 <p>This ruleset just strips virtual domain names off the addresses
2752 of incoming mail. Your sendmail must be 8.8 or newer for this to
2753 work. Michael says:</p>
2755 <blockquote>I use this scheme with 2 virtual domains and the
2756 default ISP user+domain and service about 30 mail accounts +
2757 majordomo on my inside pop3 server with fetchmail and sendmail
2760 <h2><a id="M7" name="M7">M7. Multidrop mode isn't parsing envelope
2761 addresses from my Received headers as it should.</a></h2>
2763 <p>It may happen that you're getting what appear to be well-formed
2764 sendmail Received headers, but fetchmail can't seem to extract an
2765 envelope address from them. There can be a couple of reasons for
2768 <h3>Spurious Received lines need to be skipped:</h3>
2770 <p>First, fetchmail might be looking at the wrong Received header.
2771 Normally it looks only on the first one it sees, on the theory that
2772 that one was last added and is going to be the one containing your
2773 mailserver's theory of who the message was addressed to.</p>
2775 <p>Some (unusual) mailserver configurations will generate extra
2776 Received lines which you need to skip. To arrange this, use the
2777 optional skip prefix argument of the 'envelope' option; you may
2778 need to say something like '<code>envelope 1 Received</code>' or
2779 '<code>envelope 2 Received</code>'.</p>
2781 <h3>The 'by' clause doesn't contain a mailserver alias:</h3>
2783 <p>When fetchmail parses a Received line that looks like</p>
2786 Received: from send103.yahoomail.com (send103.yahoomail.com [205.180.60.92])
2787 by iserv.ttns.net (8.8.5/8.8.5) with SMTP id RAA10088
2788 for <ksturgeon@fbceg.org>; Wed, 9 Sep 1998 17:01:59 -0700
2791 <p>it checks to see if 'iserv.ttns.net' is a DNS alias of your
2792 mailserver before accepting 'ksturgeon@fbceg.org' as an envelope
2793 address. This check might fail if your DNS were misconfigured, or
2794 if you were using 'no dns' and had failed to declare iserv.ttns.net
2795 as an alias of your server.</p>
2797 <h2><a id="M8" name="M8">M8. Users are getting multiple copies of
2800 <p>It's a consequence of multidrop. What's happening is that you
2801 have N users subscribed to the same list. The list software sends N
2802 copies, not knowing they will end up in the same multidrop box.
2803 Since they are both locally addressed to all N users, fetchmail
2804 delivers N copies to each user.</p>
2806 <p>Fetchmail tries to eliminate adjacent duplicate messages in a
2807 multidrop mailbox. However, this logic depends on the message-ID
2808 being identical in both copies. It also depends on the two copies
2809 being adjacent in the server mailbox. The former is usually the
2810 case, but the latter condition sometimes fails in a
2811 timing-dependent way if the server was processing multiple incoming
2814 <p>I could eliminate this problem by keeping a list of all
2815 message-IDs received during a poll so far and dropping any message
2816 that matches a seen mail ID. The trouble is that this is an O(N**2)
2817 operation that might significantly slow down the retrieval of large
2820 <p>The real solution however is to make sure that fetchmail can find the
2821 envelope recipient properly, which will reliably prevent this message
2825 <h1>Mangled mail</h1>
2826 <h2><a id="X1" name="X1">X1. Spurious blank lines are appearing in
2827 the headers of fetched mail.</a></h2>
2829 <p>What's probably happening is that the POP/IMAP daemon on your
2830 mailserver is inserting a non-RFC822 header (like X-POP3-Rcpt:) and
2831 something in your delivery path (most likely an old version of the
2832 <em>deliver</em> program, which sendmail often calls to do local
2833 delivery) is failing to recognize it as a header.</p>
2835 <p>This is not fetchmail's problem. The first thing to try is
2836 installing a current version of <em>deliver</em>. If this doesn't
2837 work, try to figure out which other program in your mail path is
2838 inserting the blank line and replace that. If you can't do either
2839 of these things, pick a different MDA (such as maildrop) and
2840 declare it with the '<code>mda</code>' option.</p>
2842 <h2><a id="X2" name="X2">X2. My mail client can't see a Subject
2845 <p>First, see <a href="#X1">X1</a>. This is quite probably the same
2846 problem (X-POP3-Rcpt header or something similar being inserted by
2847 the server and choked on by an old version of
2848 <em>deliver</em>).</p>
2850 <p>The O'Reilly sendmail book does warn that IDA sendmail doesn't
2851 process X- headers correctly. If this is your problem, all I can
