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">
4 <html xmlns="http://www.w3.org/1999/xhtml">
6 <title>The Fetchmail FAQ</title>
7 <meta name="description"
8 content="Frequently asked questions about fetchmail."/>
9 <meta name="keywords" content="fetchmail, POP, POP2, POP3, IMAP, remote mail"/>
12 <table width="100%" cellpadding="0" summary="Canned page footer">
14 <td width="30%">Back to <a href="index.html">Fetchmail Home
16 <td width="30%" align="right">$Date$</td>
21 <h1>Frequently Asked Questions About Fetchmail</h1>
23 <p>Before reporting any bug, please read <a href="#G3">G3</a> for
24 advice on how to include diagnostic information that will get your
25 bug fixed as quickly as possible.</p>
27 <p>If you have a question or answer you think ought to be added to
28 this FAQ list, file it to one of the trackers at <a
29 href="http://developer.berlios.de/projects/fetchmail/">our BerliOS
30 project site</a> or post to one of the fetchmail mailing lists (see
33 <h1>General questions:</h1>
35 <a href="#G1">G1. What is fetchmail and why should I bother?</a><br/>
36 <a href="#G2">G2. Where do I find the latest FAQ and fetchmail sources?</a><br/>
37 <a href="#G3">G3. I think I've found a bug. Will you fix it?</a><br/>
38 <a href="#G4">G4. I have this idea for a neat feature. Will you add it?</a><br/>
39 <a href="#G5">G5. I want to make fetchmail behave like Outlook Express.</a><br/>
40 <a href="#G6">G6. Is there a mailing list for exchanging tips?</a><br/>
41 <a href="#G7">G7. So, what's this I hear about a fetchmail paper?</a><br/>
42 <a href="#G8">G8. What is the best server to use with fetchmail?</a><br/>
43 <a href="#G9">G9. What is the best mail program to use with fetchmail?</a><br/>
44 <a href="#G10">G10. How can I avoid sending my password en clair?</a><br/>
45 <a href="#G11">G11. Is any special configuration needed to use a dynamic IP address?</a><br/>
46 <a href="#G12">G12. Is any special configuration needed to use firewalls?</a><br/>
47 <a href="#G13">G13. Is any special configuration needed to <em>send</em> mail?</a><br/>
48 <a href="#G14">G14. Is fetchmail Y2K-compliant?</a><br/>
49 <a href="#G15">G15. Is there a way in fetchmail to support disconnected IMAP mode?</a><br/>
50 <a href="#G16">G16. How will fetchmail perform under heavy loads?</a><br/>
53 <h1>Build-time problems:</h1>
55 <a href="#B1">B1. Make coughs and dies when building on FreeBSD.</a><br/>
56 <a href="#B2">B2. Lex bombs out while building the fetchmail lexer.</a><br/>
57 <a href="#B3">B3. I get link failures when I try to build fetchmail.</a><br/>
58 <a href="#B4">B4. I get build failures in the intl directory.</a><br/>
60 <h1>Fetchmail configuration file grammar questions:</h1>
62 <a href="#F1">F1. Why does my old .fetchmailrc no longer work?</a><br/>
63 <a href="#F2">F2. The .fetchmailrc parser won't accept my all-numeric user name.</a><br/>
64 <a href="#F3">F3. The .fetchmailrc parser won't accept my host or username beginning with 'no'.</a><br/>
65 <a href="#F4">F4. I'm getting a 'parse error' message I don't understand.</a><br/>
67 <h1>Configuration questions:</h1>
69 <a href="#C1">C1. Why do I need a .fetchmailrc when running as root
70 on my own machine?</a><br/>
71 <a href="#C2">C2. How can I arrange for a fetchmail daemon to get
72 killed when I log out?</a><br/>
73 <a href="#C3">C3. How do I know what interface and address to use
74 with --interface?</a><br/>
75 <a href="#C4">C4. How can I set up support for sendmail's anti-spam
77 <a href="#C5">C5. How can I poll some of my mailboxes more/less
78 often than others?</a><br/>
79 <a href="#C6">C6. Fetchmail works OK started up manually, but not
80 from an init script.</a><br/>
81 <a href="#C7">C7. How can I forward mail to another
85 <h1>How to make fetchmail play nice with various MTAs:</h1>
87 <a href="#T1">T1. How can I use fetchmail with sendmail?</a><br/>
88 <a href="#T2">T2. How can I use fetchmail with qmail?</a><br/>
89 <a href="#T3">T3. How can I use fetchmail with exim?</a><br/>
90 <a href="#T4">T4. How can I use fetchmail with smail?</a><br/>
91 <a href="#T5">T5. How can I use fetchmail with SCO's MMDF?</a><br/>
92 <a href="#T6">T6. How can I use fetchmail with Lotus Notes?</a><br/>
93 <a href="#T7">T7. How can I use fetchmail with Courier IMAP?</a><br/>
94 <a href="#T8">T8. How can I use fetchmail with vbmailshield?</a><br/>
96 <h1>How to make fetchmail work with various servers:</h1>
98 <a href="#S1">S1. How can I use fetchmail with qpopper?</a><br/>
99 <a href="#S2">S2. How can I use fetchmail with Microsoft Exchange?</a><br/>
100 <a href="#S3">S3. How can I use fetchmail with HP OpenMail?</a><br/>
101 <a href="#S4">S4. How can I use fetchmail with Novell GroupWise?</a><br/>
102 <a href="#S5">S5. How can I use fetchmail with InterChange?</a><br/>
103 <a href="#S6">S6. How can I use fetchmail with MailMax?</a><br/>
104 <a href="#S7">S7. How can I use fetchmail with FTGate?</a><br/>
106 <h1>How to fetchmail work with specific ISPs:</h1>
108 <a href="#I1">I1. How can I use fetchmail with Compuserve RPA?</a><br/>
109 <a href="#I2">I2. How can I use fetchmail with Demon Internet's SDPS?</a><br/>
110 <a href="#I3">I3. How can I use fetchmail with usa.net's servers?</a><br/>
111 <a href="#I4">I4. How can I use fetchmail with geocities POP3 servers?</a><br/>
112 <a href="#I5">I5. How can I use fetchmail with Hotmail or Lycos Webmail?</a><br/>
113 <a href="#I6">I6. How can I use fetchmail with MSN?</a><br/>
114 <a href="#I7">I7. How can I use fetchmail with SpryNet?</a><br/>
115 <a href="#I8">I8. How can I use fetchmail with comcast.net?</a><br/>
117 <h1>How to set up well-known security and authentication
120 <a href="#K1">K1. How can I use fetchmail with SOCKS?</a><br/>
121 <a href="#K2">K2. How can I use fetchmail with IPv6 and IPsec?</a><br/>
122 <a href="#K3">K3. How can I get fetchmail to work with ssh?</a><br/>
123 <a href="#K4">K4. What do I have to do to use the IMAP-GSS protocol?</a><br/>
124 <a href="#K5">K5. How can I use fetchmail with SSL?</a><br/>
125 <a href="#K6">K6. How can I tell fetchmail not to try TLS if the server
126 advertises it?</a><br/>
128 <h1>Runtime fatal errors:</h1>
130 <a href="#R1">R1. Fetchmail isn't working, and -v shows 'SMTP
131 connect failed' messages.</a><br/>
132 <a href="#R2">R2. When I try to configure an MDA, fetchmail doesn't
134 <a href="#R3">R3. Fetchmail dumps core when given an invalid rc
136 <a href="#R4">R4. Fetchmail dumps core in -V mode, but operates
137 normally otherwise.</a><br/>
138 <a href="#R5">R5. Running fetchmail in daemon mode doesn't
140 <a href="#R6">R6. Fetchmail randomly dies with socket errors.</a><br/>
141 <a href="#R7">R7. Fetchmail running as root stopped working after
142 an OS upgrade</a><br/>
143 <a href="#R8">R8. Fetchmail is timing out after fetching certain
144 messages but before deleting them</a><br/>
145 <a href="#R9">R9. Fetchmail is timing out during message fetches</a><br/>
146 <a href="#R10">R10. Fetchmail is dying with SIGPIPE.</a><br/>
147 <a href="#R11">R11. My server is hanging or emitting errors on CAPA.</a><br/>
148 <a href="#R12">R12. Fetchmail isn't working and reports getaddrinfo
150 <h1>Hangs and lockups:</h1>
152 <a href="#H1">H1. Fetchmail hangs when used with pppd.</a><br/>
153 <a href="#H2">H2. Fetchmail hangs during the MAIL FROM
155 <a href="#H3">H3. Fetchmail hangs while fetching mail.</a><br/>
158 <h1>Disappearing mail:</h1>
160 <a href="#D1">D1. I think I've set up fetchmail correctly, but I'm
161 not getting any mail.</a><br/>
162 <a href="#D2">D2. All my mail seems to disappear after a dropped
164 <a href="#D3">D3. Mail that was being fetched when I interrupted my
165 fetchmail seems to have been vanished.</a><br/>
168 <h1>Multidrop-mode problems:</h1>
170 <a href="#M1">M1. I've declared local names, but all my multidrop
171 mail is going to root anyway.</a><br/>
172 <a href="#M2">M2. I can't seem to get fetchmail to route to a local
173 domain properly.</a><br/>
174 <a href="#M3">M3. I tried to run a mailing list using multidrop,
175 and I have a mail loop!</a><br/>
176 <a href="#M4">M4. My multidrop fetchmail seems to be having DNS
178 <a href="#M5">M5. I'm seeing long DNS delays before each message is
180 <a href="#M6">M6. How do I get multidrop mode to work with
182 <a href="#M7">M7. Multidrop mode isn't parsing envelope addresses
183 from my Received headers as it should.</a><br/>
184 <a href="#M8">M8. Users are getting multiple copies of
188 <h1>Mangled mail:</h1>
190 <a href="#X1">X1. Spurious blank lines are appearing in the headers
191 of fetched mail.</a><br/>
192 <a href="#X2">X2. My mail client can't see a Subject
194 <a href="#X3">X3. Messages containing "From" at start of line are
195 being split.</a><br/>
196 <a href="#X4">X4. My mail is being mangled in a new and different
198 <a href="#X5">X5. Using POP3, retrievals seems to be fetching too
200 <a href="#X6">X6. My mail attachments are being dropped or
202 <a href="#X7">X7. Some mail attachments are hanging
204 <a href="#X8">X8. A spurious ) is being appended to my
207 <h1>Other problems:</h1>
209 <a href="#O1">O1. The --logfile option doesn't work if the logfile
210 doesn't exist.</a><br/>
211 <a href="#O2">O2. Every time I get a POP or IMAP message the header
212 is dumped to all my terminal sessions.</a><br/>
213 <a href="#O3">O3. Does fetchmail reread its rc file every poll
215 <a href="#O4">O4. Why do deleted messages show up again when I take
216 a line hit while downloading?</a><br/>
217 <a href="#O5">O5. Why is fetched mail being logged with my name,
218 not the real From address?</a><br/>
219 <a href="#O6">O6. I'm seeing long sendmail delays or hangs near the
220 start of each poll cycle.</a><br/>
221 <a href="#O7">O7. Why doesn't fetchmail deliver mail in date-sorted
223 <a href="#O8">O8. I'm using pppd. Why isn't my monitor option
225 <a href="#O9">O9. Why does fetchmail keep retrieving the same
226 messages over and over?</a><br/>
227 <a href="#O10">O10. Why is the received date on all my messages the
229 <a href="#O11">O11. I keep getting messages that say "Repoll
230 immediately" in my logs.</a><br/>
231 <a href="#O12">O12. Fetchmail no longer expunges mail on a 451 SMTP response.</a><br/>
232 <a href="#O13">O13. I want timestamp information in my fetchmail logs.</a><br/>
233 <a href="#O14">O14. Fetchmail no longer deletes oversized mails with
239 <h2><a id="G1" name="G1">G1. What is fetchmail and why should I
242 <p>Fetchmail is a one-stop solution to the remote mail retrieval
243 problem for Unix machines, quite useful to anyone with an
244 intermittent PPP or SLIP connection to a remote mailserver. It can
245 collect mail using any variant of POP or IMAP and forwards via port
246 25 to the local SMTP listener, enabling all the normal
247 forwarding/filtering/aliasing mechanisms that would apply to local
248 mail or mail arriving via a full-time TCP/IP connection.</p>
250 <p>Fetchmail is not a toy or a coder's learning exercise, but an
251 industrial-strength tool capable of transparently handling every
252 retrieval demand from those of a simple single-user ISP connection
253 up to mail retrieval and rerouting for an entire client domain.
254 Fetchmail is easy to configure, unobtrusive in operation, powerful,
255 feature-rich, and well documented.</p>
257 <p>Fetchmail is <a href="http://www.opensource.org">open-source</a>
258 software. The openness of the sources is the strongest assurance of
259 quality you can have. Extensive peer review by a large,
260 multi-platform user community has shown that fetchmail is as near
261 bulletproof as the underlying protocols permit.</p>
263 <p>Fetchmail is licensed under the <a
264 href="http://gnu.org//copyleft/gpl.html">GNU General Public
267 <p>If you found this FAQ in the distribution, see the README for
268 fetchmail's full feature list.</p>
271 <h2><a id="G2" name="G2">G2. Where do I find the latest FAQ and
272 fetchmail sources?</a></h2>
274 <p>The latest HTML FAQ is available alongside the latest fetchmail
275 sources at the fetchmail home page: <a
276 href="http://fetchmail.berlios.de/">http://fetchmail.berlios.de/</a>.
277 You can also usually find both in the <a
278 href="http://sunsite.unc.edu/pub/Linux/system/mail/pop/!INDEX.html">
279 POP mail tools directory on Sunsite</a>.</p>
281 <p>A text dump of this FAQ is included in the fetchmail
282 distribution. Because it freezes at distribution release time, it
283 may not be completely current.</p>
286 <h2><a id="G3" name="G3">G3. I think I've found a bug. Will you fix
289 <p>Yes I will, provided you include enough diagnostic information
290 for me to go on. Send bugs to <a
291 href="mailto:fetchmail-users@lists.berlios.de">fetchmail-users</a>.
292 When reporting bugs, please include the following:</p>
295 <li>Your operating system.</li>
297 <li>Your compiler version, if you built from source; otherwise, the
298 name and origin of the RPM or other binary package you
301 <li>A copy of your POP or IMAP server's greeting line.</li>
303 <li>The name and version of the SMTP listener or MDA you are
306 <li>Any command-line options you used.</li>
308 <li>The output of fetchmail -V called with whatever other
309 command-line options you used.</li>
312 <p>If you have FTP access to your remote mail account, and you have
313 any suspicion that the bug was triggered by a particular message,
314 please include a copy of the message that triggered the bug.</p>
316 <p>Often, the first thing I will do when you report a bug is tell
317 you to upgrade to the newest version of fetchmail, and then see if
318 the problem reproduces. So you'll probably save us both time if you
319 upgrade and test with the latest version <em>before</em> sending in
322 <p>If your bug is something that used to work but stopped working
323 when you upgraded, then you can help pin the bug down by trying <a
324 href="http://download.berlios.de/fetchmail/">intermediate versions
325 of fetchmail</a> until you identify the revision that broke your
326 feature. The smart way to do this is by binary search on the
327 version sequence. First, try the version halfway between your last
328 good one and the current one. If it works, the failure was
329 introduced in the upper half of the sequence; if it doesn't, the
330 failure was introduced in the lower half. Now bisect that half in
331 the same way. In a very few tries, you should be able to identify
332 the exact adjacent pair of versions between which your bug was
333 introduced -- and with information like that, I can usually come up
334 with a fix very quickly.</p>
336 <p>Another useful thing you can do, if you're using POP3, is to
337 test for IMAP4 support on your mailserver using the autoprobe
338 function of fetchmailconf. If you have IMAP4, and fetchmailconf
339 doesn't tell you it's broken, switch immediately. POP3 is a weak,
340 poorly-designed protocol with chronic problems, and the later
341 versions after RFC1725 actually get worse rather than better.
342 Changing over to IMAP4 may well make your problem go away -- and if
343 your ISP doesn't have IMAP4 support, bug them to supply it.</p>
345 <p>It is helpful if you include your .fetchmailrc file, but not
346 necessary unless your symptom seems to involve an error in
347 configuration parsing. If you do send in your .fetchmailrc, mask
348 the passwords first!</p>
350 <p>If fetchmail seems to run and fetch mail, but the headers look
351 mangled (that is, headers are missing or blank lines are inserted
352 in the headers) then read the FAQ items in section <a
353 href="#X1">X</a> before submitting a bug report. Pay special
354 attention to the item on <a href="#generic_mangling">diagnosing
355 mail mangling</a>. There are lots of ways for other programs in the
356 mail chain to screw up that look like fetchmail's fault, but you
357 may be able to fix these by tweaking your configuration.</p>
359 <p>A transcript of the failed session with -v -v (yes, that's
360 <em>two</em> -v options, enabling debug mode) will almost always be
361 useful. It is very important that the transcript include your
362 POP/IMAP server's greeting line, so I can identify it in case of
363 server problems. This transcript will not reveal your passwords,
364 which are specially masked out precisely so transcripts can be
367 <p>If you upgraded your fetchmail and something broke, you should
368 include session transcripts with -v -v of both the working and
369 failing versions. Very often, the source of the problem can
370 instantly identified by looking at the differences in protocol
373 <p>If the bug involves a core dump or hang, a gdb stack trace is
374 good to have. (Bear in mind that you can attach gdb to a running
375 but hung process by giving the process ID as a second argument.)
376 You will need to reconfigure with:</p>
379 CFLAGS=-g LDFLAGS=" " ./configure
382 <p>Then rebuild in order to generate a version that can be
385 <p>Best of all is a mail file which, when fetched, will reproduce
386 the bug under the latest (current) version.</p>
388 <p>Any bug I can reproduce will usually get fixed very quickly,
389 often within 48 hours. Bugs I can't reproduce are a crapshoot. If
390 the solution isn't obvious when I first look, it may evade me for a
391 long time (or to put it another way, fetchmail is well enough
392 tested that the easy bugs have long since been found). So if you
393 want your bug fixed rapidly, it is not just sufficient but nearly
394 <em>necessary</em> that you give me a way to reproduce it.</p>
397 <h2><a id="G4" name="G4">G4. I have this idea for a neat feature.
398 Will you add it?</a></h2>
400 <p>Probably not. Most of the feature suggestions I get are for ways
401 to set various kinds of administrative policy or add more spam
402 filtering (the most common one, which I used to get about four
403 million times a week and got <em>really</em> tired of, is for
404 tin-like kill files).</p>
406 <p>You can do spam filtering better with procmail or maildrop on
407 the server side and (if you're the server sysadmin) sendmail.cf
408 domain exclusions. If you really want fetchmail to do it from the
409 client side, use a <code>preconnect</code> command to call
410 <a href='http://mailfilter.sourceforge.net/'>mailfilter</a>.</p>
412 <p>You can do other policy things better with the
413 <code>mda</code> option and script wrappers around fetchmail. If
414 it's a prime-time-vs.-non-prime-time issue, ask yourself whether a
415 wrapper script called from crontab would do the job.</p>
417 <p>I'm not going to do these; fetchmail's job is transport, not
418 policy, and I refuse to change it from doing one thing well to
419 attempting many things badly. One of my objectives is to keep
420 fetchmail simple so it stays reliable.</p>
422 <p>For reasons fetchmail doesn't have other commonly-requested
423 features (such as password encryption, or multiple concurrent polls
424 from the same instance of fetchmail) see <a
425 href="esrs-design-notes.html">ESR's design
426 notes</a>. Note that this document is partially obsoleted by the
427 <a href="design-notes.html">updated design notes.</a></p>
429 <p>Fetchmail is a mature project, no longer in constant active
430 development. It is no longer my top project, and I am going to be
431 quite reluctant to add features that might either jeopardize its
432 stability or involve me in large amounts of coding.</p>
434 <p>All that said, if you have a feature idea that really is about a
435 transport problem that can't be handled anywhere but fetchmail, lay
436 it on me. I'm very accommodating about good ideas.</p>
439 <h2><a id="G5" name="G5">G5. I want to make fetchmail behave like
440 Outlook Express.</a></h2>
442 <p>The second-most-requested feature for fetchmail, after
443 content-based filtering, is the ability to have it remove messages
444 from a maildrop after N days, typically to be used with the
445 <code>keep</code> option as a sort of poor man's newsgroup
446 facility. Microsoft's Outlook Express supports this.</p>
448 <p>This feature won't be added either. Repeat after me: fetchmail's
449 job is transport, not policy. If you want this, write a Perl or
450 Python script, to be run from a cron job, that deletes old messages
451 off your maildrop. Send it to me and I'll put it in the contrib
455 <h2><a id="G6" name="G6">G6. Is there a mailing list for exchanging
458 <p>There is a fetchmail-users list (fetchmail-users@lists.berlios.de)
459 for bug reports and people who want to discuss configuration issues of
460 fetchmail. It's a Mailman list, see <a
461 href="http://lists.berlios.de/mailman/listinfo/fetchmail-users">http://lists.berlios.de/mailman/listinfo/fetchmail-users</a>.</p>
462 <p>There is a fetchmail-devel list
463 (fetchmail-devel@lists.berlios.de) for people who want to discuss
464 fixes and improvements in fetchmail and help co-develop it. It's a
465 Mailman list, which you can sign up for at <a
466 href="http://lists.berlios.de/mailman/listinfo/fetchmail-devel">http://lists.berlios.de/mailman/listinfo/fetchmail-devel</a>.
467 There is also an announcements-only list,
468 fetchmail-announce@lists.berlios.de, which you can sign up for at <a
469 href="http://lists.berlios.de/mailman/listinfo/fetchmail-announce">http://lists.berlios.de/mailman/listinfo/fetchmail-announce</a>.</p>
472 <h2><a id="G7" name="G7">G7. So, what's this I hear about a
473 fetchmail paper?</a></h2>
475 <p>The fetchmail development was also a sociological experiment, an
476 extended test to see if my theory about the critical features of
477 the Linux development model is correct.</p>
479 <p>The experiment was a success. I wrote a paper about it titled <a
480 href="http://www.catb.org/~esr/writings/cathedral.html">The
481 Cathedral and the Bazaar</a> which was first presented at Linux
482 Kongress '97 in Bavaria and very well received there. It was also
483 given at Atlanta Linux Expo, Linux Pro '97 in Warsaw, and the first
484 Perl Conference, at UniForum '98, and was the basis of an invited
485 presentation at Usenix '98. The folks at Netscape tell me it helped
487 href="http://www.netscape.com/newsref/pr/newsrelease558.html">give
488 away the source for Netscape Communicator</a>.</p>
490 <p>If you're reading a non-HTML dump of this FAQ, you can find the
491 paper on the Web with a search for that title.</p>
494 <h2><a id="G8" name="G8">G8. What is the best server to use with
497 <p>The short answer: IMAP 2000 running over Unix.</p>
499 <p>Here's a longer answer:</p>
501 <p>Fetchmail will work with any POP, IMAP, ETRN, or ODMR server
502 that conforms to the relevant RFCs (and even some outright broken
503 ones like <a href="#S2">Microsoft Exchange</a> and <a
504 href="#S6">Novell GroupWise</a>). This doesn't mean it works
505 equally well with all, however. POP2 servers, and POP3 servers
506 without LAST, limit fetchmail's capabilities in various ways
507 described on the manual page.</p>
509 <p>Most modern Unixes (and effectively all Linux/*BSD systems) come
510 with POP3 support preconfigured (but beware of the horribly broken
511 POP3 server mentioned in <a href="#D2">D2</a>). An increasing
512 minority also feature IMAP (you can detect IMAP support by running
513 fetchmail in AUTO mode, or by using the 'Probe for supported
514 protocols' function in the fetchmailconf utility).</p>
516 <p>If you have the option, we recommend using or installing an
517 IMAP4rev1 server; it has the best facilities for tracking message
518 'seen' states. It also recovers from interrupted connections more
519 gracefully than POP3, and enables some significant performance
520 optimizations. The new <a
521 href="ftp://ftp.cac.washington.edu/imap/imap.tar.Z">IMAP 2000</a>
522 is particularly nice, as it supports CRAM-MD5 so you don't have to
523 ship your mail password over the net en clair (fetchmail
524 autodetects this capability). Older versions had support for GSSAPI
525 giving a similar effect.</p>
527 <p>Don't be fooled by NT/Exchange propaganda. M$ Exchange is just
528 plain broken (see item <a href="#S2">S2</a>) and NT cannot handle
529 the sustained load of a high-volume remote mail server. Even
530 Microsoft itself knows better than to try this; their own Hotmail
531 service runs over Solaris! For extended discussion, see John
532 Kirch's excellent <a href="http://unix-vs-nt.org/kirch/">white
533 paper</a> on Unix vs. NT performance.</p>
535 <p>Source for a high-quality supported implementation of POP is
536 available from the <a
537 href="ftp://ftp.qualcomm.com/eudora/servers/unix/popper/">Eudora
538 FTP site</a>. Don't use 2.5, which has a rather restrictive
539 license. The 2.5.2 version appears to restore the open-source
540 license of previous versions.</p>
543 <h2><a id="G9" name="G9">G9. What is the best mail program to use
544 with fetchmail?</a></h2>
546 <p>Fetchmail will work with all popular <a href="#T1">mail
547 transport programs</a>. It also doesn't care which user agent you
548 use, and user agents are as a rule almost equally indifferent to
549 how mail is delivered into your system mailbox. So any of the
550 popular Unix mail agents -- <a
551 href="http://www.myxa.com/old/elm.html">elm</a>, <a
552 href="http://www.washington.edu/pine/">pine</a>, <a
553 href="http://www.cs.indiana.edu/docproject/mail/mh.html">mh</a>, or
554 <a href="http://www.mutt.org">mutt</a> -- will work fine with
557 <p>All this having been said, I can't resist putting in a discreet
558 plug for <a href="http://www.mutt.org">mutt</a>. My own personal
559 mail setup is sendmail plus fetchmail plus mutt. Mutt's interface
560 is only a little different from that of its now-moribund ancestor
561 elm, but its excellent handling of MIME and PGP put it in a class
562 by itself. You won't need its built-in POP3 support, though; most
563 of the mutt developers will cheerfully admit that fetchmail's is
567 <h2><a id="G10" name="G10">G10. How can I avoid sending my password
570 <p>Depending on what your mail server you are talking to, this
571 ranges from trivial to impossible. It may even be next to
574 <p>Most people use fetchmail over phone wires (whether plain old
575 copper or DSL), which are hard to tap. Anybody with the skill and
576 resources to do this could get into your server mailbox with much less
577 effort by subverting the server host. So if your provider setup is
578 phone-company wire going straight into a service box, you probably
579 don't need to worry.</p>
581 <p>In general there is little point in trying to secure your fetchmail
582 transaction unless you trust the security of the server host you are
583 retrieving mail from. Your vulnerability is more likely to be an
584 insecure local network on the server end (e.g. to somebody with a
585 TCP/IP packet sniffer intercepting Ethernet traffic between the modem
586 concentrator or DSL POP you dial in to and the mailserver host).</p>
588 <p>Having realized this, you need to ask whether password
589 encryption alone will really address your security exposure. If you
590 think you might be snooped between server and client, it's better
591 to use end-to-end encryption on your whole mail stream so none of
592 it can be read. One of the advantages of fetchmail over
593 conventional SMTP-push delivery is that you may be able to arrange
594 this by using ssh(1); see <a href="#K3">K3</a>.</p>
596 <p>Note that ssh is not a complete privacy solution either, as your
597 mail could have been snooped in transit to your POP server from
598 wherever it originated. For best security, agree with your
599 correspondents to use a tool such as <a
600 href="http://www.gnupg.org/">GPG</a> (Gnu Privacy Guard) or PGP
601 (Pretty Good Privacy).</p>
603 <p>If ssh/sshd isn't available, or you find it too complicated for
604 you to set up, password encryption will at least keep a malicious
605 cracker from deleting your mail, and require him to either tap your
606 connection continuously or crack root on the server in order to
609 <p>You can deduce what encryptions your mail server has available
610 by looking at the server greeting line (and, for IMAP, the response
611 to a CAPABILITY query). Do a <code>fetchmail -v</code> to see
612 these, or telnet direct to the server port (110 for POP3, 143 for
615 <p>If your mailserver is using IMAP 2000, you'll have CRAM-MD5
616 support built in. Fetchmail autodetects this; you can skip the rest
619 <p>The POP3 facility you are most likely to have available is APOP.
620 This is a POP3 feature supported by many servers (fetchmailconf's
621 autoprobe facility will detect it and tell you if you have it). If
622 you see something in the greeting line that looks like an
623 angle-bracket-enclosed Internet address with a numeric left-hand
624 part, that's an APOP challenge (it will vary each time you log in).
625 You can register a secret on the host (using
626 <code>popauth(8)</code> or some program like it). Specify the
627 secret as your password in your .fetchmailrc; it will be used to
628 encrypt the current challenge, and the encrypted form will be sent
629 back the the server for verification.</p>
631 <p>Alternatively, you may have Kerberos available. This may require
632 you to set up some magic files in your home directory on your
633 client machine, but means you can omit specifying any password at
636 <p>Fetchmail supports two different Kerberos schemes. One is a POP3
637 variant called KPOP; consult the documentation of your mail server
638 to see if you have it (one clue is the string "krb-IV" in the
639 greeting line on port 110). The other is an IMAP and POP3 facility
640 described by RFC1731 and RFC1734. You can tell if this one is
641 present by looking for AUTH=KERBEROS_V4 in the CAPABILITY
644 <p>If you are fetching mail from a CompuServe POP3 account, you can
645 use their RPA authentication (which works much like APOP). See <a
646 href="#I1">I1</a> for details. If you are fetching mail from
647 Microsoft Exchange using IMAP, you will be able to use NTLM.</p>
649 <p>Your POP3 server may have the RFC1938 OTP capability to use
650 one-time passwords (if it doesn't, you can get OTP patches for the
651 2.2 version of the Qualcomm popper from <a href="#cmetz">Craig
652 Metz</a>). To check this, look for the string "otp-" in the
653 greeting line. If you see it, and your fetchmail was built with
654 OPIE support compiled in (see the distribution INSTALL file),
655 fetchmail will detect it also. When using OTP, you will specify a
656 password but it will not be sent en clair.</p>
658 <p>You can get both POP3 and IMAP OTP patches from <a id="cmetz"
659 name="cmetz">Craig Metz</a> at <a
660 href="http://www.inner.net/opie">http://www.inner.net/opie</a>.</p>
662 <p>These patches use a SASL authentication method named "X-OTP"
663 because there is not currently a standard way to do this; fetchmail
664 also uses this method, so the two will interoperate happily. They
665 better, because this is how Craig gets his mail ;-)</p>
667 <p>Finally, you can use <a href="#K5">SSL</a> for complete
668 end-to-end encryption if you have an SSL-enabled mailserver.</p>
671 <h2><a id="G11" name="G11">G11. Is any special configuration needed
672 to use a dynamic IP address?</a></h2>
674 <p>Yes. In order to avoid giving indigestion to certain picky MTAs
675 (notably <a href="#T3">exim</a>), fetchmail always makes the RCPT
676 TO address it feeds the MTA a fully qualified one with a hostname
677 part. Normally it does this by appending @ and "localhost", but
678 when you are using Kerberos or ETRN mode it will append @ and your
679 machine's fully-qualified domain name (FQDN).</p>
681 <p>Appending the FQDN can create problems when fetchmail is running
682 in daemon mode and outlasts the dynamic IP address assignment your
683 client machine had when it started up.</p>
685 <p>Since the new IP address (looked up at RCPT TO interpretation
686 time) doesn't match the original, the most benign possible result
687 is that your MTA thinks it's seeing a relaying attempt and refuses.
688 More frequently, fetchmail will try to connect to a nonexistent
689 host address and time out. Worst case, you could up forwarding your
690 mail to the wrong machine!</p>
692 <p>Use the <code>smtpaddress</code> option to force the appended
693 hostname to one with a (fixed) IP address of 127.0.0.1 in your
694 <code>/etc/hosts</code>. (The name 'localhost' will usually work;
695 or you can use the IP address itself).</p>
697 <p>Only one fetchmail option interacts directly with your IP
698 address, '<code>interface</code>'. This option can be used to set
699 the gateway device and restrict the IP address range fetchmail will
700 use. Such a restriction is sometimes useful for security reasons,
701 especially on multihomed sites. See <a href="#C3">C3</a>.</p>
703 <p>I recommend against trying to set up the <code>interface</code>
704 option when initially developing your poll configuration -- it's
705 never necessary to do this just to get a link working. Get the link
706 working first, observe the actual address range you see on
707 connections, and add an <code>interface</code> option (if you need
710 <p>You can't use ETRN if you have a dynamic IP address (your ISP
711 changes your IP address occasionally, possibly with every connect).
712 You need to have your own registered domain and a definite IP
713 address registered for that domain. The server needs to be
714 configured to accept mail for your domain but then queue it to
715 forward to your machine. ETRN just tells to server to flush its
716 queue for your domain. Fetchmail doesn't actually get the mail in
719 <p>You can use On-Demand Mail Relay (ODMR) with a dynamic IP
720 address; that's what it was designed for, and it provides
721 capabilities very similar to ETRN. Unfortunately ODMR servers are
722 not yet widely deployed, as of early 2001.</p>
724 <p>If you're using a dynamic-IP configuration, one other
725 (non-fetchmail) problem you may run into with outgoing mail is that
726 some sites will bounce your email because the hostname your giving
727 them isn't real (and doesn't match what they get doing a reverse
728 DNS on your dynamically-assigned IP address). If this happens, you
729 need to hack your sendmail so it masquerades as your host.
736 <p>in your <code>sendmail.cf</code> will work, or you can set</p>
739 MASQUERADE_AS(smarthost.here)
742 <p>in the m4 configuration and do a reconfigure. (In both cases,
743 replace <code>smarthost.here</code> with the actual name of your
744 mailhost.) See the <a
745 href="http://www.lege.com/sendmail-FAQ.txt">sendmail FAQ</a> for
749 <h2><a id="G12" name="G12">G12. Is any special configuration needed
750 to use firewalls?</a></h2>
752 <p>No. You can use fetchmail with SOCKS, the standard tool for
753 indirecting TCP/IP through a firewall. You can find out about
754 SOCKS, and download the SOCKS software including server and client
755 code, at the <a href="http://www.socks.nec.com/">SOCKS distribution
758 <p>The specific recipe for using fetchmail with a firewall is at <a
759 href="#K1">K1</a></p>
762 <h2><a id="G13" name="G13">G13. Is any special configuration needed
763 to <em>send</em> mail?</a></h2>
765 <p>A user asks: but how do we send mail out to the POP3 server? Do
766 I need to implement another tool or will fetchmail do this too?</p>
768 <p>Fetchmail only handles the receiving side. The sendmail or other
769 preinstalled MTA on your client machine will handle sending mail
770 automatically; it will ship mail that is submitted while the
771 connection is active, and put mail that is submitted while the
772 connection is inactive into the outgoing queue.</p>
774 <p>Normally, sendmail is also run periodically (every 15 minutes on
775 most Linux systems) in a mode that tries to ship all the mail in
776 the outgoing queue. If you have set up something like pppd to
777 automatically dial out when your kernel is called to open a TCP/IP
778 connection, this will ensure that the mail gets out.</p>
781 <h2><a id="G14" name="G14">G14. Is fetchmail
782 Y2K-compliant?</a></h2>
784 <p>Fetchmail is fully Y2K-compliant.</p>
786 <p>Fetchmail could theoretically have problems when the 32-bit
787 time_t counters roll over in 2038, but I doubt it. Timestamps
788 aren't used for anything but log entry generation. Anyway, if you
789 aren't running on a 64-bit machine by then, you'll deserve to
793 <h2><a id="G15" name="G15">G15. Is there a way in fetchmail to
794 support disconnected IMAP mode?</a></h2>
796 <p>No. Fetchmail is a mail transport agent, best understood as a
797 protocol gateway between POP3/IMAP servers and SMTP. Disconnected
798 operation requires an elaborate interactive client. It's a very
799 different problem.</p>
802 <h2><a id="G16" name="G16">G16. How will fetchmail perform under
803 heavy loads?</a></h2>
805 <p>Fetchmail streams message bodies line-by-line; the most core it
806 ever requires per message is enough memory to hold the RFC822
807 header, and that storage is freed when body processing begins. It
808 is, accordingly, quite economical in its use of memory.</p>
810 <p>After startup time, a fetchmail running in daemon mode stats its
811 configuration file once per poll cycle to see whether it has
812 changed and should be rescanned. Other than that, a fetchmail in
813 normal operation doesn't touch the disk at all; that job is left up
814 to the MTA or MDA the fetchmail talks to.</p>
816 <p>Fetchmail's performance is usually bottlenecked by latency on
817 the POP server or (less often) on the TCP/IP link to the server.