2852 suggest is replacing IDA sendmail, because it's broken and not
2853 RFC822 conformant.</p>
2855 <h2><a id="X3" name="X3">X3. Messages containing "From" at the start of
2856 line are being split.</a></h2>
2858 <p>If you know the messages aren't split in your server mailbox,
2859 then this is a problem with your POP/IMAP server, your client-side
2860 SMTP listener or your local delivery agent. Fetchmail cannot split
2863 <p>Some POP server daemons ignore Content-Length headers and split
2864 messages on From lines. We have one report that the 2.1 version of
2865 the BSD popper program (as distributed on Solaris 2.5 and
2866 elsewhere) is broken this way.</p>
2868 <p>You can test this. Declare an mda of 'cat' and send yourself one
2869 piece of mail containing "From" at start of a line. If you see a
2870 split message, your POP/IMAP server is at fault. Upgrade to a more
2873 <p>Sendmail and other SMTP listeners don't split RFC822 messages
2874 either. What's probably happening is either sendmail's local
2875 delivery agent or your mail reader are not quite RFC822-conformant
2876 and are breaking messages on what it thinks are Unix-style From
2877 headers. You can figure out which by looking at your client-side
2878 mailbox with vi or more. If the message is already split in your
2879 mailbox, your local delivery agent is the problem. If it's not,
2880 your mailreader is the problem.</p>
2882 <p>If you can't replace the offending program, take a look at your
2883 sendmail.cf file. There will likely be a line something like</p>
2886 Mlocal, P=/usr/bin/procmail, F=lsDFMShP, S=10, R=20/40, A=procmail -Y -d $u
2889 <p>describing your local delivery agent. Try inserting the 'E'
2890 option in the flags part (the F= string). This will make sendmail
2891 turn each dangerous start-of-line From into a >From, preventing
2892 programs further downstream from acting up.</p>
2894 <h2><a id="X4" name="X4">X4.</a> <a id="generic_mangling"
2895 name="generic_mangling">My mail is being mangled in a new and
2896 different way</a></h2>
2898 <p>The first thing you need to do is pin down what program is doing
2899 the mangling. We don't like getting bug reports about fetchmail
2900 that are actually due to some other program's malfeasance, so
2901 please go through this diagnostic sequence before sending us a
2904 <p>There are five possible culprits to consider, listed here in the
2905 order they pass your mail:</p>
2908 <li>Programs upstream of your server mailbox.</li>
2910 <li>The POP or IMAP server on your mailserver host.</li>
2912 <li>The fetchmail program itself.</li>
2914 <li>Your local sendmail.</li>
2916 <li>Your LDA (local delivery agent), as called by sendmail or
2917 specified by <code>mda</code>.</li>
2920 <p>Often it happens that fetchmail itself is OK, but using it
2921 exposes pre-existing bugs in your downstream software, or your
2922 downstream software has a bad interaction with POP/IMAP. You need
2923 to pin down exactly where the message is being garbled in order to
2924 deduce what is actually going on.</p>
2926 <p>The first thing to do is send yourself a test message, and
2927 retrieve it with a .fetchmailrc entry containing the following (or
2928 by running with the equivalent command-line options):</p>
2931 mda "cat >MBOX" keep fetchall
2934 <p>This will capture what fetchmail gets from the server, except
2935 for (a) the extra Received header line fetchmail prepends, (b)
2936 header address changes due to <code>rewrite</code>, and (c) any
2937 end-of-line changes due to the <code>forcecr</code> and
2938 <code>stripcr</code> options. MBOX will in fact contain what
2939 programs downstream of fetchmail see.</p>
2941 <p>The most common causes of mangling are bugs and
2942 misconfigurations in those downstream programs. If MBOX looks
2943 unmangled, you will know that is what is going on and that it is
2944 not fetchmail's problem. Take a look at the other FAQ items in this
2945 section for possible clues about how to fix your problem.</p>
2947 <p>If MBOX looks mangled, the next thing to do is compare it with
2948 your actual server mailbox (if possible). That's why you specified
2949 <code>keep</code>, so the server copy would not be deleted. If your
2950 server mailbox looks mangled, programs upstream of your server
2951 mailbox are at fault. Unfortunately there is probably little you
2952 can do about this aside from complaining to your site postmaster,
2953 and nothing at all fetchmail can do about it!</p>
2955 <p>More likely you'll find that the server copy looks OK. In that
2956 case either the POP/IMAP server or fetchmail is doing the mangling.