818 This is not a problem readily solved by tuning fetchmail, or even
819 by buying more TCP/IP capacity (which tends to improve bandwidth
820 but not necessarily latency).</p>
823 <h2><a id="B1" name="B1">B1. Make coughs and dies when building on
826 <p>The vendor-supplied make on FreeBSD systems can only be used
827 within FreeBSD's "scope", e.g. the ports collection. Type "gmake"
828 to run GNU make and better things will happen.</p>
831 <h2><a id="B2" name="B2">B2. Lex bombs out while building the
832 fetchmail lexer.</a></h2>
834 <p>In the immortal words of Alan Cox the last time this came up:
835 "Take the Solaris lex and stick it up the backside of a passing
836 Sun salesman, then install <a
837 href="ftp://ftp.gnu.org/pub/non-gnu/flex/">flex</a> and use that.
838 All will be happier."</p>
840 <p>I couldn't have put it better myself, and ain't going to try
843 <p>(The same problem has been reported under HP-UX v10.20 and
847 <h2><a id="B3" name="B3">B3. I get link failures when I try to
848 build fetchmail.</a></h2>
850 <p>If you get errors resembling these</p>
853 mxget.o(.text+0x35): undefined referenceto '__res_search'
854 mxget.o(.text+0x99): undefined reference to '__dn_skipname'
855 mxget.o(.text+0x11c): undefined reference to '__dn_expand'
856 mxget.o(.text+0x187): undefined reference to '__dn_expand'
857 make: *** [fetchmail] Error 1
860 <p>then you must add "-lresolv" to the LOADLIBS line in your
861 Makefile once you have installed the 'bind' package.</p>
863 <p>If you get link errors involving <tt>dcgettext</tt>, like
867 rcfile_y.o: In function 'yyparse':
868 rcfile_y.o(.text+0x3aa): undefined reference to 'dcgettext__'
869 rcfile_y.o(.text+0x4f2): undefined reference to 'dcgettext__'
870 rcfile_y.o(.text+0x5ee): undefined reference to 'dcgettext__'
871 rcfile_y.o: In function 'yyerror':
872 rcfile_y.o(.text+0xc7c): undefined reference to 'dcgettext__'
873 rcfile_y.o(.text+0xcc8): undefined reference to 'dcgettext__'
874 rcfile_y.o(.text+0xdf9): more undefined references to 'dcgettext__' follow
877 <p>reconfigure with <tt>configure --with-included-gettext</tt>.
878 This is due to some brain-damage in the GNU internationalization
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>
887 <p>GNU gettext is an overengineered, fragile pile of crap. I have
888 teetered on the brink of removing support for it entirely several
892 <h2><a id="F1" name="F1">F1. Why does my old .fetchmailrc file no
893 longer work?</a></h2>
895 <h3>If your file predates 5.8.9</h3>
897 <p>If you were using ETRN mode, change your <tt>smtphost</tt>
898 option to a <tt>fetchdomains</tt> option.</p>
900 <h3>If your file predates 5.8.3</h3>
902 <p>The 'via localhost' special case for use with ssh tunnelling is
903 gone. Use the %h feature of <tt>plugin</tt> instead.</p>
905 <h3>If your file predates 5.6.8</h3>
907 <p>In 5.6.8, the <tt>preauth</tt> keyword and option were changed
908 back to <tt>auth</tt>. The <tt>preauth</tt> synonym will still be
909 supported through a few more point releases.</p>
911 <h3>If your file predates 5.6.5</h3>
913 <p>The <tt>imap-gss</tt>, <tt>imap-k4</tt>, and <tt>imap-login</tt>
914 protocol types are gone. This is a result of a major re-factoring
915 of the authentication machinery; fetchmail can now use Kerberos V4
916 and GSSAPI not just with IMAP but with POP3 servers that have
917 RFC1734 support for the AUTH command.</p>
919 <p>When trying to identify you to an IMAP or POP mailserver,
920 fetchmail now first tries methods that don't require a password
921 (GSSAPI, KERBEROS_IV); then it looks for methods that mask your
922 password (CRAM-MD5, X-OTP); and only if it the server doesn't
923 support any of those will it ship your password en clair.</p>
925 <p>Setting the <tt>preauth</tt> option to any value other than
926 'password' will prevent from looking for a password in your
927 <tt>.netrc</tt> file or querying for it at startup time.</p>
929 <h3>If your file predates 5.1.0</h3>
931 <p>In 5.1.0, the <tt>auth</tt> keyword and option were changed to
932 <tt>preauth</tt>.</p>
934 <h3>If your file predates 4.5.5</h3>
936 <p>If the <code>dns</code> option is on (the default), you may need
937 to make sure that any hostname you specify (for mail hosts or for
938 an SMTP target) is a canonical fully-qualified hostname). In order
939 to avoid DNS overhead and complications, fetchmail no longer tries
940 to derive the fetchmail client machine's canonical DNS name at
943 <h3>If your file predates 4.0.6:</h3>
945 <p>Just after the '<code>via</code>' option was introduced, I
946 realized that the interactions between the '<code>via</code>',
947 '<code>aka</code>', and '<code>localdomains</code>' options were
948 out of control. Their behavior had become complex and confusing, so
949 much so that I was no longer sure I understood it myself. Users
950 were being unpleasantly surprised.</p>
952 <p>Rather than add more options or crock the code, I re-thought it.
953 The redesign simplified the code and made the options more
954 orthogonal, but may have broken some complex multidrop
957 <p>Any multidrop configurations that depended on the name just
958 after the '<code>poll</code>' or '<code>skip</code>' keyword being
959 still interpreted as a DNS name for address-matching purposes, even
960 in the presence of a '<code>via</code>' option, will break.</p>
962 <p>It is theoretically possible that other unusual configurations
963 (such as those using a non-FQDN poll name to generate Kerberos IV
964 tickets) might also break; the old behavior was sufficiently murky
965 that we can't be sure. If you think this has happened to you,
966 contact the maintainer.</p>
968 <h3>If your file predates 3.9.5:</h3>
970 <p>The '<code>remote</code>' keyword has been changed to
971 '<code>folder</code>'. If you try to use the old keyword, the
972 parser will utter a warning.</p>
974 <h3>If your file predates 3.9:</h3>
976 <p>It could be because you're using a .fetchmailrc that's written
977 in the old popclient syntax without an explicit
978 '<code>username</code>' keyword leading the first user entry
979 attached to a server entry.</p>
981 <p>This error can be triggered by having a user option such as
982 '<code>keep</code>' or '<code>fetchall</code>' before the first
983 explicit username. For example, if you write</p>
986 poll openmail protocol pop3
987 keep user "Hal DeVore" there is hdevore here
990 <p>the '<code>keep</code>' option will generate an entire user
991 entry with the default username (the name of fetchmail's invoking
994 <p>The popclient compatibility syntax was removed in 4.0. It
995 complicated the configuration file grammar and confused users.</p>
997 <h3>If your file predates 2.8:</h3>
999 <p>The '<code>interface</code>', '<code>monitor</code>' and
1000 '<code>batchlimit</code>' options changed after 2.8.</p>
1002 <p>They used to be global options with '<code>set</code>' syntax
1003 like the batchlimit and logfile options. Now they're per-server
1004 options, like '<code>protocol</code>'.</p>
1006 <p>If you had something like</p>
1009 set interface = "sl0/10.0.2.15"
1012 <p>in your .fetchmailrc file, simply delete that line and insert
1013 'interface sl0/10.0.2.15' in the server options part of your
1014 'defaults' declaration.</p>
1016 <p>Do similarly for any '<code>monitor</code>' or
1017 '<code>batchlimit</code>' options.</p>
1020 <h2><a id="F2" name="F2">F2. The .fetchmailrc parser won't accept
1021 my all-numeric user name.</a></h2>
1023 <p>Either upgrade to a post-5.0.5 fetchmail or put string quotes
1026 <p>The configuration file parser in older fetchmail versions
1027 treated any all-numeric token as a number, which confused it when
1028 it was expecting a name. String quoting forces the token's
1031 <p>The lexical analyzer in 5.0.6 and beyond is smarter and assumes
1032 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>
1047 <h2><a id="F4" name="F4">F4. I'm getting a 'parse error' message I
1048 don't understand.</a></h2>
1050 <p>The most common cause of mysterious parse errors is putting a
1051 server option after a user option. Check the manual page; you'll
1052 probably find that by moving one or more options closer to the
1053 'poll' keyword you can eliminate the problem.</p>
1055 <p>Yes, I know these ordering restrictions are hard to understand.
1056 Unfortunately, they're necessary in order to allow the 'defaults'
1057 feature to work.</p>
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>
1118 <h2><a id="C2" name="C2">C2. How can I arrange for a fetchmail
1119 daemon to get killed when I log out?</a></h2>
1121 <p>The easiest way to dispatch fetchmail on logout (which will work
1122 reliably only if you have just one login going at any time) is to
1123 arrange for the command 'fetchmail -q' to be called on logout.
1124 Under bash, you can arrange this by putting 'fetchmail -q' in the
1125 file '~/.bash_logout'. Most csh variants execute '~/.logout' on
1126 logout. For other shells, consult your shell manual page.</p>
1128 <p>Automatic startup/shutdown of fetchmail is a little harder to
1129 arrange if you may have multiple login sessions going. In the
1130 contrib subdirectory of the fetchmail distribution there is some
1131 shell code you can add to your .bash_login and .bash_logout
1132 profiles that will accomplish this. Thank James Laferriere
1133 <babydr@nwrain.net> for it.</p>
1135 <p>Some people start up and shut down fetchmail using the ppp-up
1136 and ppp-down scripts of pppd.</p>
1139 <h2><a id="C3" name="C3">C3. How do I know what interface and
1140 address to use with --interface?</a></h2>
1142 <p>This depends a lot on your local networking configuration (and
1143 right now you can't use it at all except under Linux and the newer
1144 BSDs). However, here are some important rules of thumb that can
1145 help. If they don't work, ask your local sysop or your Internet
1148 <p>First, you may not need to use --interface at all. If your
1149 machine only ever does SLIP or PPP to one provider, it's almost
1150 certainly by a point to point modem connection to your provider's
1151 local subnet that's pretty secure against snooping (unless someone
1152 can tap your phone or the provider's local subnet!). Under these
1153 circumstances, specifying an interface address is fairly
1156 <p>What the option is really for is sites that use more than one
1157 provider. Under these circumstances, typically one of your provider
1158 IP addresses is your mailserver (reachable fairly securely via the
1159 modem and provider's subnet) but the others might ship your packets
1160 (including your password) over unknown portions of the general
1161 Internet that could be vulnerable to snooping. What you'll use
1162 --interface for is to make sure your password only goes over the
1163 one secure link.</p>
1165 <p>To determine the device:</p>
1168 <li>If you're using a SLIP link, the correct device is probably
1171 <li>If you're using a PPP link, the correct device is probably
1174 <li>If you're using a direct connection over a local network such
1175 as an ethernet, use the command 'netstat -r' to look at your
1176 routing table. Try to match your mailserver name to a destination
1177 entry; if you don't see it in the first column, use the 'default'
1178 entry. The device name will be in the rightmost column.</li>
1181 <p>To determine the address and netmask:</p>
1184 <li>If you're talking to slirp, the correct address is probably
1185 10.0.2.15, with no netmask specified. (It's possible to configure
1186 slirp to present other addresses, but that's the default.)</li>
1188 <li>If you have a static IP address, run 'ifconfig <device>',
1189 where <device> is whichever one you've determined. Use the IP
1190 address given after "inet addr:". That is the IP address for your
1191 end of the link, and is what you need. You won't need to specify a
1194 <li>If you have a dynamic IP address, your connection IP will vary
1195 randomly over some given range (that is, some number of the least
1196 significant bits change from connection to connection). You need to
1197 declare an address with the variable bits zero and a complementary
1198 netmask that sets the range.</li>
1201 <p>To illustrate the rule for dynamic IP addresses, let's suppose
1202 you're hooked up via SLIP and your IP provider tells you that the
1203 dynamic address pool is 255 addresses ranging from 205.164.136.1 to
1204 205.164.136.255. Then</p>
1207 interface "sl0/205.164.136.0/255.255.255.0"
1210 <p>would work. To range over any value of the last two octets
1211 (65536 addresses) you would use</p>
1214 interface "sl0/205.164.0.0/255.255.0.0"
1218 <h2><a id="C4" name="C4">C4. How can I set up support for
1219 sendmail's anti-spam features?</a></h2>
1221 <p>This answer covers versions of sendmail from 8.9.3-20 (the
1222 version installed in Red Hat 6.2) upwards. If you have an older
1223 version, upgrade to sendmail 8.9.</p>
1225 <p>Stock sendmails can now do anti-spam exclusions based on a
1226 database of filter rules. The human-readable form of the database
1227 is at <tt>/etc/mail/access</tt>. The database itself is at
1228 <tt>/etc/mail/access.db</tt>.</p>
1230 <p>The table itself uses email addresses, domain names, and network
1231 numbers as keys. For example,</p>
1234 spammer@aol.com REJECT
1235 cyberspammer.com REJECT
1239 <p>would refuse mail from spammer@aol.com, any user from
1240 cyberspammer.com (or any host within the cyberspammer.com domain),
1241 and any host on the 192.168.212.* network. (This feature can be
1242 used to do other things as well; see the <a
1243 href="http://www.sendmail.org/m4/anti-spam.html">sendmail
1244 documentation</a> for details)</p>
1246 <p>To actually set up the database, run</p>
1249 makemap hash deny <deny
1252 <p>in /etc/mail.</p>
1254 <p>To test, send a message to your mailing address from that host
1255 and then pop off the message with fetchmail, using the -v argument.
1256 You can monitor the SMTP transaction, and when the FROM address is
1257 parsed, if sendmail sees that it is an address in spamlist,
1258 fetchmail will flush and delete it.</p>
1260 <p>Under no circumstances put your <strong>mailhost</strong> or
1261 <strong>any host you accept mail from</strong> using fetchmail into
1262 your reject file. You <strong>will</strong> lose mail if you do
1266 <h2><a id="C5" name="C5">C5. How can I poll some of my mailboxes
1267 more/less often than others?</a></h2>
1269 <p>Use the <cite>interval</cite> keyword on the ones that should be
1270 checked less often. For example, if you do a poll every 5 minutes,
1271 and want to poll some mailboxes every 5 minutes and some every 30
1272 minutes, use something like this:</p>
1275 poll mainsite.example.com proto pop3 user ....
1276 poll secondary.example.com proto pop3 interval 6 user ...
1279 <p>Then secondary.example.com will be polled every 6th time that
1280 mainsite.example.com is polled, which with a polling interval of
1281 every 5 minutes means that secondary.example.com will be polled
1282 every 30 minutes.</p>
1285 <h2><a id="C6" name="C6">Fetchmail works OK started up manually,
1286 but not from an init script.</a></h2>
1288 <p>Often, startup scripts have a different environment than an
1289 interactive login shell. For instance, $HOME might point to "/root"
1290 when you are logged in as root, but it might be either unset, or
1291 set to "/" when the startup scripts are running. That means
1292 fetchmail at startup can't find the .fetchmailrc.</p>
1294 <p>Pick a location (such as /etc/fetchmailrc) and use fetchmail's
1295 -f option to point fetchmail at it. That should solve the
1299 <h2><a id="C7" name="C7">C7. How can I forward mail to another
1302 <p>To forward mail to a host other than the one you are running
1303 fetchmail on, use the <code>smtphost</code> or
1304 <code>smtpname</code> option. See the manual page for details.</p>
1307 <h2><a id="T1" name="T1">T1. How can I use fetchmail with
1310 <p>For most sendmails, no special configuration is required. Eric
1311 Allman tells me that if <code>FEATURE(always_add_domain)</code> is
1312 included in sendmail's configuration, you can leave the
1313 <code>rewrite</code> option off.</p>
1315 <p>If your sendmail complains "sendmail does not relay", make
1316 sure your sendmail.cf file says <code>Cwlocalhost</code> so that
1317 sendmail recognizes 'localhost' as a name of its host.</p>
1319 <p>If you're mailing from another machine on your local network,
1320 also ensure that its IP address is listed in ip_allow or name in
1321 name_allow (usually in /etc/mail/)</p>
1323 <p>If you find that your sendmail doesn't like the address
1324 'FETCHMAIL-DAEMON@localhost' (which is used in the bouncemail that
1325 fetchmail generates), you may have to set
1326 <code>FEATURE(accept_unqualified_senders)</code>.</p>
1328 <p>Günther Leber reports that Digital Unix sendmails won't
1329 work with fetchmail. The symptom is an error message "<code>553
1330 Local configuration error, hostname not recognized as
1331 local</code>". The problem is that fetchmail normally feeds
1332 sendmail with the client machine's host address in the MAIL FROM
1333 line. These sendmails think this means they're seeing the result of
1334 a mail loop and suppress the mail. You may be able to work around
1335 this by running in <code>--invisible</code> mode.</p>
1337 <p>If you want to support multidrop mode, and you can get access to
1338 your mailserver's sendmail.cf file, it's a good idea to add this
1342 H?l?Delivered-To: $h
1345 <p>This will cause the mailserver's sendmail to reliably write the
1346 appropriate envelope address into each message before fetchmail
1347 sees it, and tell fetchmail which header it is.  With this
1348 change, multidrop mode should work reliably even when the Received
1349 header omits the envelope address (which will typically be the case
1350 when the message has multiple recipients).  However it will
1351 still not distinguish the recipients, your only advantage is that
1352 no bounce will be sent if a message is BCC addressed to multiple
1353 users at your site.  To fix even that problem, you might want
1354 to try the following hack, which is however untested and quite
1358 H?J?Delivered-To: $u
1360 Mmdrop, P=/usr/bin/procmail, F=lsDFMqSPfhnu9J,
1361 S=EnvFromSMTP/HdrFromSMTP, R=EnvToSMTP/HdrToSMTP,
1362 T=DNS/RFC822/X-Unix,
1363 A=procmail -Y -a $u -d $h
1366 <p>For both hacks, you have to declare '<code>envelope
1367 "Delivered-To:"</code>' on the fetchmail side, to put the virtual
1368 domain (e.g. 'domain.com') with RELAY permission into your access
1369 file and to add a line reading '<code>domain.com
1370 local:local-pop-user</code>' for the first and '<code>domain.com
1371 mdrop:local-pop-user</code>' for the second hack to your
1374 <p>You will notice that if the mail already has a Delivered-To
1375 header, sendmail will not add another.  Further, editing
1376 sendmail.cf directly is not very comfortable.  Solutions for
1377 both problems can be found in Peter 'Rattacresh' Backes' 'hybrid'
1378 patch against sendmail.  Have a look at it, you can find it in
1379 the contrib subdirectory.</p>
1381 <p>Feel free to try Martijn Lievaart's detailed recipe in the
1382 contrib subdirectory of the fetchmail source distribution, it
1383 attempts to realize multidrop mailboxes with an external
1386 <p>If for some reason you are invoking sendmail via the
1387 <tt>mda</tt> option (rather than delivering to port 25 via smtp),
1388 don't forget to include the -i switch. Otherwise you will
1389 occasionally get mysterious delivery failures with a SIGPIPE as the
1390 sendmail instance dies. The problem is messages with a single dot
1391 at start of a text line.</p>
1394 <h2><a id="T2" name="T2">T2. How can I use fetchmail with
1397 <p>Turn on the <code>forcecr</code> option; qmail's listener mode
1398 doesn't like header or message lines terminated with bare
1401 <p>(This information is thanks to Robert de Bath
1402 <robert@mayday.cix.co.uk>.)</p>
1404 <p>If a mailhost is using the qmail package (see <a
1405 href="http://pobox.com/~djb/qmail.html">http://pobox.com/~djb/qmail.html</a>)
1406 then, providing the local hosts are also using qmail, it is
1407 possible to set up one fetchmail link to be reliably collect the
1408 mail for an entire domain.</p>
1410 <p>One of the basic features of qmail is the 'Delivered-To:'
1411 message header. Whenever qmail delivers a message to a local
1412 mailbox it puts the username and hostname of the envelope recipient
1413 on this line. The major reason for this is to prevent mail
1416 <p>To set up qmail to batch mail for a disconnected site the
1417 ISP-mailhost will have normally put that site in its 'virtualhosts'
1418 control file so it will add a prefix to all mail addresses for this
1419 site. This results in mail sent to
1420 'username@userhost.userdom.dom.com' having a 'Delivered-To:' line
1424 Delivered-To: mbox-userstr-username@userhost.userdom.dom.com
1427 <p>A single host maildrop will be slightly simpler:</p>
1430 Delivered-To: mbox-userstr-username@userhost.dom.com
1433 <p>The ISP can make the 'mbox-userstr-' prefix anything they choose
1434 but a string matching the user host name is likely.</p>
1436 <p>To use this line you must:</p>
1439 <li>Ensure the option 'envelope Delivered-To:' is in the fetchmail
1442 <li>Ensure you have a localdomains containing 'userdom.dom.com' or
1443 'userhost.dom.com' respectively.</li>
1446 <p>So far this reliably delivers messages to the correct machine of
1447 the local network, to deliver to the correct user the
1448 'mbox-userstr-' prefix must be stripped off of the user name. This
1449 can be done by setting up an alias within the qmail MTA on each
1450 local machine. Simply create a dot-qmail file called
1451 '.qmail-mbox-userstr-default' in the alias directory (normally
1452 /var/qmail/alias) with the contents:</p>
1455 | ../bin/qmail-inject -a -f"$SENDER" "${LOCAL#mbox-userstr-}@$HOST"
1458 <p>Note this <em>does</em> require a modern /bin/sh.</p>
1460 <p>Peter Wilson adds:</p>
1462 <p>"My ISP uses "alias-unzzippedcom-" as the prefix, which means
1463 that I need to name my file ".qmail-unzzippedcom-default". This is
1464 due to qmail's assumption that a message sent to user-xyz is
1465 handled by the file ~user/.qmail-xyz (or
1466 ~user/.qmail-default)."</p>
1468 <p>Luca Olivetti adds:</p>
1470 <p>If you aren't using qmail locally, or you don't want to set up
1471 the alias mechanism described above, you can use the option
1472 '<code>qvirtual "mbox-userstr-"</code>' in your fetchmail config
1473 file to strip the prefix from the local user name.</p>
1476 <h2><a id="T3" name="T3">T3. How can I use fetchmail with
1479 <p>If you have <code>rewrite</code> on:</p>
1481 <p>There is an RFC1123 requirement that MAIL FROM and RCPT TO
1482 addresses you pass to it have to be canonical (e.g. with a fully
1483 qualified hostname part). Therefore fetchmail tries to pass fully
1484 qualified RCPT TO addresses. But exim does not by default accept
1485 'localhost' as a fully qualified domain. This can be fixed.</p>
1487 <p>In exim.conf, add 'localhost' to your local_domains declaration
1488 if it's not already present. For example, the author's site at
1489 thyrsus.com would have a line reading:</p>
1492 local_domains = thyrsus.com:localhost
1495 <p>If you have <code>rewrite</code> off:</p>
1497 <p>MAIL FROM is a potential problem if the MTAs upstream from your
1498 fetchmail don't necessarily pass canonicalized From and Return-Path
1499 addresses, and fetchmail's <code>rewrite</code> option is off. The
1500 specific case where this has come up involves bounce messages
1501 generated by sendmail on your mailer host, which have the
1502 (un-canonicalized) origin address MAILER-DAEMON.</p>
1504 <p>The right way to fix this is to enable the <code>rewrite</code>
1505 option and have fetchmail canonicalize From and Return-Path
1506 addresses with the mailserver hostname before exim sees them. This
1507 option is enabled by default, so it won't be off unless you turned
1510 <p>If you must run with <code>rewrite</code> off, there is a switch
1511 in exim's configuration files that allows it to accept domainless
1512 MAIL FROM addresses; you will have to flip it by putting the
1516 sender_unqualified_hosts = localhost
1519 <p>in the main section of the exim configuration file. Note that
1520 this will result in such messages having an incorrect domain name
1521 attached to their return address (your SMTP listener's hostname
1522 rather than that of the remote mail server).</p>
1525 <h2><a id="T4" name="T4">T4. How can I use fetchmail with
1528 <p>Smail 3.2 is very nearly plug-compatible with sendmail, and may
1529 work fine out of the box.</p>
1531 <p>We have one report that when processing multiple messages from a
1532 single fetchmail session, smail sometimes delivers them in an order
1533 other than received-date order. This can be annoying because it
1534 scrambles conversational threads. This is not fetchmail's problem,
1535 it is an smail 'feature' and has been reported to the maintainers
1538 <p>Very recent smail versions require an
1539 <code>-smtp_hello_verify</code> option in the smail config file.
1540 This overrides smail's check to see that the HELO address is
1541 actually that of the client machine, which is never going to be the
1542 case when fetchmail is in the picture. According to RFC1123 an SMTP
1543 listener <em>must</em> allow this mismatch, so smail's new behavior
1544 (introduced sometime between 3.2.0.90 and 3.2.0.95) is a bug.</p>
1546 <p>You may also need to say
1547 <code>-smtp_hello_broken_allow=127.0.0.1</code> in order for smail
1548 to accept the "localhost" that fetchmail normally appends to
1549 recipient addresses.</p>
1552 <h2><a id="T5" name="T5">T5. How can I use fetchmail with SCO's
1555 <p>MMDF itself is difficult to configure, but it turns out that
1556 connecting fetchmail to MMDF's SMTP channel isn't that hard. You
1558 href="http://www.aplawrence.com/Unixart/uucptofetch.html">MMDF
1559 recipe</a> that describes replacing a UUCP link with fetchmail
1563 <h2><a id="T6" name="T6">T6. How can I use fetchmail with Lotus
1566 <p>The Lotus Notes SMTP gateway tries to deduce when it should
1567 convert \n to \r\n, but its rules are not the intuitive and
1568 correct-for-RFC822 ones. Use 'forcecr'.</p>
1571 <h2><a id="T7" name="T7">T7. How can I use fetchmail with Courier
1574 <p>The courier mta doesn't like RCPT addresses that look like
1575 <code>someone@localhost</code>. Work around this with an
1576 <code>smtphost</code> or <code>smtpaddress</code>.</p>
1579 <h2><a name="T8">T8. How can I use fetchmail with vbmailshield?</a></h2>
1581 <p>vbmailshield's SMTP interpreter is broken. It doesn't understand RSET.</p>
1583 <p>As a workaround, you can set batchlimit to 1 so RSET is never used.</p>
1586 <h2><a id="S1" name="S1">S1. How can I use fetchmail with
1589 <p>Qualcomm's qpopper is probably the best-of-breed among POP3
1590 servers, and is very widely deployed. Nevertheless, it has some
1591 problems which fetchmail exposes. We recommend using <a
1592 href="#G8">IMAP</a> instead if at all possible. If you must talk to
1593 qpopper, here are some problems to be aware of:</p>
1595 <h3>Problems with retrieving large messages from qpopper 2.53</h3>
1598 href="mailto:tony@atn.com.hk"><tony@atn.com.hk></a> reports
1599 that there is a bad intercation between fetchmail and qpopper 2.5.3
1600 under Red Hat Linux versions 5.0 to 5.2, kernels 2.0.34 to 2.0.35.
1601 When fetching very large messages (over 700K) from 2.5.3, fetchmail
1602 will hang with a socket error.</p>
1604 <p>This is probably not a fetchmail bug, but rather a symptom of
1605 some problem in the networking stack that qpopper's transmission
1606 pattern is tickling, as fetchpop (another Linux POP client) also
1607 displays the hang but Netscape running under Win95 does not. The
1608 problem can also be banished by <a
1609 href="http://www.eudora.com/freeware/qpop.html">upgrading to
1610 qpopper 3.0b1</a>.</p>
1612 <h3>Bad interaction with fetchmail 4.4.2 to 4.4.7</h3>
1614 <p>Versions of fetchmail from 4.4.2 through 4.4.7 had a bad
1615 interaction with Eudora qpopper versions 2.3 and later. See <a
1616 href="#X5">X5</a> for details. The solution is to upgrade your
1620 <h2><a id="S2" name="S2">S2. How can I use fetchmail with Microsoft
1623 <p>It's been reliably reported that Exchange 2000's POP3 support is
1624 so broken that it's unusable. One symptom is that messages without
1625 a terminating newline get the POP3 message termination dot emitted
1626 -- you guessed it -- right after the last character of the message,
1627 with no terminating newline added. This will hang fetchmail or any
1628 other RFC-compliant server. IMAP is alleged to work OK, though.</p>
1630 <p>Older versions of Exchange are semi-usable. They randomly drop
1631 attachments on the floor, though. Microsoft acknowledges this
1632 as a known bug and apparently has no plans to fix it.</p>
1634 <p>Fetchmail using IMAP supports the proprietary NTLM mode used
1635 with M$ Exchange servers. To enable this, configure fetchmail with
1636 the --enable-NTLM option and recompile it. Specify a user option
1637 value that looks like 'user@domain': the part to the left of the @
1638 will be passed as the username and the part to the right as the
1641 <p>M$ Exchange violates the POP3 and IMAP RFCs. Its LIST command
1642 does not reveal the real sizes of mail in the pop mailbox, but the
1643 sizes of the compressed versions in the exchange mail database
1644 (thanks to Arjan De Vet and Guido Van Rooij for alerting us to this
1647 <p>Fetchmail works with M$ Exchange, despite this brain damage. Two
1648 features are compromised. One is that the --limit option will not
1649 work right (it will check against compressed and not actual sizes).
1650 The other is that a too-small SIZE argument may be passed to your
1651 ESMTP listener, assuming you're using one (this should not be a
1652 problem unless the actual size of the message is above the
1653 listener's configured length limit).</p>
1655 <p>Somewhat belatedly, I've learned that there's supposed to be a
1656 registry bit that can fix this breakage:</p>
1659 HKEY_LOCAL_MACHINE\System\CurrentControlSet\Services\MsExchangeIs\Parameters
1660 System\Pop3 Compatibility
1663 <p>This is a bitmask that controls the variations from the standard
1664 protocol. The bits defined are:</p>
1667 <dt>0x00000001:</dt>
1669 <dd>Report exact message sizes for the LIST command</dd>
1671 <dt>0x00000002:</dt>
1673 <dd>Allow arbitrary linear whitespace between commands and
1676 <dt>0x00000004:</dt>
1678 <dd>Enable the LAST command</dd>
1680 <dt>0x00000008:</dt>
1682 <dd>Allow an empty PASS command (needed for users with blank
1683 passwords, but illegal in the protocol)</dd>
1685 <dt>0x00000010:</dt>
1687 <dd>Relax the length restrictions for arguments to commands
1688 (protocol requires 40, but some user names may be longer than
1691 <dt>0x00000020:</dt>
1693 <dd>Allow spaces in the argument to the USER command.</dd>
1696 <p>There's another one that may be useful to know about:</p>
1699 KEY_LOCAL_MACHINE\System\CurrentControlSet\Services\MsExchangeIs\Parameters
1700 System\Pop3 Performance
1704 <dt>0x00000001:</dt>
1706 <dd>Render messages to a temporary stream instead of sending
1707 directly from the database (should always be on)</dd>
1709 <dt>0x00000002: Flag unrenderable messages (instead of just failing
1710 commands) (should only be on if you are seeing the problems
1711 reported in KB Q168109)</dt>
1713 <dt>0x00000004:</dt>
1715 <dd>Return from the QUIT command before all messages have been
1719 <p>The Microsoft pod-person who revealed this information to me
1720 admitted that he couldn't find it anywhere in their public
1723 <p>Another specific problem we have seen with Exchange servers has
1724 as its symptom a response to LOGIN that says "NO Ambiguous Alias".
1725 Grant Edwards writes:</p>
1727 <p>This means that Exchange Server is too f*&#ing stupid to
1728 figure out which mailbox belongs to you. Instead of actually
1729 keeping track of which inbox belongs to which user, it uses some
1730 half-witted, guess-o-matic heuristic to try to guess your mailbox
1731 name from your username.</p>
1733 <p>In your case it doesn't work because your username maps to more
1734 than one mailbox. For some people it doesn't work because their
1735 username maps to zero mailboxes. This is yet another inept, lame,
1736 almost criminally negligent design decision from our friends in
1739 <p>You've got several options:</p>
1742 <li>Get your administrator to configure the server so that
1743 usernames and mailbox names are the same.</li>
1745 <li>Get your administrator to add an alias that maps your username
1746 explicitly to your mailbox name.</li>
1749 <p>But, the best option involves a tactical nuclear weapon (an old
1750 ASROC will do), pissing off a lot people who live downwind from
1751 Redmond, and your choice of any Linux, NetBSD, FreeBSD, or Solaris
1755 <h2><a id="S3" name="S3">S3. How can I use fetchmail with HP
1758 <p>No special configuration is required, but OpenMail versions
1759 prior to 6.0 have an annoying bug similar to the big one in <a
1760 href="#S2">Microsoft Exchange</a>. The message sizes it gives in
1761 the LIST are rounded to the nearest 1024 bytes. It also has a nasty
1762 habit of discarding headers it doesn't recognize, such as X- and
1763 Resent- headers.</p>
1765 <p>As with M$ Exchange, the only real fix for these problems is to
1766 get a POP (or preferably IMAP) server that isn't brain-dead.
1767 OpenMail's project manager claims these bugs have been fixed in
1770 <p>We've had a more recent report (December 2001) that the TOP
1771 command fails, returning only one line regrardless of its argument,
1772 on something identifying itself as "OpenMail POP3 interface".</p>
1775 <h2><a id="S4" name="S4">S4. How can I use fetchmail with Novell GroupWise?</a></h2>
1777 <p>The Novell GroupWise IMAP server would be better named
1778 GroupFoolish; it is (according to the designer of IMAP) unusably
1779 broken. Among other things, it doesn't include a required content
1780 length in its BODY[TEXT] response.</p>
1782 <p>Fetchmail works around this problem, but we strongly recommend
1783 voting with your dollars for a server that isn't brain-dead. If you
1784 stick with code as shoddy as GroupWise seems to be, you will
1785 probably pay for it with other problems.</p>
1788 <h2><a id="S5" name="S5">S5. How can I use fetchmail with
1789 InterChange?</a></h2>
1791 <p>You can't. At least not if you want to be able to see
1792 attachments. InterChange has a bug similar to the MailMax server;
1793 it reports the message length with attachments but doesn't download
1794 them on TOP or RETR.</p>
1796 <p>On Jan 9 2001, the people at InfiniteMail sent me mail informing
1797 me that their new 3.61.08 release of InterChange fixes this
1798 problem. I don't have any reports one way or the other yet.</p>
1801 <h2><a id="S6" name="S6">S6. How can I use fetchmail with MailMax?</a></h2>
1803 <p>You can't. At least not if you want to be able to see
1804 attachments. MailMax has a bug; it reports the message length with
1805 attachments but doesn't download them on TOP or RETR.</p>
1807 <p>Also, we're told that TOP sometimes fails to retrieve the entire
1808 message even when enough lines have been specified. The MailMax
1809 developers have acknowledged this bug as of 4 May 2000, but there
1810 is no fix yet. If you must use this server, force RETR with the
1811 <tt>fetchall</tt> option.</p>
1814 <h2><a id="S7" name="S7">S7. How can I use fetchmail with FTGate?</a></h2>
1816 <p>The FTGate V2 server (and possibly older versions as well) has a
1817 weird bug. It answers OK twice to a TOP request! Use the
1818 <code>fetchall</code> option to force use of RETR and work around
1822 <h2><a id="I1" name="I1">I1. How can I use fetchmail with CompuServe RPA?</a></h2>
1824 <p>First, make sure your fetchmail has the RPA support compiled in.