2957 To determine which, you'll need to telnet to the server port and
2958 simulate a fetchmail session yourself. This is not actually hard
2959 (both POP3 and IMAP are simple, text-only, line-oriented protocols)
2960 but requires some attention to detail. You should be able to use a
2961 fetchmail -v log as a model for a session, but remember that the
2962 "*" in your LOGIN or PASS command dump has to be replaced with your
2963 actual password.</p>
2965 <p>The objective of manually simulating fetchmail is so you can see
2966 exactly what fetchmail sees. If you see a mangled message, then
2967 your server is at fault, and you probably need to complain to your
2968 mailserver administrators. However, we like to know what the broken
2969 servers are so we can warn people away from them. So please send us
2970 a transcript of the session including the mangling <em>and the
2971 server's initial greeting line</em>. Please tell us anything else
2972 you think might be useful about the server, like the server host's
2973 operating system.</p>
2975 <p>If your manual fetchmail simulation shows an unmangled message,
2976 congratulations. You've found an actual fetchmail bug, which is a
2977 pretty rare thing these days. Complain to us and we'll fix it.
2978 Please include the session transcript of your manual fetchmail
2979 simulation along with the other things described in the FAQ entry
2980 on <a href="#G3">reporting bugs</a>.</p>
2982 <h2><a id="X5" name="X5"><strike>X5. Using POP3, retrievals seems to be
2983 fetching too much!</strike></a></h2>
2985 <p><em>The information that used to be here pertained to fetchmail 4.4.7 or
2986 older, which should not be used. Use a recent fetchmail version.</em></p>
2988 <h2><a id="X6" name="X6">X6. My mail attachments are being dropped
2989 or mangled.</a></h2>
2991 <p>Fetchmail doesn't discard attachments; fetchmail doesn't have any idea
2992 that attachments are there. Fetchmail treats the body of each message as
2993 an uninterpreted byte stream and passes it through without alteration.
2994 If you are not receiving attachments through fetchmail, it is because
2995 your mailserver is not sending them to you.</p>
2997 <p>The fix for this is to replace your mailserver with one that works.
2998 If its operating system makes this difficult, you should replace its
2999 operating system with one that works. Windows- and NT-based POP servers
3000 seem especially prone to mangle attachments. If you are running one
3001 of these, replacing your server with a Unix machine is probably the
3002 only effective solution.</p>
3004 <p>We've had sporadic reports of problems with Microsoft Exchange and
3005 Outlook servers. These sometimes randomly fail to ship
3006 attachments to your client. This is a known bug, acknowledged by
3009 <p>They may also mangle the attachments they do pass through. If you
3010 see unreadable attachments with a ContentType of "application/x-tnef",
3011 you're having this problem. The <a
3012 href="http://world.std.com/~damned/software.html">TNEF</a> utility may
3015 <p>The Mail Max POP3 server and the InterChange and Imail IMAP
3016 servers are known to simply drop MIME attachments when uploading
3019 <p>We've also had a report that Lotus Notes sometimes trashes the
3020 MIME type of messages. In particular, it seems to modify MIME
3021 headers of type application/pdf, mangling the type to
3022 application/octet-stream. It may corrupt other MIME types as
3025 <p>The IMAP service of Lotus Domino has a known bug in the way it
3026 generates MIME Content-type headers (observed on Lotus Domino
3027 5.0.2b). It's a subtle one that doesn't show up when Netscape
3028 Messenger and other clients use a FETCH BODY[] to grab the whole
3029 message. When fetchmail uses FETCH RFC822.HEADER and FETCH
3030 RFC822.TEXT to get first the header and then the body, Domino
3031 generates different Boundary tags for each part, e.g. one tag is