1825 Stock fetchmail binaries (such as you might get from an RPM) don't.
1826 You can check this by looking at the output of <code>fetchmail
1827 -V</code>; if you see the string "+RPA" after the version ID you're
1828 good to go, otherwise you'll have to build your own from sources
1829 (see the INSTALL file in the source distribution for
1832 <p>Give your CompuServe pass-phrase in lower case as your password.
1833 Add '@compuserve.com' to your user ID so that it looks like 'user
1834 <UserID>@compuserve.com', where <UserID> can be either
1835 your numerical userID or your E-mail nickname. An RPA-enabled
1836 fetchmail will automatically check for csi.com in the POP server's
1837 greeting line. If that's found, and your user ID ends with
1838 '@compuserve.com', it will query the server to see if it is
1839 RPA-capable, and if so do an RPA transaction rather than a
1840 plain-text password handshake.</p>
1842 <p><strong>Warning:</strong> the debug (-v -v) output of fetchmail
1843 will show your pass-phrase in Unicode!</p>
1845 <p>These two .fetchmailrc entries show the difference between an
1846 RPA and non-RPA configuration:</p>
1849 # This version will use RPA
1850 poll csi.com via "pop.site1.csi.com" with proto POP3 and options no dns
1851 user "CSERVE_USER@compuserve.com" there with password "CSERVE_PASSWORD"
1852 is LOCAL_USER here options fetchall stripcr
1854 # This version will not use RPA
1855 poll non-rpa.csi.com via "pop.site1.csi.com" with proto POP3 and options no dns
1856 user "CSERVE_USER" there with password "CSERVE_POP3_PASSWORD"
1857 is LOCAL_USER here options fetchall stripcr
1861 <h2><a id="I2" name="I2">I2. How can I use fetchmail with Demon
1862 Internet's SDPS?</a></h2>
1864 <h3>Single-drop mode</h3>
1866 <p>You can get fetchmail to download the email for just one user
1867 from Demon Internet's POP3 server by giving it a username
1868 consisting of your Demon user name followed by your account name,
1869 with an at-sign between them.</p>
1871 <p>For example, to download email for the user
1872 <philh@vision25.demon.co.uk>, you could use the following
1873 .fetchmailrc file:</p>
1876 set postmaster "philh"
1877 poll pop3.demon.co.uk with protocol POP3:
1878 user "philh@vision25" is philh
1881 <h3>Multi-drop mode</h3>
1883 <p>Demon Internet's SDPS service is an implementation of POP3. All
1884 messages have a Received: header added when they enter the
1885 maildrop, like this:</p>
1888 Received: from punt-1.mail.demon.net by mailstore for fred@xyz.demon.co.uk
1889 id 899963657:10:27896:0; Thu, 09 Jul 98 05:54:17 GMT
1892 <p>To enable multi-drop mode you need to tell fetchmail that
1893 'mailstore' is the name of the host which accepted the mail, and
1894 let it know the hostname part(s) of your E-mail address. The
1895 following example assumes that your hostname is xyz.demon.co.uk,
1896 and that you have also bought "mail forwarding" for the domain
1897 my-company.co.uk (in which case your MTA must also be configured to
1898 accept mail sent to user@my-company.co.uk)</p>
1901 poll pop3.demon.co.uk proto pop3 aka mailstore no dns:
1902 localdomains xyz.demon.co.uk my-company.co.uk
1906 <p>Note that Demon may delete mail on the server which is more than
1907 30 days old; see their <a
1908 href="http://www.demon.net/helpdesk/products/mail/sdps-tech.shtml">POP3
1909 page</a> for details.</p>
1911 <h3>The SDPS extension</h3>
1913 <p>There's a different way to do multidrop. It's not necessary on
1914 Demon Internet, since fetchmail can parse Received addresses, but
1915 the person who implemented this didn't know that. It may be useful
1916 if Demon Internet ever changes mail transports.</p>
1918 <p>SDPS includes a non-standard extension for retrieving the
1919 envelope of a message (*ENV), which fetchmail optionally supports
1920 if compiled with the --enable-SDPS option. If you have it, the
1921 first line of the fetchmail -V response will include the string
1924 <p>Once you have SDPS compiled in, fetchmail in POP3 mode will
1925 automatically detect when it's talking to a Demon Internet host in
1926 multidrop mode, and use the *ENV extension to get an envelope To
1929 <p>The autodetection works by looking at the hostname in the POP3
1930 greeting line; if you're accessing Demon Internet through a proxy
1931 it may fail. To force SDPS mode, pick "sdps" as your protocol.</p>
1934 <h2><a id="I3" name="I3">I3. How can I use fetchmail with usa.net's
1937 <p>Enable '<code>fetchall</code>'. A user reports that the 2.2
1938 version of USA.NET's POP server reports that you must use the
1939 '<code>fetchall</code>' option to make sure that all of the mail is
1940 retrieved, otherwise some may be left on the server. This is almost
1941 certainly a server bug.</p>
1943 <p>The usa.net servers (at least in their 2.2 version, June 1998)
1944 don't handle the TOP command properly, either. Regardless of the
1945 argument you give it, they retrieve only about 10 lines of the
1946 message. Fetchmail normally uses TOP for message retrieval in order
1947 to avoid marking messages seen, but '<code>fetchall</code>' forces
1948 it to use RETR instead.</p>
1950 <p>Also, we're told USA.NET adds a ton of hops to your messages.
1951 You may need to raise the MaxHopCount parameter in your sendmail.cf
1952 to avoid having fetched mail rejected.</p>
1954 <p>(Note: Other failure modes have been reported on usa.net's
1955 servers. They seem to be chronically flaky. We recommend finding
1956 another provider.)</p>
1959 <h2><a id="I4" name="I4">I4. How can I use fetchmail with geocities
1960 POP3 servers?</a></h2>
1962 <p>Nathan Cutler reports that the the mail.geocities.com POP3
1963 servers fail to include the first Received line of the message in
1964 the send to fetchmail. This can solve problems if your MUA
1965 interprets Received continuations as body lines and doesn't parse
1966 any of the following headers.</p>
1968 <p>Workaround is to use "mda" keyword or "-mda" switch:</p>
1971 mda "sed -e '1s/^\t/Received: /' | formail | /usr/bin/procmail -d <user>"
1974 <p>Replace \t with exactly one tabulation character.</p>
1976 <p>You should also consider using "fetchall" option because
1977 Geocities' servers sometimes think that the first 45 messages have
1978 already been read.</p>
1980 <p>Fix: Get an email provider that doesn't suck. The pop-up ads on
1981 Geocities are lame, you should boycott them anyway.</p>
1984 <h2><a id="I5" name="I5">I5. How can I use fetchmail with Hotmail or Lycos Webmail?</a></h2>
1986 <p>You can't directly. But you can use fetchmail with hotmail or lycos
1987 webmail with the help of the <a
1988 href='http://people.freenet.de/courierdave/'>HotWayDaemon</a>
1989 daemon. You don't even need to install hotwayd as a daemon in
1990 <samp>inetd.conf</samp> but can use it as a plugin. Your
1991 configuration should look like this:</p>
1994 poll localhost protocol pop3 tracepolls
1995 plugin "/usr/local/sbin/hotwayd -l 0 -p yourproxy:yourproxyport"
1996 username "youremail@hotmail.com" password "yourpassword"
2000 <p>As a second option you may consider using <a
2001 href="http://linux.cudeso.be/linuxdoc/gotmail.php">gotmail</a>.</p>
2004 <h2><a id="I6" name="I6">I6. How can I use fetchmail with MSN?</a></h2>
2006 <p>You can't. MSN uses something that looks like POP3, except the
2007 authentication part is nonstandard. And of course they don't
2008 document it, so nobody but their Windows clients can speak it.</p>
2010 <p>This is a customer lock-in tactic; we recommend boycotting MSN
2011 as the only appropriate response.</p>
2013 <p>As of 5.0.8, we have support for the client side of NTLM
2014 authentication. It's possible this may enable fetchmail to talk to
2015 MSN; if so, somebody should report it so this FAQ can be
2019 <h2><a id="I7" name="I7">I7. How can I use fetchmail with SpryNet?</a></h2>
2021 <p>The SpryNet POP3 servers mark a message queried with TOP as
2022 seen. This means that if your connection drops in mid-message, it
2023 may end up invisibly stuck on your mail spool. Use the
2024 <code>fetchall</code> flag to ensure that it's recovered on the
2028 <h2><a id="I8" name="I8">I8. How can I use fetchmail with comcast.net?</a></h2>
2030 <p>Stock fetchmail will work with a comcast.net server...<em>but</em>
2031 the Maillennium POP3 server comcat uses seems to have an 80K limit on
2032 the length of downloaded messages if you use POP3 TOP to retrieve.
2033 Anything larger is silently truncated. Don't mistake this for a
2034 fetchmail bug. (Reported July 2003.)</p>
2036 <p>Workaround: use the <tt>fetchall</tt> option.</p>
2039 <h2><a id="K1" name="K1">K1. How can I use fetchmail with SOCKS?</a></h2>
2041 <p>Giuseppe Guerini added a --with-socks option that supports
2042 linking with socks library. If you specify the value of this option
2043 as "yes", the configure script will try to find the Rconnect
2044 library and set the makefile up to link it. You can also specify a
2045 directory containing the Rconnect library.</p>
2047 <p>Alan Schmitt has added a similar --with-socks5 option that may
2048 work better if you have a recent version of the SOCKS library.</p>
2051 <h2><a id="K2" name="K2">K2. How can I use fetchmail with IPv6 and
2054 <p>To use fetchmail with IPv6, you need a system that supports
2055 IPv6, the "Basic Socket Interface Extensions for IPv6" (RFC 2133).
2058 <p>The NRL IPv6+IPsec software distribution can be obtained from:
2060 href="http://web.mit.edu/network/isakmp">http://web.mit.edu/network/isakmp</a></p>
2062 <p>More information on using IPv6 with Linux can be obtained
2067 href="http://www.bieringer.de/linux/IPv6/IPv6-HOWTO/IPv6-HOWTO.html">
2068 http://www.bieringer.de/linux/IPv6/IPv6-HOWTO/IPv6-HOWTO.html</a></li>
2072 <h2><a id="K3" name="K3">K3. How can I get fetchmail to work with
2075 <p>Use the <tt>plugin</tt> option. This is dead simple with
2079 plugin "ssh %h /usr/sbin/imapd"
2082 <p>You may have to use a different absolute pathname, whatever the
2083 location of imapd on your mailserver is. This option tells
2084 fetchmail that instead of opening a connection on the server's port
2085 143 and doing standard IMAP authentication, fetchmail should ssh to
2086 the server and run imapd, using the more secure ssh authentication
2087 (as well as getting ssh's end-to-end encryption). Most IMAP daemons
2088 will detect that they've been called from the command line and
2089 assume the connection is preauthenticated.</p>
2091 <p>POP3 daemons aren't quite as smart. They won't know they are
2092 preauthenticated in this mode, so you'll actually have to ship your
2093 password. It will be under ssh encryption, though, so that
2094 shouldn't be a problem.</p>
2097 <h2><a id="K4" name="K4">K4. What do I have to do to use the
2098 IMAP-GSS protocol?</a></h2>
2100 <p>Fetchmail can use RFC1731 GSSAPI authorization to safely
2101 identify you to your IMAP server, as long as you can share Kerberos
2102 V credentials with your mail host and you have a GSSAPI-capable
2103 IMAP server. UW-IMAP (available via FTP at <a
2104 href="ftp://ftp.cac.washington.edu/mail/">ftp.cac.washington.edu</a>)
2105 is the only one I'm aware of and the one I recommend anyway for
2106 other reasons. You'll need version 4.1-FINAL or greater though, and
2107 it has to have GSS support compiled in.</p>
2109 <p>Neither UW-IMAP nor fetchmail compile in support for GSS by
2110 default, since it requires libraries from the Kerberos V
2111 distribution (available via FTP at <a
2112 href="ftp://athena-dist.mit.edu/pub/ATHENA/kerberos">athena-dist.mit.edu</a>).
2113 If you have these, compiling in GSS support is simple: add a
2114 <code>--with-gssapi=[/path/to/krb5/root]</code> option to
2115 configure. For instance, I have all of my Kerberos V libraries
2116 installed under /usr/krb5 so I run <code>configure
2117 --with-gssapi=/usr/krb5</code></p>
2119 <p>Setting up Kerberos V authentication is beyond the scope of this
2120 FAQ (you may find Jim Rome's paper <a
2121 href="http://www.ornl.gov/~jar/HowToKerb.html">How to Kerberize
2122 your site</a> helpful), but you'll at least need to add a
2123 credential for imap/[mailhost] to the keytab of the mail server
2124 (IMAP doesn't just use the host key). Then you'll need to have your
2125 credentials ready on your machine (cf. kinit).</p>
2127 <p>After that things are very simple. Set your protocol to imap-gss
2128 in your .fetchmailrc, and omit the password, since imap-gss doesn't
2129 need one. You can specify a username if you want, but this is only
2130 useful if your mailbox belongs to a username different from your
2131 Kerberos principal.</p>
2133 <p>Now you don't have to worry about your password appearing in
2134 cleartext in your .fetchmailrc, or across the network.</p>
2137 <h2><a id="K5" name="K5">K5. How can I use fetchmail with
2140 <p>You'll need to have the <a
2141 href="http://www.openssl.org/">OpenSSL</a> libraries installed.
2142 Configure with --with-ssl. If you have the OpenSSL libraries
2143 installed in the default location (/usr/local/ssl) ths will
2144 suffice. If you have them installed in a non-default location,
2145 you'll need to specify it as an argument to --with-ssl after an
2148 <p>Note that there is a known bug in the implementation of SSL_peek
2149 under OpenSSL versions 0.9.5 and older that fetchmail occasionally
2150 tripped over, causing hangs. It is recommended that you install
2153 <p>Fetchmail binaries built this way support <code>ssl</code>,
2154 <code>sslkey</code>, and <code>sslcert</code> options that control
2155 SSL encryption. You will need to have an SSL-enabled mailserver to
2156 use these options. See the manual page for details and some words
2157 of care on the limited security provided.</p>
2159 <p>If your open OpenSSL session dies with a message that complains
2160 "PRNG not seeded", update or improve your operating system. This
2161 means that the OpenSSL library on your machine has been unable to
2162 locate a source of random bits from which to seed its random-number
2163 generator; normally these come from the <tt>/dev/urandom</tt>, and
2164 this message probably means your OS doesn't have that device.</p>
2166 <p>An interactive program could seed the random number generator
2167 from keystroke timings or some other form of user input. Because
2168 fetchmail is primarily designed to run forever as a background
2169 daemon, that option is not available in this case.</p>
2171 <p>If you don't have the libraries installed, but do have the
2172 OpenSSL utility toolkit, something like this may work:</p>
2175 poll MYSERVER port 993 plugin "openssl s_client -connect %h:%p"
2176 protocol imap username MYUSERNAME password MYPASSWORD
2179 <p>You should note that SSL is only secure against a "man-in-the-middle"
2180 attack if the client is able to verify that the peer's public key is the
2181 correct one, and has not been substituted by an attacker. fetchmail can do
2182 this in one of two ways: by verifying the SSL certificate, or by checking
2183 the fingerprint of the peer's public key.</p>
2185 <p>There are three parts to SSL certificate verification: checking that the
2186 domain name in the certificate matches the hostname you asked to connect to;
2187 checking that the certificate expiry date has not passed; and checking that
2188 the certificate has been signed by a known Certificate Authority (CA). This
2189 last step takes some preparation, as you need to install the root
2190 certificates of all the CA's which you might come across.</p>
2192 <p>The easiest way to do this is using the root CA keys supplied in the
2193 OpenSSL distribution, which means you need to download and unpack the
2194 source tarball from www.openssl.org. Once you have done that:</p>
2197 <li><code>mkdir /etc/ssl/certs</code></li>
2198 <li>in the openssl-x.x.x/certs directory: <code>cp *.pem /etc/ssl/certs/</code></li>
2199 <li>in the openssl-x.x.x/tools directory: edit c_rehash and set
2200 <code>$dir="/etc/ssl"</code></li>
2201 <li>run "perl c_rehash". This generates a number of symlinks within the
2202 /etc/ssl/certs/ directory</li>
2205 <p>Now in .fetchmailrc, set option sslcertpath to point to this
2209 poll pop3.example.com proto pop3 uidl no dns
2210 user foobar@example.com password xyzzy is foobar ssl sslcertpath /etc/ssl/certs
2213 <p>If the server certificate has not been signed by a known CA (e.g. it is a
2214 self-signed certificate), then this certificate validation will always
2217 <p>Certificate verification is always attempted. If it fails, by default a
2218 warning is printed but the connection carries on (which means you are not
2219 protected against attack). If your server's certificate has been properly
2220 set up and verifies correctly, then add the "sslcertck" option to enforce
2221 validation. If your server doesn't have a valid certificate though (e.g. it
2222 has a self-signed certificate) then it will never verify, and the only way
2223 you can protect yourself is by checking the fingerprint.</p>
2225 <p>To check the peer fingerprint: first use fetchmail -v once to connect to
2226 the host, at a time when you are pretty sure that there is no attack in
2227 progress (e.g. you are not traversing any untrusted network to reach the
2228 server). Make a note of the fingerprint shown. Now embed this in your
2229 .fetchmailrc using the sslfingerprint option: e.g.</p>
2232 poll pop3.example.com proto pop3 uidl no dns
2233 user foobar@example.com password xyzzy is foobar ssl sslfingerprint "67:3E:02:94:D3:5B:C3:16:86:71:37:01:B1:3B:BC:E2"
2236 <p>When you next connect, the public key presented by the server will be
2237 verified against the fingerprint given. If it's different, it may mean that
2238 a man-in-the-middle attack is in progress - or it might just mean that the
2239 server changed its key. It's up to you to determine which has happened.</p>
2242 <h2><a id="K6" name="K6">K6. How can I tell fetchmail not to use TLS
2243 if the server advertises it?</a></h2>
2245 <p>Some servers advertise STLS (POP3) or STARTTLS (IMAP), and fetchmail
2246 will automatically attempt TLS negotiation if SSL was enabled at compile
2247 time. This can however cause problems if the upstream didn't configure
2248 his certificates properly.</p>
2250 <p>In order to prevent fetchmail from trying TLS (STLS, STARTTLS)
2251 negotiation, add this option:</p>
2253 <pre>sslproto ssl23</pre>
2255 <p>This restricts fetchmail's SSL/TLS protocol choice from the default
2256 "SSLv2, SSLv3, TLSv1" to the two SSL variants, disabling TLSv1. Note
2257 however that this causes the connection to be unencrypted unless an
2258 encrypting "plugin" is used or SSL is requested explicitly.</p>
2261 <h2><a id="R1" name="R1">R1. Fetchmail isn't working, and -v shows
2262 'SMTP connect failed' messages.</a></h2>
2264 <p>Fetchmail itself is probably working, but your SMTP port 25
2265 listener is down or inaccessible.</p>
2267 <p>The first thing to check is if you can telnet to port 25 on your
2268 smtp host (which is normally 'localhost' unless you've specified an
2269 smtp option in your .fetchmailrc or on the command line) and get a
2270 greeting line from the listener. If the SMTP host is inaccessible
2271 or the listener is down, fix that first.</p>
2273 <p>In Red Hat Linux 6.x, SMTP is disabled by default. To fix this,
2274 set "DAEMON=yes" in your /etc/sysconfig/sendmail file, then restart
2275 sendmail by running "/sbin/service sendmail restart".</p>
2277 <p>If the listener seems to be up when you test with telnet, the
2278 most benign and typical problem is that the listener had a
2279 momentary seizure due to resource exhaustion while fetchmail was
2280 polling it -- process table full or some other problem that stopped
2281 the listener process from forking. If your SMTP host is not
2282 'localhost' or something else in /etc/hosts, the fetchmail glitch
2283 could also have been caused by transient nameserver failure.</p>
2285 <p>Try running fetchmail -v again; if it succeeds, you had one of
2286 these kinds of transient glitch. You can ignore these hiccups,
2287 because a future fetchmail run will get the mail through.</p>
2289 <p>If the listener tests up, but you have chronic failures trying
2290 to connect to it anyway, your problem is more serious. One way to
2291 work around chronic SMTP connect problems is to use --mda. But this
2292 only attacks the symptom; you may have a DNS or TCP routing
2293 problem. You should really try to figure out what's going on
2294 underneath before it bites you some other way.</p>
2296 <p>We have one report (from toby@eskimo.com) that you can sometimes
2297 solve such problems by doing an <code>smtp</code> declaration with
2298 an IP address that your routing table maps to something other than
2299 the loopback device (he used ppp0).</p>
2301 <p>We also have a report that this error can be caused by having an
2302 /etc/hosts file that associates your client host name with more
2303 than one IP address.</p>
2305 <p>It's also possible that your DNS configuration isn't looking at
2306 <code>/etc/hosts</code> at all. If you're using libc5, look at
2307 <code>/etc/resolv.conf</code>; it should say something like:</p>
2313 <p>so your <code>/etc/hosts</code> file is checked first. If you're
2314 running GNU libc6, check your <code>/etc/nsswitch.conf</code> file.