3032 declared in the Content-type header and another is used to separate
3033 the MIME parts in the body. This doesn't work. (I have heard a
3034 rumor that this bug is scheduled to be fixed in Domino release 6;
3035 you can find a workaround at contrib/domino.)</p>
3037 <p>Rob Funk explains: Unfortunately there also remain many mail
3038 user agents that don't write correct MIME messages. One big
3039 offender is Sun MailTool attachments, which are formatted enough
3040 like MIME that some programs could get confused; these are
3041 generated by the mailtool and dtmail programs (the mail programs in
3042 Sun's OpenWindows and CDE environments).</p>
3044 <p>One solution to problems related to misformatted MIME
3045 attachments is the <a
3046 href="ftp://ftp.uu.se/pub/unix/networking/mail/emil/">emil</a>
3048 href="ftp://ftp.uu.se/pub/unix/networking/mail/emil/TUTORIAL.html">tutorial</a>
3049 file at that site for details on emil. It is useful for converting
3050 character sets, attachment encodings, and attachment formats. At
3051 this writing, emil does not appear to have been maintained since a
3052 patch to version 2.1.0beta9 in late 1997, but it is still
3055 <p>One good way of using emil is from within procmail. You can have
3056 procmail look for signs of problematic message formatting, and pipe
3057 those messages through emil to be fixed. emil will not always be
3058 able to fix the problem, in which case the message is
3061 <p>A possible rule to be inserted into a .procmailrc file for using
3066 * 1^1 ^Content-Type: \/X-sun[^;]*
3067 * 1^1 ^Content-Type: \/application/mac-binhex[^;]*
3068 * 1^1 ^Content-Transfer-Encoding: \/x-binhex[^;]*
3069 * 1^1 ^Content-Transfer-Encoding: \/x-uuencode[^;]*
3071 LOG="Converting $MATCH
3074 | emil -A B -T Q -B BA -C iso-8859-1 -H Q -F MIME \
3075 | gawk '{gsub(/\r\n?/,"\n");print $0}'
3079 <p>The "1^1" in the conditions is a way of specifying to procmail
3080 that if any one of the four listed expressions is found in the
3081 message, the total condition is considered true, and the message
3082 gets passed into emil. These four subconditions check whether the
3083 message has a Sun attachment, a binhex attachment, or a uuencoded
3084 attachment; there are others that could be added to check these
3085 things better and to check other relevant conditions. The "LOG="
3086 line writes a line into the procmail log; the lone double-quote
3087 beginning the following line makes sure the log entry gets an
3088 end-of-line character. The call to gawk (GNU awk) is for fixing
3089 end-of-line conventions, since emil sometimes leaves those in the
3090 format of the originating machine; it could probably be replaced
3091 with a sed subsitution.</p>
3093 <p>The emil call itself tries to ensure that the message uses:</p>
3096 <li>BinHex encoding for any Apple Macintosh-only attachments</li>
3098 <li>Quoted-Printable encoding for text (when necessary)</li>
3100 <li>Base64 Encoding for binary attachments</li>
3102 <li>iso-8859-1 character set for text (unfortunately emil can't yet
3103 convert from windows-1252 to iso-8859-1)</li>
3105 <li>Quoted-Printable encoding for headers</li>
3107 <li>MIME attachment format</li>
3110 <p>Most of these (the primary exceptions being the character set
3111 and the Apple binary format) are as they should be for good
3112 internet interoperability.</p>
3114 <p>Some mail servers (Lotus Domino is a suspect here) mangle
3115 Sun-formatted messages, so the conversion to MIME needs to happen
3116 before such programs see the message. The ideal is to rid the world
3117 of Sun-formatted messages: don't use mailtool for sending
3118 attachments (it doesn't understand MIME anyway, and most of the
3119 world doesn't understand its attachments, so it really shouldn't be
3120 used at all), and make sure dtmail is set to use MIME rather than
3121 mailtool's format.</p>
3123 <h2><a id="X7" name="X7">X7. Some mail attachments are hanging