2315 Make sure it says something like</p>
2321 <p>again, in order to make sure <code>/etc/hosts</code> is seen
2324 <p>If you have a hostname set for your machine, and this hostname
2325 does not appear in /etc/hosts, you will be able to telnet to port
2326 25 and even send a mail with rcpt to: user@host-not-in-/etc/hosts,
2327 but fetchmail can't seem to get in touch with sendmail, no matter
2328 what you set smtpaddress to.</p>
2330 <p>We had another report from a Linux user of fetchmail 2.1 who
2331 solved his SMTP connection problem by removing the reference to
2332 -lresolv from his link line and relinking. Apparently in some older
2333 Linux distributions the libc bind library version works better.</p>
2335 <p>As of 2.2, the configure script has been hacked so the bind
2336 library is linked only if it is actually needed. So under Linux it
2337 won't be, and this particular cause should go away.</p>
2340 <h2><a id="R2" name="R2">R2. When I try to configure an MDA,
2341 fetchmail doesn't work.</a></h2>
2343 <p>(I hear this one from people who have run into the blank-line
2344 problem in <a href="#X1">X1</a>.)</p>
2346 <p>Try sending yourself test mail and retrieving it using the
2347 command-line options '<code>-k -m cat</code>'. This will dump
2348 exactly what fetchmail retrieves to standard output (plus the
2349 Received line fetchmail itself adds to the headers).</p>
2351 <p>If the dump doesn't match what shows up in your mailbox when you
2352 configure an MDA, your MDA is mangling the message. If it doesn't
2353 match what you sent, then fetchmail or something on the server is
2357 <h2><a id="R3" name="R3">R3. Fetchmail dumps core when given an
2358 invalid rc file.</a></h2>
2360 <p>This is usually reported from AIX or Ultrix, but has even been
2361 known to happen on Linuxes without a recent version of
2362 <code>flex</code> installed. The problem appears to be a result of
2363 building with an archaic version of lex.</p>
2365 <p>Workaround: fix the syntax of your .fetchmailrc file.</p>
2367 <p>Fix: build and install the latest version of <a
2368 href="ftp://prep.ai.mit.edu/~ftp/pub/gnu">flex</a> from the Free
2369 Software Foundation. An FSF <a
2370 href="http://www.gnu.ai.mit.edu/order/ftp.html">mirror site</a>
2371 will help you get it faster.</p>
2374 <h2><a id="R4" name="R4">R4. Fetchmail dumps core in -V mode, but
2375 operates normally otherwise.</a></h2>
2377 <p>We've had this reported to us under Linux using libc-5.4.17 and
2378 gcc-2.7.2. It does not occur with libc-5.3.12 or earlier
2381 <p>Workaround: link with GNU malloc rather than the stock C library
2384 <p>We're told there is some problem with the malloc() code in that
2385 version which makes it fragile in the presence of multiple free()
2386 calls on the same pointer (the malloc arena gets corrupted).
2387 Unfortunately it appears from doing gdb traces that whatever free()
2388 calls producing the problem are being made by the C library itself,
2389 not the fetchmail code (they're all from within fclose, and not an
2390 fclose called directly by fetchmail, either).</p>
2393 <h2><a id="R5" name="R5">R5. Running fetchmail in daemon mode
2394 doesn't work.</a><br/>
2397 <p>We have one report from a SunOS 4.1.4 user that trying to run
2398 fetchmail in detached daemon mode doesn't work, but that using the
2399 same options with -N (nodetach) is OK. We have another report of
2400 similar behavior from one Linux user, but many other Linux users
2401 reportt no problem.</p>
2403 <p>If this happens, you have a specific portability problem with
2404 the code in daemon.c that detaches and backgrounds the daemon
2405 fetchmail. The isolated Linux case has been chased down to a
2406 failure in dup(2) that may reflect a glibc bug.</p>
2408 <p>As a workaround, you can start fetchmail with -N and an
2409 ampersand to background it. A Sun user recommends this:</p>
2412 (fetchmail --nodetach <other params> &)
2415 <p>The extra pair of parens is significant --- it makes sure that
2416 the process detaches from the initial shell (one more shell is
2417 started and dies immediately, detaching fetchmail and making it
2418 child of PID 1). This is important when you start fetchmail
2419 interactively and than quit interactive shell. The line above makes
2420 sure fetchmail lives after that!</p>
2423 <h2><a id="R6" name="R6">R6. Fetchmail randomly dies with socket
2426 <p>Check the MTU value in your PPP interface reported by
2427 <code>/sbin/ifconfig</code>. If it's over 600, change it in your
2428 PPP options file. (<code>/etc/ppp/options</code> on my box). Here
2429 are option values that work:</p>
2436 <p>Another circumstance that can trigger this is if you are polling
2437 a virtual-mail-server name that is round-robin connected to
2438 different actual servers, so you get different IP addresses on
2439 different poll cycles. To work around this, change the poll name
2440 either to the real name of one of the servers in the ring or to a
2441 corresponding IP address.</p>
2444 <h2><a id="R7" name="R7">R7. Fetchmail running as root stopped
2445 working after an OS upgrade</a></h2>
2447 <p>In RH 6.0, the HOME value in the boot-time root environment
2448 changed from /root to / as the result of a change in init. Move
2449 your .fetchmailrc or use a -f option to explicitly point at the
2450 file. (Oddly, a similar problem has been reported from Debian
2454 <h2><a id="R8" name="R8">R8. Fetchmail is timing out after fetching
2455 certain messages but before deleting them</a></h2>
2457 <p>There's a TCP/IP stalling problem under Redhat 6.0 (and possibly
2458 other recent Linuxes) that can cause this symptom. Brian Boutel
2462 <p>TCP timestamps are turned on on my Linux boxes (I assume it's
2463 now the default). This uses 12 extra bytes per segment. When the
2464 tcp connection starts, the other end agrees a MSS of 1460, and then
2465 fragments 1460 byte chunks into 1448 and 12, because is is not
2466 allowing for the timestamp.</p>
2468 <p>Then, for reasons I can't explain, it waits a long time
2469 (typically 2 minutes) after the ack is sent before sending the next
2470 (fragmented) packet. Turning off tcp timestamps avoids the
2471 fragmentation and restores normal behaviour. To do this,
2474 <p>echo 0 > /proc/sys/net/ipv4/tcp_timestamps</p>
2476 <p>I'm still unclear about the details of why this is happening. At
2477 least [now] I am now getting good performance and no queue
2482 <h2><a id="R9" name="R9">R9. Fetchmail is timing out during message
2485 <p>This is probably a general networking issue. Sending a "RETR"
2486 command will cause the server to start sending large amounts of
2487 data, which means large packets. If your networking layer has a
2488 packet-fragmentation problem, that's where you'll see it.</p>
2491 <h2><a id="R10" name="R10">R10. Fetchmail is dying with
2494 <p>This probably means you have an <code>mda</code> option. Your
2495 MDA is croaking while being passed a message. Best fix is to remove
2496 the <code>mda</code> option and pass mail to your port 25 SMTP
2499 <p>If for some reason you are invoking sendmail via the
2500 <tt>mda</tt> option (rather than delivering to port 25 via smtp),
2501 don't forget to include the -i switch. Otherwise you will
2502 occasionally get mysterious delivery failures with a SIGPIPE as the
2503 sendmail instance dies. The problem is messages with a single dot
2504 at start of a text line.</p>
2507 <h2><a id="R11" name="R11">R11. My server is hanging or emitting
2508 errors on CAPA.</a></h2>
2510 <p>Your POP3 server is broken. You can work around this with the
2511 declaration <tt>auth password</tt> in your .fetchmailrc.</p>
2514 <h2><a id="R12" name="R12">R12. Fetchmail isn't working and reports
2515 getaddrinfo errors.</a></h2>
2516 <ol><li>Make sure you haven't mistyped the host name or address, and that
2517 your DNS is working. If you cannot fix DNS, give the numeric host
2518 literal, for instance, 192.168.0.1</li>
2519 <li>Make sure your <code>/etc/services</code> file (or other
2520 services database) contains the necessary service entries. If you
2521 cannot fix the services database, use the --service option and give the
2522 numeric port address. Common port addresses are:<table>
2523 <tr><th>service</th><th>port</th></tr>
2524 <tr><td>IMAP</td><td>143</td></tr>
2525 <tr><td>IMAP+SSL</td><td>993</td></tr>
2526 <tr><td>POP3</td><td>110</td></tr>
2527 <tr><td>POP3+SSL</td><td>995</td></tr>
2530 <h2><a id="H1" name="H1">H1. Fetchmail hangs when used with
2533 <p>Your problem may be with pppd's 'demand' option. We have a
2534 report that fetchmail doesn't play well with it, but works with
2535 pppd if 'demand' is turned off. We have no idea why this is.</p>
2538 <h2><a id="H2" name="H2">H2. Fetchmail hangs during the MAIL FROM
2541 <p>The symptom: 'fetchmail -v' retrieves mail fine, but appears to
2542 hang after sending the MAIL FROM command</p>
2545 SMTP> MAIL FROM: <someone@somewhere>
2548 <p>The hang is actually occuring when sendmail looks up a sender's
2549 address in DNS. The problem isn't in fetchmail but in the
2550 configuration of sendmail. You must enable the 'nodns' and
2551 'nocanonify' features of sendmail.</p>
2553 <p>Here was my fix for RedHat 7.2:</p>
2556 <li># cd /etc/mail</li>
2558 <li># cp sendmail.mc sendmail-mine.mc</li>
2560 <li>Edit sendmail-mine.mc and add lines:
2568 <li>Build a new sendmail.cf
2571 # m4 sendmail-mine.cf > /etc/sendmail.cf
2575 <li>Restart sendmail.</li>
2578 <p>For more details consult the file
2579 /usr/share/sendmail-cf/README.</p>
2582 <h2><a id="H3" name="H3">H3. Fetchmail hangs while fetching
2585 <p>The symption: 'fetchmail -v' retrieves the first few messages,
2586 but hangs returning:</p>
2589 fetchmail: SMTP< 550 5.0.0 Access denied
2590 fetchmail: SMTP> RSET
2591 fetchmail: SMTP< 250 2.0.0 Reset state
2592 .......fetchmail: flushed
2593 fetchmail: POP3> DELE 1
2594 fetchmail: POP3< +OK marked deleted
2597 <p>Check and see if you're allowing sendmail connections through
2600 <p>Adding 'sendmail : 127.0.0.1' to /etc/hosts.allow could solve
2604 <h2><a id="D1" name="D1">D1. I think I've set up fetchmail
2605 correctly, but I'm not getting any mail.</a></h2>
2607 <p>Maybe you have a .forward or alias set up that you've forgotten
2608 about. You should probably remove it.</p>
2610 <p>Or maybe you're trying to run fetchmail in multidrop mode as
2611 root without a .fetchmailrc file. This doesn't do what you think it
2612 should; see question <a href="#C1">C1</a>.</p>
2614 <p>Or you may not be connecting to the SMTP listener. Run fetchmail
2615 -v and see <a href="#R1">R1</a>.</p>
2617 <p>Or you may have your local user set incorrectly. In the
2621 user 'remoteuser' there with password '*' is 'localuser' here
2624 <p>make sure that 'localuser' does exist and can receive mail.</p>
2627 <h2><a id="D2" name="D2">D2. All my mail seems to disappear after a
2628 dropped connection.</a></h2>
2630 <p>One POP3 daemon used in the Berkeley Unix world that reports
2631 itself as POP3 version 1.004 actually throws the queue away. 1.005
2632 fixed that. If you're running this one, upgrade immediately. (It
2633 also truncates long lines at column 1024)</p>
2635 <p>Many POP servers, if an interruption occurs, will restore the
2636 whole mail queue after about 10 minutes. Others will restore it
2637 right away. If you have an interruption and don't see it right
2638 away, cross your fingers and wait ten minutes before retrying.</p>
2640 <p>Some servers (such as Microsoft's NTMail) are mis-designed to
2641 restore the entire queue, including messages you have deleted. If
2642 you have one of these and it flakes out on you a lot, try setting a
2643 small <code>--fetchlimit</code> value. This will result in more IP
2644 connects to the server, but will mean it actually executes changes
2645 to the queue more often.</p>
2647 <p>Qualcomm's qpopper, used at many BSD Unix sites, is better
2648 behaved. If its connection is dropped, it will first execute all
2649 DELE commands as though you had issued a QUIT (this is a technical
2650 violation of the POP3 RFCs, but a good idea in a world of flaky
2651 phone lines). Then it will re-queue any message that was being
2652 downloaded at hangup time. Still, qpopper may require a noticeable
2653 amount of time to do deletions and clean up its queue. (Fetchmail
2654 waits a bit before retrying in order to avoid a 'lock busy'
2658 <h2><a id="D3" name="D3">D3. Mail that was being fetched when I
2659 interrupted my fetchmail seems to have been vanished.</a></h2>
2661 <p>Fetchmail only sends a delete mail request to the server when
2662 either (a) it gets a positive delivery acknowledgment from the SMTP
2663 listener, or (b) it gets one of the spam-filter errors (see the
2664 description of the <code>antispam></code> option) from the
2665 listener. No interrupt can cause it to lose mail.</p>
2667 <p>However, IMAP2bis has a design problem in that its normal fetch
2668 command marks a message 'seen' as soon as the fetch command to get
2669 it is sent down. If for some reason the message isn't actually
2670 delivered (you take a line hit during the download, or your port 25
2671 listener can't find enough free disk space, or you interrupt the
2672 delivery in mid-message) that 'seen' message can lurk invisibly in
2673 your server mailbox forever.</p>
2675 <p>Workaround: add the '<code>fetchall</code>' keyword to your
2678 <p>Solution: switch to an <a href="http://www.imap.org">IMAP4</a>
2682 <h2><a id="M1" name="M1">M1. I've declared local names, but all my
2683 multidrop mail is going to root anyway.</a></h2>
2685 <p>Somehow your fetchmail is never recognizing the hostname part of
2686 recipient names it parses out of To/Cc/envelope-header lines as
2687 matching the name of the mailserver machine. To check this, run
2688 fetchmail in foreground with -v -v on. You will probably see a lot
2689 of messages with the format "line rejected, %s is not an alias of
2690 the mailserver" or "no address matches; forwarding to %s."</p>
2692 <p>These errors usually indicate some kind of DNS configuration
2693 problem either on the server or your client machine.</p>
2695 <p>The easiest workaround is to add a '<code>via</code>' option (if
2696 necessary) and add enough aka declarations to cover all of your
2697 mailserver's aliases, then say '<code>no dns</code>'. This will
2698 take DNS out of the picture (though it means mail may be
2699 uncollected if it's sent to an alias of the mailserver that you
2700 don't have listed).</p>
2702 <p>It would be better to fix your DNS, however. DNS problems can
2703 hurt you in lots of ways, for example by making your machines
2704 intermittently or permanently unreachable to the rest of the
2707 <p>Occasionally these errors indicate the sort of header-parsing
2708 problem described in <a href="#M7">M7</a>.</p>
2711 <h2><a id="M2" name="M2">M2. I can't seem to get fetchmail to route
2712 to a local domain properly.</a></h2>
2714 <p>A lot of people want to use fetchmail as a poor man's
2715 internetwork mail gateway, picking up mail accumulated for a whole
2716 domain in a single server mailbox and then routing based on what's
2717 in the To/Cc/Bcc lines.</p>
2719 <p>In general, this is not really a good idea. It would be smarter
2720 to just let the mail sit in the mailserver's queue and use
2721 fetchmail's ETRN or ODMR modes to trigger SMTP sends periodically
2722 (of course, this means you have to poll more frequently than the
2723 mailserver's expiration period). If you can't arrange this, try
2724 setting up a UUCP feed.</p>
2726 <p>If neither of these alternatives is available, multidrop mode
2727 may do (though you <em>are</em> going to get hurt by some mailing
2728 list software; see the caveats under THE USE AND ABUSE OF MULTIDROP
2729 MAILBOXES on the man page). If you want to try it, the way to do it
2730 is with the '<code>localdomains</code>' option.</p>
2732 <p>In general, if you use localdomains you need to make sure of two
2735 <p><strong>1. You've actually set up your .fetchmailrc entry to
2736 invoke multidrop mode.</strong></p>
2738 <p>Many people set a '<code>localdomains</code>' list and then
2739 forget that fetchmail wants to see more than one name (or the
2740 wildcard '*') in a '<code>here</code>' list before it will do
2741 multidrop routing.</p>
2743 <p><strong>2. You may have to set 'no envelope'.</strong></p>
2745 <p>Normally, multidrop mode tries to deduce an envelope address
2746 from a message before parsing the To/Cc/Bcc lines (this enables it
2747 to avoid losing to mailing list software that doesn't put a
2748 recipient address in the To lines).</p>
2750 <p>Some ways of accumulating a whole domain's messages in a single
2751 server mailbox mean it all ends up with a single envelope address
2752 that is useless for rerouting purposes. You may have to set
2753 '<code>no envelope</code>' to prevent fetchmail from being
2754 bamboozled by this.</p>
2756 <p>Check also answer <a href="#T1">T1</a> on a reliable way to do
2757 multidrop delivery if your ISP (or your mail redirection provider)
2761 <h2><a id="M3" name="M3">M3. I tried to run a mailing list using
2762 multidrop, and I have a mail loop!</a></h2>
2764 <p>This isn't fetchmail's fault. Check your mailing list. If the
2765 list expansion includes yourself or anybody else at your mailserver
2766 (that is, not on the client side) you've created a mail loop. Just
2767 chop the host part off any local addresses in the list.</p>
2769 <p>If you use sendmail, you can check the list expansion with
2770 <code>sendmail -bv</code>.</p>
2773 <h2><a id="M4" name="M4">M4. My multidrop fetchmail seems to be
2774 having DNS problems.</a></h2>
2776 <p>We have one report from a Linux user (not the same one as in <a
2777 href="#R1">R1</a>!) who solved this problem by removing the
2778 reference to -lresolv from his link line and relinking. Apparently
2779 in some older Linux distributions the libc5 bind library version
2782 <p>As of 2.2, the configure script has been hacked so the bind
2783 library is linked only if it is actually needed. So under Linux it
2784 won't be, and this problem should go away.</p>
2787 <h2><a id="M5" name="M5">M5. I'm seeing long DNS delays before each
2788 message is processed.</a></h2>
2790 <p>Use the '<code>aka</code>' option to pre-declare as many of your
2791 mailserver's DNS names as you can. When an address's host part
2792 matches an aka name, no DNS lookup needs to be done to check
2795 <p>If you're sure you've pre-declared all of your mailserver's DNS
2796 names, you can use the '<code>no dns</code>' option to prevent
2797 other hostname parts from being looked up at all.</p>
2799 <p>Sometimes delays are unavoidable. Some SMTP listeners try to
2800 call DNS on the From-address hostname as a way of checking that the
2801 address is valid.</p>
2804 <h2><a id="M6" name="M6">M6. How do I get multidrop mode to work
2805 with majordomo?</a></h2>
2807 <p>In order for sendmail to execute the command strings in the
2808 majordomo alias file, it is necessary for sendmail to think that
2809 the mail it receives via SMTP really is destined for a local user
2810 name. A normal virtual-domain setup results in delivery to the
2811 default mailbox, rather than expansion through majordomo.</p>
2813 <p>Michael <michael@bizsystems.com> gave us a recipe for
2814 dealing with this case that pairs a run control file like this:</p>
2817 poll your.pop3.server proto pop3:
2819 localdomains virtual.localdomain1.com virtual.localdomain2.com ...