3126 <p>This isn't fetchmail's problem either; fetchmail doesn't know
3127 anything about mail attachments and doesn't treat them any
3128 differently from plain message data.</p>
3130 <p>The most usual cause of this problem seems to be bugs in your
3131 network transport layer's capability to handle the very large
3132 TCP/IP packets that attachments tend to turn into. You can test
3133 this theory by trying to download the offending message through a
3134 webmail account; using HTTP for the message tends to simulate
3135 large-packet stress rather well, and you will probably find that
3136 the messages that seem to be choking fetchmail will make your HTTP
3137 download speed drop to zero.</p>
3139 <p>This problem can be caused by subtle bugs in the
3140 packet-reassembly layer of your TCP/IP stack; these often don't
3141 manifest at normal packet sizes. It may also be caused by
3142 malfunctioning path-MTU discovery on the mailserver. Or, if there's
3143 a modem in the link, it may be because the attachment contains the
3144 Hayes mode escape "+++".</p>
3146 <h2><a id="X8" name="X8">X8. A spurious ) is being appended to my
3149 <p>Blame it on that rancid pile of dung and offal called Microsoft
3150 Exchange. Due to the problem described in <a href="#S2">S2</a>, the
3151 IMAP support in fetchmail cannot follow the IMAP protocol 100%.
3152 Most of the time it doesn't matter, but if you combine it with an
3153 SMTP server that behaves unusually, you'll get a spurious ) at
3156 <p>One piece of software that can trigger this is the Interchange
3157 mail server, as used by, e.g., mailandnews.com. Here's what
3160 <p>1. Someone sends mail to your account. The last line of the
3161 message contains text. So at the SMTP level, the message ends with,
3162 e.g. "blahblah\r\n.\r\n"</p>
3164 <p>2. The SMTP handler sees the final "\r\n.\r\n" and recognizes
3165 the end of the message. However, instead of doing the normal thing,
3166 which is tossing out the ".\r\n" and leaving the first '\r\n' as
3167 part of the email body, Interchange throws out the whole
3168 "\r\n.\r\n", and leaves the email body without any line terminator
3169 at the end of it. RFC821 does not forbid this, though it probably
3172 <p>3. Fetchmail, or some other IMAP client, asks for the message.
3173 IMAP returns it, but it's enclosed inside parentheses, according to
3174 the protocol. The message size in bytes is also present. Because
3175 the message doesn't end with a line terminator, the IMAP client
3183 <p>where the ')' is from IMAP.</p>
3185 <p>4. Fetchmail only deals with complete lines, and can't trust the
3186 stated message size because Microsoft Exchange fscks it up.</p>
3188 <p>5. As a result, fetchmail takes the final 'blahblah)' and puts
3189 it at the end of the message it forwards on. If you have verbosity
3190 on, you'll get a message about actual != expected.</p>
3192 <p>There is no fix for this. The nuke mentioned in <a
3193 href="#S2">S2</a> looks more tempting all the time.</p>
3195 <h2><a id="X9" name="X9">X9. Missing "Content-Transfer-Encoding" header
3196 with Domino IMAP</a></h2>
3198 <p>Domino 6 IMAP was found by Anthony Kim in February 2006 to
3199 erroneously omit the "Content-Transfer-Encoding" header in messages
3200 downloaded through IMAP, causing messages to display improperly. This
3201 happened with Domino's incoming mail format configured to "Prefers
3202 MIME". Solution: switch Domino to "Keep in Sender's format".</p>
3205 href="http://lists.ccil.org/pipermail/fetchmail-friends/2006-March/010015.html">Anthony
3210 <h1>Other problems</h1>
3211 <h2><a id="O1" name="O1">O1. The --logfile option doesn't work if
3212 the logfile doesn't exist.</a></h2>
3214 <p>This is a feature, not a bug. It's in line with normal practice
3215 for system daemons and allows you to suppress logging by removing
3216 the log file, without hacking potentially fragile startup scripts.
3217 To get around it, just touch(1) the logfile before you run fetchmail
3218 (this will have no effect on the contents of the logfile if it already
3221 <h2><a id="O2" name="O2">O2. Every time I get a POP or IMAP message,
3222 the header is dumped to all my terminal sessions.</a></h2>
3224 <p>Fetchmail uses the local sendmail to perform final delivery,
3225 which Mozilla and other clients don't do; the announcement of
3226 new messages is done by a daemon that sendmail pokes. There should
3227 be a "biff" command to control this. Type</p>
3233 <p>to turn it off. If this doesn't work, try the command</p>
3239 <p>which is essentially what <code>biff -n</code> will do. If this
3240 doesn't work, comment out any reference to "comsat" in your
3241 /etc/inetd.conf file and reload (or restart) inetd.</p>
3243 <p>In Slackware Linux distributions, the last line in /etc/profile
3256 to solve the problem system-wide.
3258 <h2><a id="O3" name="O3">O3. Does fetchmail reread its rc file
3259 every poll cycle?</a></h2>
3261 <p>No, but versions 5.2.2 and later will notice when you modify
3262 your rc file and restart, reading it. Note that this causes troubles if
3263 you need to provide a password via the console, unless you're running in
3264 --nodetach mode.</p>
3266 <h2><a id="O4" name="O4">O4. Why do deleted messages show up again
3267 when I take a line hit while downloading?</a></h2>
3269 <p>According to the POP3 RFCs, deletes aren't actually performed
3270 until you issue the end-of-session QUIT command. Fetchmail cannot
3271 fix this, but there is a workaround: use the --expunge option with a
3272 reasonably low figure that works for you. Try 10 for a start.</p>
3274 <p>IMAP is less susceptible to this problem, because the "deleted"
3275 message marks are persistent, but they aren't in POP3. Note that the
3276 --expunge default for IMAP is different than the default for POP3.</p>
3278 <p>If you get very unlucky, you might take a line hit in the window
3279 between the delete and the expunge. If you've set a longer expunge
3280 interval, the window gets wider. This problem should correct itself
3281 the next time you complete a successful query.</p>
3283 <h2><a id="O5" name="O5">O5. Why is fetched mail being logged with
3284 my name, not the real From address?</a></h2>
3286 <p>Because logging is done based on the address indicated by the
3287 sending SMTP's MAIL FROM, and some listeners are picky about that
3290 <p>Some SMTP listeners get upset if you try to hand them a MAIL
3291 FROM address naming a different host than the originating site for
3292 your connection. This is a feature, not a bug -- it's supposed to
3293 help prevent people from forging mail with a bogus origin site.
3294 (RFC 1123 says you shouldn't do this exclusion...)</p>
3296 <p>Since the originating site of a fetchmail delivery connection is
3297 localhost, this effectively means these picky listeners will barf
3298 on any MAIL FROM address fetchmail hands them with an @ in it!</p>
3300 <p>Versions 2.1 and up try the header From address first and fall
3301 back to the calling-user ID. So if your SMTP listener isn't picky,
3302 the log will look right.</p>
3304 <h2><a id="O6" name="O6">O6. I'm seeing long sendmail delays or
3305 hangs near the start of each poll cycle.</a></h2>
3307 <p>Sendmail does a hostname lookup when it first starts up, and
3308 also each time it gets a HELO in listener mode.</p>
3310 <p>Your resolver configuration may be causing one of these lookups
3311 to fail and time out. Check your <code>/etc/resolv.conf</code>,
3312 <code>/etc/host.conf</code>, <code>/etc/nsswitch.conf</code> (if you
3313 have the latter two) and you <code>/etc/hosts</code> files. Make sure
3314 your hostname and fully-qualified domain name are both in
3315 <code>/etc/hosts</code>, and that hosts is looked at before DNS is
3316 queried. You probably also want your remote mail server(s) to be in the
3319 <p>You can suppress the startup-time lookup if need to by reconfiguring
3320 with <code>FEATURE(nodns)</code>.</p>
3322 <p>Configuring your bind library to cache DNS lookups locally may
3323 help, and is a good idea for speeding up other services as well.