2820 user yourISPusername is root * here,
2821 password yourISPpassword fetchall
2824 <p>with a hack on your local sendmail.cf like this:</p>
2827 #############################################
2828 # virtual info, local hack for ruleset 98 #
2829 #############################################
2831 # domains to treat as direct mapped local domain
2833 CVvirtual.localdomain1.com virtual.localdomain2.com ...
2834 ---------------------------
2836 -------------------------
2837 # handle virtual users
2839 R$+ <@ $=V . > $: $1 < @ $j . >
2840 R< @ > $+ < @ $=V . > $: $1 < @ $j . >
2841 R< @ > $+ $: $1
2842 R< error : $- $+ > $* $#error $@ $1 $: $2
2843 R< $+ > $+ < @ $+ > $: $>97 $1
2846 <p>This ruleset just strips virtual domain names off the addresses
2847 of incoming mail. Your sendmail must be 8.8 or newer for this to
2848 work. Michael says:</p>
2850 <blockquote>I use this scheme with 2 virtual domains and the
2851 default ISP user+domain and service about 30 mail accounts +
2852 majordomo on my inside pop3 server with fetchmail and sendmail
2856 <h2><a id="M7" name="M7">M7. Multidrop mode isn't parsing envelope
2857 addresses from my Received headers as it should.</a></h2>
2859 <p>It may happen that you're getting what appear to be well-formed
2860 sendmail Received headers, but fetchmail can't seem to extract an
2861 envelope address from them. There can be a couple of reasons for
2864 <h3>Spurious Received lines need to be skipped:</h3>
2866 <p>First, fetchmail might be looking at the wrong Received header.
2867 Normally it looks only on the first one it sees, on the theory that
2868 that one was last added and is going to be the one containing your
2869 mailserver's theory of who the message was addressed to.</p>
2871 <p>Some (unusual) mailserver configurations will generate extra
2872 Received lines which you need to skip. To arrange this, use the
2873 optional skip prefix argument of the 'envelope' option; you may
2874 need to say something like '<code>envelope 1 Received</code>' or
2875 '<code>envelope 2 Received</code>'.</p>
2877 <h3>The 'by' clause doesn't contain a mailserver alias:</h3>
2879 <p>When fetchmail parses a Received line that looks like</p>
2882 Received: from send103.yahoomail.com (send103.yahoomail.com [205.180.60.92])
2883 by iserv.ttns.net (8.8.5/8.8.5) with SMTP id RAA10088
2884 for <ksturgeon@fbceg.org>; Wed, 9 Sep 1998 17:01:59 -0700
2887 <p>it checks to see if 'iserv.ttns.net' is a DNS alias of your
2888 mailserver before accepting 'ksturgeon@fbceg.org' as an envelope
2889 address. This check might fail if your DNS were misconfigured, or
2890 if you were using 'no dns' and had failed to declare iserv.ttns.net
2891 as an alias of your server.</p>
2894 <h2><a id="M8" name="M8">M8. Users are getting multiple copies of
2897 <p>It's a consequence of multidrop. What's happening is that you
2898 have N users subscribed to the same list. The list software sends N
2899 copies, not knowing they will end up in the same multidrop box.
2900 Since they are both locally addressed to all N users, fetchmail
2901 delivers N copies to each user.</p>
2903 <p>Fetchmail tries to eliminate adjacent duplicate messages in a
2904 multidrop mailbox. However, this logic depends on the message-ID
2905 being identical in both copies. It also depends on the two copies
2906 being adjacent in the server mailbox. The former is usually the
2907 case, but the latter condition sometimes fails in a
2908 timing-dependent way if the server was processing multiple incoming
2911 <p>I could eliminate this problem by keeping a list of all
2912 message-IDs received during a poll so far and dropping any message
2913 that matches a seen mail ID. The trouble is that this is an O(N**2)
2914 operation that might significantly slow down the retrieval of large
2918 <h2><a id="X1" name="X1">X1. Spurious blank lines are appearing in
2919 the headers of fetched mail.</a></h2>
2921 <p>What's probably happening is that the POP/IMAP daemon on your
2922 mailserver is inserting a non-RFC822 header (like X-POP3-Rcpt:) and
2923 something in your delivery path (most likely an old version of the
2924 <em>deliver</em> program, which sendmail often calls to do local
2925 delivery) is failing to recognize it as a header.</p>
2927 <p>This is not fetchmail's problem. The first thing to try is
2928 installing a current version of <em>deliver</em>. If this doesn't
2929 work, try to figure out which other program in your mail path is
2930 inserting the blank line and replace that. If you can't do either
2931 of these things, pick a different MDA (such as procmail) and
2932 declare it with the '<code>mda</code>' option.</p>
2935 <h2><a id="X2" name="X2">X2. My mail client can't see a Subject
2938 <p>First, see <a href="#X1">X1</a>. This is quite probably the same
2939 problem (X-POP3-Rcpt header or something similar being inserted by
2940 the server and choked on by an old version of
2941 <em>deliver</em>).</p>
2943 <p>The O'Reilly sendmail book does warn that IDA sendmail doesn't
2944 process X- headers correctly. If this is your problem, all I can
2945 suggest is replacing IDA sendmail, because it's broken and not
2946 RFC822 conformant.</p>
2949 <h2><a id="X3" name="X3">X3. Messages containing "From" at start of
2950 line are being split.</a></h2>
2952 <p>If you know the messages aren't split in your server mailbox,
2953 then this is a problem with your POP/IMAP server, your client-side
2954 SMTP listener or your local delivery agent. Fetchmail cannot split
2957 <p>Some POP server daemons ignore Content-Length headers and split
2958 messages on From lines. We have one report that the 2.1 version of
2959 the BSD popper program (as distributed on Solaris 2.5 and
2960 elsewhere) is broken this way.</p>
2962 <p>You can test this. Declare an mda of 'cat' and send yourself one
2963 piece of mail containing "From" at start of a line. If you see a
2964 split message, your POP/IMAP server is at fault. Upgrade to a more
2967 <p>Sendmail and other SMTP listeners don't split RFC822 messages
2968 either. What's probably happening is either sendmail's local
2969 delivery agent or your mail reader are not quite RFC822-conformant
2970 and are breaking messages on what it thinks are Unix-style From
2971 headers. You can figure out which by looking at your client-side
2972 mailbox with vi or more. If the message is already split in your
2973 mailbox, your local delivery agent is the problem. If it's not,
2974 your mailreader is the problem.</p>
2976 <p>If you can't replace the offending program, take a look at your
2977 sendmail.cf file. There will likely be a line something like</p>
2980 Mlocal, P=/usr/bin/procmail, F=lsDFMShP, S=10, R=20/40, A=procmail -Y -d $u
2983 <p>describing your local delivery agent. Try inserting the 'E'
2984 option in the flags part (the F= string). This will make sendmail
2985 turn each dangerous start-of-line From into a >From, preventing
2986 programs further downstream from acting up.</p>
2989 <h2><a id="X4" name="X4">X4.</a> <a id="generic_mangling"
2990 name="generic_mangling">My mail is being mangled in a new and
2991 different way</a></h2>
2993 <p>The first thing you need to do is pin down what program is doing
2994 the mangling. We don't like getting bug reports about fetchmail
2995 that are actually due to some other program's malfeasance, so
2996 please go through this diagnostic sequence before sending us a
2999 <p>There are five possible culprits to consider, listed here in the
3000 order they pass your mail:</p>
3003 <li>Programs upstream of your server mailbox.</li>
3005 <li>The POP or IMAP server on your mailserver host.</li>
3007 <li>The fetchmail program itself.</li>
3009 <li>Your local sendmail.</li>
3011 <li>Your LDA (local delivery agent), as called by sendmail or
3012 specified by <code>mda</code>.</li>
3015 <p>Often it happens that fetchmail itself is OK, but using it
3016 exposes pre-existing bugs in your downstream software, or your
3017 downstream software has a bad interaction with POP/IMAP. You need
3018 to pin down exactly where the message is being garbled in order to
3019 deduce what is actually going on.</p>
3021 <p>The first thing to do is send yourself a test message, and
3022 retrieve it with a .fetchmailrc entry containing the following (or
3023 by running with the equivalent command-line options):</p>
3026 mda "cat >MBOX" keep fetchall
3029 <p>This will capture what fetchmail gets from the server, except
3030 for (a) the extra Received header line fetchmail prepends, (b)
3031 header address changes due to <code>rewrite</code>, and (c) any
3032 end-of-line changes due to the <code>forcecr</code> and
3033 <code>stripcr</code> options. MBOX will in fact contain what
3034 programs downstream of fetchmail see.</p>
3036 <p>The most common causes of mangling are bugs and
3037 misconfigurations in those downstream programs. If MBOX looks
3038 unmangled, you will know that is what is going on and that it is
3039 not fetchmail's problem. Take a look at the other FAQ items in this
3040 section for possible clues about how to fix your problem.</p>
3042 <p>If MBOX looks mangled, the next thing to do is compare it with
3043 your actual server mailbox (if possible). That's why you specified
3044 <code>keep</code>, so the server copy would not be deleted. If your
3045 server mailbox looks mangled, programs upstream of your server
3046 mailbox are at fault. Unfortunately there is probably little you
3047 can do about this aside from complaining to your site postmaster,
3048 and nothing at all fetchmail can do about it!</p>
3050 <p>More likely you'll find that the server copy looks OK. In that
3051 case either the POP/IMAP server or fetchmail is doing the mangling.
3052 To determine which, you'll need to telnet to the server port and
3053 simulate a fetchmail session yourself. This is not actually hard
3054 (both POP3 and IMAP are simple, text-only, line-oriented protocols)
3055 but requires some attention to detail. You should be able to use a
3056 fetchmail -v log as a model for a session, but remember that the
3057 "*" in your LOGIN or PASS command dump has to be replaced with your
3058 actual password.</p>
3060 <p>The objective of manually simulating fetchmail is so you can see
3061 exactly what fetchmail sees. If you see a mangled message, then
3062 your server is at fault, and you probably need to complain to your
3063 mailserver administrators. However, we like to know what the broken
3064 servers are so we can warn people away from them. So please send us
3065 a transcript of the session including the mangling <em>and the
3066 server's initial greeting line</em>. Please tell us anything else
3067 you think might be useful about the server, like the server host's
3068 operating system.</p>
3070 <p>If your manual fetchmail simulation shows an unmangled message,
3071 congratulations. You've found an actual fetchmail bug, which is a
3072 pretty rare thing these days. Complain to us and we'll fix it.
3073 Please include the session transcript of your manual fetchmail
3074 simulation along with the other things described in the FAQ entry
3075 on <a href="#G3">reporting bugs</a>.</p>
3078 <h2><a id="X5" name="X5">X5. Using POP3, retrievals seems to be
3079 fetching too much!</a></h2>
3081 <p>This may happen in versions of fetchmail after 4.4.1 and before
3082 4.4.8. Versions after 4.4.1 use POP3's TOP command rather than
3083 RETR, in order to avoid marking the message seen (leaving it unseen
3084 is helpful for later recovery if you lose your connection in the
3085 middle of a retrieval).</p>
3087 <p>Versions of fetchmail from 4.4.2 through 4.4.7 had a bad
3088 interaction with Eudora qpopper versions 2.3 and later. The TOP
3089 bounds check was fooled by an overflow condition in the TOP
3090 argument. Decrementing the TOP argument in 4.4.7 fixed this.</p>
3092 <p>Fix: Upgrade to a later version of fetchmail.</p>
3094 <p>Workaround: set the <code>fetchall</code> option. Under POP3
3095 this has the side effect of forcing RETR use.</p>
3098 <h2><a id="X6" name="X6">X6. My mail attachments are being dropped
3099 or mangled.</a></h2>
3101 <p>Fetchmail doesn't discard attachments; fetchmail doesn't have any idea
3102 that attachments are there. Fetchmail treats the body of each message as
3103 an uninterpreted byte stream and passes it through without alteration.
3104 If you are not receiving attachments through fetchmail, it is because
3105 your mailserver is not sending them to you.</p>
3107 <p>The fix for this is to replace your mailserver with one that works.