3324 Switching to a faster MTA like <a
3325 href="http://www.postfix.org/">Postfix</a> might help.</p>
3327 <h2><a id="O7" name="O7">O7. Why doesn't fetchmail deliver mail in
3328 date-sorted order?</a></h2>
3330 <p>Because that's not the order the server hands it to fetchmail
3333 <p>Fetchmail getting mail from a POP server delivers mail in the
3334 order that your server delivers mail. Fetchmail can't do anything
3335 about this; it's a limitation of the underlying POP protocol.</p>
3337 <p>In theory it might be possible for fetchmail in IMAP mode to
3338 sort messages by date, but this would be in violation of two basics
3339 of fetchmail's design philosophy: (a) to be as simple and
3340 transparent a pipe as possible, and (b) to <em>hide</em>, rather
3341 than emphasize, the differences between the remote-fetch protocols
3344 <p>Re-ordering messages is a user-agent function, anyway.</p>
3346 <h2><a id="O8" name="O8">O8. I'm using pppd. Why isn't my monitor
3347 option working?</a></h2>
3349 <p>There is a combination of circumstances that can confuse
3350 fetchmail. If you have set up demand dialing with pppd, and pppd
3351 has an idle timeout, and you have lcp-echo-interval set, then the
3352 lcp-echo-interval time must be longer than the pppd idle timeout.
3353 Otherwise it is going keep increasing the packet counters that
3354 fetchmail relies upon, triggering fetchmail into polling after its
3355 own delay interval and thus preventing the pppd link from ever
3356 reaching its inactivity timeout.</p>
3358 <h2><a id="O9" name="O9">O9. Why does fetchmail keep retrieving the
3359 same messages over and over?</a></h2>
3361 <p>First, check to see that you haven't enabled the
3362 <cite>keep</cite> and <cite>fetchall</cite> option. If you have,
3363 turn one of them off - which one, depends on why they have been set in
3364 the first place, and to a lesser degree on the upstream server.</p>
3366 <p>This can also happen when some other mail client is logged in to
3367 your mail server, if it uses a simple exclusive-locking scheme (and
3368 many, especially most POP3 servers, do exactly that). Your
3369 fetchmail is able to retrieve the messages, but because the mailbox
3370 is write-locked by the other instance yours can neither mark
3371 messages seen or delete them. The solution is to either (a) wait
3372 for the other client to finish, or (b) terminate it.</p>
3374 <h2><a id="O10" name="O10"><strike>O10. Why is the received date on all my
3375 messages the same?</strike></a></h2>
3377 <p><em>The answer that used to be here made no sense and was dropped.</em></p>
3379 <h2><a name="O11">O11. I keep getting messages that say "Repoll
3380 immediately" in my logs.</a></h2>
3382 <p>This is your server barfing on the CAPA probe that fetchmail sends.
3383 Because some servers like to drop the connection after that probe,
3384 fetchmail will re-poll immediately with this probe defeated.</p>
3386 <p>If you run fetchmail in daemon mode (say "set daemon 600"), you will
3387 get the message only once per run.</p>
3389 <p>If you set an authentication method explicitly (say, with
3390 <code>auth password</code>), you will never get the message.</p>
3392 <h2><a name="O12">O12. Fetchmail no longer expunges mail on a 451 SMTP response.</a></h2>
3394 <p>This is a feature, not a bug.</p>
3396 <p>Any 4xx response (like 451) indicates a transient (temporary) error.
3397 This means that the mail could be accepted if retried later. Lookup
3398 failures are normally transient errors as a mail should not get
3399 rejected if a dns server is unreachable or down.</p>
3401 <p>A permanent reject response is of the form 5xx (like 550).</p>
3403 <p>You could tell your SMTP server to not lookup any addresses if you are
3404 not keen on checking the sender addresses. This problem typically
3405 occurs if your mail server is not checking the sender addresses, but
3406 your local server is.</p>
3408 <p>Or you could declare <code>antispam 451</code>, which is not
3409 recommended though, as it may cause mail loss.</p>
3411 <p>Or, you could check your nameserver configuration and query logs for
3414 <p>All these issues are not related to fetchmail directly.</p>
3416 <h2><a name="O13">O13. I want timestamp information in my fetchmail logs.</a></h2>
3418 <p>Write a <code>preconnect</code> command in your configuration file that
3419 does something like "date >> $HOME/fetchmail.log".</p>
3421 <h2><a name="O14">O14. Fetchmail no longer deletes oversized mails with
3424 <p>Use <code>--limitflush</code> (available since release 6.3.0) to
3425 delete oversized mails along with the <code>--limit</code> option. If
3426 you are already having <code>flush</code> in your rcfile to delete
3427 oversized mails, <em>replace</em> it with <code>limitflush</code> to
3428 avoid losing mails unintentionally.</p>
3430 <p>The <code>--flush</code> option is primarily designed to delete
3431 mails which have been read/downloaded but not deleted yet. This option
3432 cannot be overloaded to delete oversized mails as it cannot be guessed
3433 whether the user wants to delete only read/downloaded mails or only
3434 oversized mails or both when a user specifies both
3435 <code>--limit</code> and <code>--flush</code>. Hence, a separate
3436 <code>--limitflush</code> has been added to resolve the ambiguity.</p>
3438 <h2><a name="O15">O15. Fetchmail always retains the first message in the
3441 <p>This happens when fetchmail sees an "X-IMAP:" header in the very
3442 first message in your mailbox. This usually stems from a message like
3443 the one shown below, which is automatically created on your server. This
3444 message shows up if the University of Washington IMAP or PINE software
3445 is used on the server together with a POP2 or POP3 daemon that is not
3446 aware of these messages, such as some versions of Qualcomm Popper
3451 From MAILER-DAEMON Wed Nov 23 11:38:42 2005
3452 Date: 23 Nov 2005 11:38:42 +0100
3453 From: Mail System Internal Data <MAILER-DAEMON@imap.example.org>
3454 Subject: DON'T DELETE THIS MESSAGE -- FOLDER INTERNAL DATA
3455 Message-ID: <1132742322@imap.example.org>
3456 X-IMAP: 1132742306 0000000001
3459 This text is part of the internal format of your mail folder, and is not
3460 a real message. It is created automatically by the mail system software.
3461 If deleted, important folder data will be lost, and it will be re-created
3462 with the data reset to initial values.
3466 <p>As this message does not contain useful information, fetchmail is not
3467 retrieving it. And deleting it might slow down the server if you are
3468 keeping messages on the server, and the server would recreate it
3469 anyways, that's why fetchmail does not bother to delete it either.</p>
3471 <h2><a name="O16">O16. Why is the Fetchmail FAQ only available in
3472 ISO-216 A4 format? How do I get the FAQ in Letter format?</a></h2>
3474 <p>All the world uses ISO-216:1975 "A4" paper except for North America.
3475 Using A4 format reaches far more people than (formerly known as DIN A4,
3476 from DIN 476) format. Besides that, A4 paper <em>is</em> available in North
3478 For further information on the Letter-vs-A4 story, see:</p>
3479 <ul><li><a href="http://www.cl.cam.ac.uk/~mgk25/iso-paper.html">Markus
3480 Kuhn: "International standard paper sizes"</a></li>
3482 href="http://betweenborders.com/wordsmithing/a4-vs-us-letter/">Brian
3483 Forte: "A4 vs US Letter"</a></li></ul>
3485 <p>Offering the document formatted for two different paper sizes would
3486 bloat the package beyond reason, and formatting in a way that fits A4
3487 and Letter paper formats would be a waste of paper in most parts of the
3488 world. For that reason, fetchmail only ships with an A4 formatted PDF
3491 <p>To create a letter-sized PDF, install <a
3492 href="http://www.htmldoc.org/">HTMLDOC</a>, edit
3493 <code>fetchmail-FAQ.book</code> in the source directory with your
3494 favorite text editor, replace <samp>--size A4</samp> by <samp>--size
3495 letter</samp>, and type:
3498 make fetchmail-FAQ.pdf
3503 <table width="100%" cellpadding="0" summary="Canned page footer">
3505 <td width="30%">Back to <a href="index.html">Fetchmail Home
3507 <td width="30%" align="right">$Date$</td>
3512 <address>Eric S. Raymond <a
3513 href="mailto:esr@thyrsus.com"><esr@thyrsus.com></a><br />
3514 Matthias Andree</address>