3108 If its operating system makes this difficult, you should replace its
3109 operating system with one that works. Windows- and NT-based POP servers
3110 seem especially prone to mangle attachments. If you are running one
3111 of these, replacing your server with a Unix machine is probably the
3112 only effective solution.</p>
3114 <p>We've had sporadic reports of problems with Microsoft Exchange and
3115 Outlook servers. These sometimes randomly fail to ship
3116 attachments to your client. This is a known bug, acknowledged by
3119 <p>They may also mangle the attachments they do pass through. If you
3120 see unreadable attachments with a ContentType of "application/x-tnef",
3121 you're having this problem. The <a
3122 href="http://world.std.com/~damned/software.html">TNEF</a> utility may
3125 <p>The Mail Max POP3 server and the InterChange and Imail IMAP
3126 servers are known to simply drop MIME attachments when uploading
3129 <p>We've also had a report that Lotus Notes sometimes trashes the
3130 MIME type of messages. In particular, it seems to modify MIME
3131 headers of type application/pdf, mangling the type to
3132 application/octet-stream. It may corrupt other MIME types as
3135 <p>The IMAP service of Lotus Domino has a known bug in the way it
3136 generates MIME Content-type headers (observed on Lotus Domino
3137 5.0.2b). It's a subtle one that doesn't show up when Netscape
3138 Messenger and other clients use a FETCH BODY[] to grab the whole
3139 message. When fetchmail uses FETCH RFC822.HEADER and FETCH
3140 RFC822.TEXT to get first the header and then the body, Domino
3141 generates different Boundary tags for each part, .e.g. one tag is
3142 declared in the Content-type header and another is used to separate
3143 the MIME parts in the body. This doesn't work. (I have heard a
3144 rumor that this bug is scheduled to be fixed in Domino release 6;
3145 you can find a workaround at contrib/domino.)</p>
3147 <p>Rob Funk explains: Unfortunately there also remain many mail
3148 user agents that don't write correct MIME messages. One big
3149 offender is Sun MailTool attachments, which are formatted enough
3150 like MIME that some programs could get confused; these are
3151 generated by the mailtool and dtmail programs (the mail programs in
3152 Sun's OpenWindows and CDE environments).</p>
3154 <p>One solution to problems related to misformatted MIME
3155 attachments is the <a
3156 href="ftp://ftp.uu.se/pub/unix/networking/mail/emil/">emil</a>
3158 href="ftp://ftp.uu.se/pub/unix/networking/mail/emil/TUTORIAL.html">tutorial</a>
3159 file at that site for details on emil. It is useful for converting
3160 character sets, attachment encodings, and attachment formats. At
3161 this writing, emil does not appear to have been maintained since a
3162 patch to version 2.1.0beta9 in late 1997, but it is still
3165 <p>One good way of using emil is from within procmail. You can have
3166 procmail look for signs of problematic message formatting, and pipe
3167 those messages through emil to be fixed. emil will not always be
3168 able to fix the problem, in which case the message is
3171 <p>A possible rule to be inserted into a .procmailrc file for using
3176 * 1^1 ^Content-Type: \/X-sun[^;]*
3177 * 1^1 ^Content-Type: \/application/mac-binhex[^;]*
3178 * 1^1 ^Content-Transfer-Encoding: \/x-binhex[^;]*
3179 * 1^1 ^Content-Transfer-Encoding: \/x-uuencode[^;]*
3181 LOG="Converting $MATCH
3184 | emil -A B -T Q -B BA -C iso-8859-1 -H Q -F MIME \
3185 | gawk '{gsub(/\r\n?/,"\n");print $0}'
3189 <p>The "1^1" in the conditions is a way of specifying to procmail
3190 that if any one of the four listed expressions is found in the
3191 message, the total condition is considered true, and the message
3192 gets passed into emil. These four subconditions check whether the
3193 message has a Sun attachment, a binhex attachment, or a uuencoded
3194 attachment; there are others that could be added to check these
3195 things better and to check other relevant conditions. The "LOG="
3196 line writes a line into the procmail log; the lone double-quote
3197 beginning the following line makes sure the log entry gets an
3198 end-of-line character. The call to gawk (GNU awk) is for fixing
3199 end-of-line conventions, since emil sometimes leaves those in the
3200 format of the originating machine; it could probably be replaced
3201 with a sed subsitution.</p>
3203 <p>The emil call itself tries to ensure that the message uses:</p>
3206 <li>BinHex encoding for any Apple Macintosh-only attachments</li>
3208 <li>Quoted-Printable encoding for text (when necessary)</li>
3210 <li>Base64 Encoding for binary attachments</li>
3212 <li>iso-8859-1 character set for text (unfortunately emil can't yet
3213 convert from windows-1252 to iso-8859-1)</li>
3215 <li>Quoted-Printable encoding for headers</li>
3217 <li>MIME attachment format</li>
3220 <p>Most of these (the primary exceptions being the character set
3221 and the Apple binary format) are as they should be for good
3222 internet interoperability.</p>
3224 <p>Some mail servers (Lotus Domino is a suspect here) mangle
3225 Sun-formatted messages, so the conversion to MIME needs to happen
3226 before such programs see the message. The ideal is to rid the world
3227 of Sun-formatted messages: don't use mailtool for sending
3228 attachments (it doesn't understand MIME anyway, and most of the
3229 world doesn't understand its attachments, so it really shouldn't be
3230 used at all), and make sure dtmail is set to use MIME rather than
3231 mailtool's format.</p>
3234 <h2><a id="X7" name="X7">X7. Some mail attachments are hanging
3237 <p>This isn't fetchmail's problem either; fetchmail doesn't know
3238 anything about mail attachments and doesn't treat them any
3239 differently from plain message data.</p>
3241 <p>The most usual cause of this problem seems to be bugs in your
3242 network transport layer's capability to handle the very large
3243 TCP/IP packets that attachments tend to turn into. You can test
3244 this theory by trying to download the offending message through a
3245 webmail account; using HTTP for the message tends to simulate
3246 large-packet stress rather well, and you will probably find that
3247 the messages that seem to be choking fetchmail will make your HTTP
3248 download speed drop to zero.</p>
3250 <p>This problem can be caused by subtle bugs in the
3251 packet-reassembly layer of your TCP/IP stack; these often don't
3252 manifest at normal packet sizes. It may also be caused by
3253 malfunctioning path-MTU discovery on the mailserver. Or, if there's
3254 a modem in the link, it may be because the attachment contains the
3255 Hayes mode escape "+++".</p>
3258 <h2><a id="X8" name="X8">X8. A spurious ) is being appended to my
3261 <p>Blame it on that rancid pile of dung and offal called Microsoft
3262 Exchange. Due to the problem described in <a href="#S2">S2</a>, the
3263 IMAP support in fetchmail cannot follow the IMAP protocol 100%.
3264 Most of the time it doesn't matter, but if you combine it with an
3265 SMTP server that behaves unusually, you'll get a spurious ) at
3268 <p>One piece of software that can trigger this is the Interchange
3269 mail server, as used by, e.g., mailandnews.com. Here's what
3272 <p>1. Someone sends mail to your account. The last line of the
3273 message contains text. So at the SMTP level, the message ends with,
3274 e.g. "blahblah\r\n.\r\n"</p>
3276 <p>2. The SMTP handler sees the final "\r\n.\r\n" and recognizes
3277 the end of the message. However, instead of doing the normal thing,
3278 which is tossing out the ".\r\n" and leaving the first '\r\n' as
3279 part of the email body, Interchange throws out the whole
3280 "\r\n.\r\n", and leaves the email body without any line terminator
3281 at the end of it. RFC821 does not forbid this, though it probably
3284 <p>3. Fetchmail, or some other IMAP client, asks for the message.
3285 IMAP returns it, but it's enclosed inside parentheses, according to
3286 the protocol. The message size in bytes is also present. Because
3287 the message doesn't end with a line terminator, the IMAP client
3295 <p>where the ')' is from IMAP.</p>
3297 <p>4. Fetchmail only deals with complete lines, and can't trust the
3298 stated message size because Microsoft Exchange fscks it up.</p>
3300 <p>5. As a result, fetchmail takes the final 'blahblah)' and puts
3301 it at the end of the message it forwards on. If you have verbosity
3302 on, you'll get a message about actual != expected.</p>
3304 <p>There is no fix for this. The nuke mentioned in <a
3305 href="#S2">S2</a> looks more tempting all the time.</p>
3308 <h2><a id="O1" name="O1">O1. The --logfile option doesn't work if
3309 the logfile doesn't exist.</a></h2>
3311 <p>This is a feature, not a bug. It's in line with normal practice
3312 for system daemons and allows you to suppress logging by removing
3313 the log, without hacking potentially fragile startup scripts. To
3314 get around it, just touch(1) the logfile before you run fetchmail
3315 (this will have no effect on the contents of the logfile if it
3316 already exists).</p>
3319 <h2><a id="O2" name="O2">O2. Every time I get a POP or IMAP message
3320 the header is dumped to all my terminal sessions.</a></h2>
3322 <p>Fetchmail uses the local sendmail to perform final delivery,
3323 which Netscape and other clients doesn't do; the announcement of
3324 new messages is done by a daemon that sendmail pokes. There should
3325 be a "biff" command to control this. Type</p>
3331 <p>to turn it off. If this doesn't work, try the command</p>
3337 <p>which is essentially what <code>biff -n</code> will do. If this
3338 doesn't work, comment out any reference to "comsat" in your
3339 /etc/inetd.conf file and restart inetd.</p>
3341 <p>In Slackware Linux distributions, the last line in /etc/profile
3354 to solve the problem system-wide.
3357 <h2><a id="O3" name="O3">O3. Does fetchmail reread its rc file
3358 every poll cycle?</a></h2>
3360 <p>No, but versions 5.2.2 and later will notice when you modify
3361 your rc file and restart, reading it.</p>
3364 <h2><a id="O4" name="O4">O4. Why do deleted messages show up again
3365 when I take a line hit while downloading?</a></h2>
3367 <p>Because you're using a POP3 other than Qualcomm qpopper, or an
3368 IMAP with a long expunge interval.</p>
3370 <p>According to the POP3 RFCs, deletes aren't actually performed
3371 until you issue the end-of-session QUIT command. Fetchmail cannot
3372 fix this, because doing it right takes cooperation from the server.
3373 There are two possible remedies:</p>
3375 <p>One is to switch to qpopper (the free POP3 server from Qualcomm,
3376 the Eudora people). The qpopper software violates the POP3 RFCs by
3377 doing an expunge (removing deleted messages) on a line hangup, as
3378 well as on processing a QUIT command.</p>
3380 <p>The other (which we recommend) is to switch to <a
3381 href="http://www.imap.org">IMAP</a>. IMAP has an explicit expunge
3382 command and fetchmail normally uses it to delete messages
3383 immediately after they are downloaded.</p>
3385 <p>If you get very unlucky, you might take a line hit in the window
3386 between the delete and the expunge. If you've set a longer expunge
3387 interval, the window gets wider. This problem should correct itself
3388 the next time you complete a successful query.</p>
3391 <h2><a id="O5" name="O5">O5. Why is fetched mail being logged with
3392 my name, not the real From address?</a></h2>
3394 <p>Because logging is done based on the address indicated by the
3395 sending SMTP's MAIL FROM, and some listeners are picky about that
3398 <p>Some SMTP listeners get upset if you try to hand them a MAIL
3399 FROM address naming a different host than the originating site for
3400 your connection. This is a feature, not a bug -- it's supposed to
3401 help prevent people from forging mail with a bogus origin site.
3402 (RFC 1123 says you shouldn't do this exclusion...)</p>
3404 <p>Since the originating site of a fetchmail delivery connection is
3405 localhost, this effectively means these picky listeners will barf
3406 on any MAIL FROM address fetchmail hands them with an @ in it!</p>
3408 <p>Versions 2.1 and up try the header From address first and fall
3409 back to the calling-user ID. So if your SMTP listener isn't picky,
3410 the log will look right.</p>
3413 <h2><a id="O6" name="O6">O6. I'm seeing long sendmail delays or
3414 hangs near the start of each poll cycle.</a></h2>
3416 <p>Sendmail does a hostname lookup when it first starts up, and
3417 also each time it gets a HELO in listener mode.</p>
3419 <p>Your resolver configuration may be causing one of these lookups
3420 to fail and time out. Check <code>/etc/resolv.conf</code> and
3421 <code>/etc/hosts</code> file. Make sure your hostname and
3422 fully-qualified domain name are both in <code>/etc/hosts</code>,
3423 and that hosts is looked at before DNS is queried. You probably
3424 also want your remote mail server(s) to be in the hosts file.</p>
3426 <p>You can suppress the startup-time lookup if need to by
3427 reconfiguring with <code>FEATURE(nodns)</code>.</p>
3429 <p>Configuring your bind library to cache DNS lookups locally may
3430 help, and is a good idea for speeding up other services as well.
3431 Switching to a faster MTA like qmail or exim might help.</p>
3434 <h2><a id="O7" name="O7">O7. Why doesn't fetchmail deliver mail in
3435 date-sorted order?</a></h2>
3437 <p>Because that's not the order the server hands it to fetchmail
3440 <p>Fetchmail getting mail from a POP server delivers mail in the
3441 order that your server delivers mail. Fetchmail can't do anything
3442 about this; it's a limitation of the underlying POP protocol.</p>
3444 <p>In theory it might be possible for fetchmail in IMAP mode to
3445 sort messages by date, but this would be in violation of two basics
3446 of fetchmail's design philosophy: (a) to be as simple and
3447 transparent a pipe as possible, and (b) to <em>hide</em>, rather
3448 than emphasize, the differences between the remote-fetch protocols
3451 <p>Re-ordering messages is a user-agent function, anyway.</p>
3454 <h2><a id="O8" name="O8">O8. I'm using pppd. Why isn't my monitor
3455 option working?</a></h2>
3457 <p>There is a combination of circumstances that can confuse
3458 fetchmail. If you have set up demand dialing with pppd, and pppd
3459 has an idle timeout, and you have lcp-echo-interval set, then the
3460 lcp-echo-interval time must be longer than the pppd idle timeout.
3461 Otherwise it is going keep increasing the packet counters that
3462 fetchmail relies upon, triggering fetchmail into polling after its
3463 own delay interval and thus preventing the pppd link from ever
3464 reaching its inactivity timeout.</p>
3467 <h2><a id="O9" name="O9">O9. Why does fetchmail keep retrieving the
3468 same messages over and over?</a></h2>
3470 <p>First, check to see that you haven't enabled the
3471 <cite>keep</cite> and <cite>fetchall</cite> option. If you have,
3472 turn <cite>keep</cite> off.</p>
3474 <p>There are various forms of lossage involving the POP3 UIDL
3475 feature that can lead to all your old messages being seen again
3476 after a line drop. I have given up trying to fix these, as the UIDL
3477 code breaks worse every time I touch it. The problem is
3478 fundamental; maintaining and garbage-collecting the right kind of
3479 client-side state is just hard. Whoever put UIDLs in RFC1725 and
3480 removed LAST should be hung up by his thumbs and whipped with
3481 scorpions. The right answers are either (a) live with the
3482 occasional breakage, (b) switch to IMAP4, or (c) fix the code
3483 yourself and send me a patch. Unless you choose (c), I don't want
3484 to hear about it.</p>
3486 <p>This can also happen when some other mail client is logged in to
3487 your mail server, if it uses a simple exclusive-locking scheme (and
3488 many, especially most POP3 servers, do exactly that). Your
3489 fetchmail is able to retrieve the messages, but because the mailbox
3490 is write-locked by the other instance yours can neither mark
3491 messages seen or delete them. The solution is to either (a) wait
3492 for the other client to finish, or (b) terminate it.</p>
3494 <p>James Stevens <James.Stevens at kyzo.com> writes:</p>
3496 <p><em>We had a Linux box dialing the Net and collecting mail from
3497 an NT POP3 server. Fetchmail was correctly collecting and deleting
3498 each e-mail one by one. However,the dial-up connection was very
3499 unreliable and would often just drop out in the middle of a
3502 <p><em>Interestingly, unless the TCP POP3 connection was terminated
3503 normally (I guess with a POP3 "QUIT" command) NT would then roll
3504 back all the deletes !!!</em></p>
3506 <p><em>This meant if the first e-mail was very large it might just
3507 end up continuously collecting it, basically jamming the queue. Or,
3508 if the queue became very full itmight never get a long enough phone
3509 connection to retrieve the entire mailbox, and NT would roll back
3510 any deletes, so it would end up collecting (and delivering) the
3511 first few e-mails again and again. As the POP3 mailbox became
3512 fuller and fuller the chances of getting a connection long enough
3513 to collect theentire mailbox became smaller and smaller.</em></p>
3515 <p><em>Our solution was to make fetchmail only collect a few (say 5
3516 or 10) e-mails at atime, thus trying to ensure that the POP3
3517 connection is terminated correctly.</em></p>
3519 <p>Unfortunately, this is exactly the way POP3 servers are supposed
3520 to behave on a line drop, according to the RFCs. I recommend
3521 switching to IMAP and using a short expunge interval.</p>
3524 <h2><a id="O10" name="O10">O10. Why is the received date on all my
3525 messages the same?</a></h2>
3527 <p>This is a design choice in your MTA, not fetchmail. It's taking
3528 the received date from the last Received header.</p>
3531 <h2><a name="O11">O11. I keep getting messages that say "Repoll
3532 immediately" in my logs.</a></h2>
3534 <p>This is your server barfing on the CAPA probe that fetchmail sends.</p>
3536 <p>If you run fetchmail in daemon mode (say "set daemon 600"), you will
3537 get the message only once per run.</p>
3539 <p>If you set an authentication method explicitly (say, with
3540 <code>auth password</code>), you will never get the message.</p>
3543 <h2><a name="O12">O12. Fetchmail no longer expunges mail on a 451 SMTP response.</a></h2>
3545 <p>This is a feature, not a bug.</p>
3547 <p>Any 4xx response (like 451) indicates a transient (temporary) error.
3548 This means that the mail could be accepted if retried later. Lookup
3549 failures are normally transient errors as a mail should not get
3550 rejected if a dns server is unreachable or down.</p>
3552 <p>A permanent reject response is of the form 5xx (like 550).</p>
3554 <p>You could tell your SMTP server to not lookup any addresses if you are
3555 not keen on checking the sender addresses. This problem typically
3556 occurs if your mail server is not checking the sender addresses, but
3557 your local server is.</p>
3559 <p>Or you could declare <code>antispam 451</code>.</p>
3561 <p>Or, you could check your nameserver configuration and query logs for
3564 <p>All these issues are not related to fetchmail directly.</p>
3567 <h2><a name="O13">O13. I want timestamp information in my fetchmail logs.</a></h2>
3569 <p>Write a <code>preconnect</code> command in your configuration file that
3570 does something like "date >> $HOME/Procmail/fetchmail.log".</p>
3573 <h2><a name="O14">O14. Fetchmail no longer deletes oversized mails with
3576 <p>Use <code>--limitflush</code> to delete oversized mails along with
3577 the <code>--limit</code> option. If you are already having
3578 <code>flush</code> in your rcfile to delete oversized mails,
3579 <em>replace</em> it with <code>limitflush</code> to avoid losing mails
3580 unintentionally.</p>
3582 <p>The <code>--flush</code> option is primarily designed to delete
3583 mails which have been read/downloaded but not deleted yet. This option
3584 cannot be overloaded to delete oversized mails as it cannot be guessed
3585 whether the user wants to delete only read/downloaded mails or only
3586 oversized mails or both when a user specifies both
3587 <code>--limit</code> and <code>--flush</code>. Hence, a separate
3588 <code>--limitflush</code> has been added to resolve the ambiguity.</p>
3591 <table width="100%" cellpadding="0" summary="Canned page footer">
3593 <td width="30%">Back to <a href="index.html">Fetchmail Home
3595 <td width="30%" align="right">$Date$</td>
3600 <address>Eric S. Raymond <a
3601 href="mailto:esr@thyrsus.com"><esr@thyrsus.com></a><br />
3602 Matthias Andree</address>