- dovecot - dovecot.org

[Dovecot] deliver: Panic: file sieve-cmu.c: line 90 (unfold_header): assertion failed: (str[i] == ' ' || str[i] == '\t')
by Anton Yuzhaninov 19 Feb '09

19 Feb '09

I use dovecot-1.1.10 and dovecot-sieve-1.1.6 /usr/local/libexec/dovecot/deliver coredups several times per day with bt like this: (gdb) bt #0 0x0000000800a3a91c in kill () at kill.S:2 #1 0x0000000800a3a88a in __raise (s=6) at /usr/src/lib/libc/gen/raise.c:46 #2 0x0000000800a38f83 in abort () at /usr/src/lib/libc/stdlib/abort.c:65 #3 0x00000000004a7de5 in default_fatal_finish (type=LOG_TYPE_PANIC, status=0) at failures.c:151 #4 0x00000000004a8777 in i_syslog_fatal_handler (type=Could not find the frame base for "i_syslog_fatal_handler". ) at failures.c:309 #5 0x00000000004a80ca in i_panic (format=Could not find the frame base for "i_panic". ) at failures.c:198 #6 0x0000000800e0b265 in unfold_header ( str=0x800c55a40 "\"firm ru\nÐ\217Ð\201Ð±ÐŒÂ¬Â® Ñ\236Ò\220Ð\216Â¬Â Ð±Ð²Ò\220Ð°Ð³\" <our_email@ example.ru>") at sieve-cmu.c:90 #7 0x0000000800e0b381 in unfold_multiline_headers (headers=0x800c55a30) at sieve-cmu.c:119 #8 0x0000000800e0b406 in getheader (v=0x7fffffffeab0, phead=0x800c5a160 "to", body=0x7fffffffc048) at sieve-cmu.c:132 #9 0x0000000800e1affc in eval_bc_test (interp=0x800c2d380, body_cache=0x0, m=0x7fffffffeab0, bc=0x800c5a000, ip=0x7fffffffc300) at bc_eval.c:541 #10 0x0000000800e1ac96 in eval_bc_test (interp=0x800c2d380, body_cache=0x0, m=0x7fffffffeab0, bc=0x800c5a000, ip=0x7fffffffd8e4) at bc_eval.c:445 #11 0x0000000800e1c131 in sieve_eval_bc (exe=0x800c240b0, is_incl=0, i=0x800c2d380, body_cache=0x0, sc=0x7fffffffeb60, m=0x7fffffffeab0, imapflags=0x7fffffffda20, actions=0x800c660f0, notify_list=0x800c13a40, errmsg=0x7fffffffda38) at bc_eval.c:1009 #12 0x0000000800e23ece in sieve_execute_bytecode (exe=0x800c240b0, interp=0x800c2d380, script_context=0x7fffffffeb60, message_context=0x7fffffffeab0) at script.c:868 #13 0x0000000800e0dee4 in cmu_sieve_run (namespaces=0x800c0b848, storage_r=0x7fffffffec98, mail=0x800c1d848, script_path=0x800c02d28 "/home/public/.dovecot.sieve", destaddr=0x7fffffffef3e "user_xxxxx", username=0x800c1ce35 "user_xxxxx", mailbox=0x4c581c "INBOX") at sieve-cmu.c:1038 #14 0x0000000800e0ae6d in cmusieve_deliver_mail (namespaces=0x800c0b848, storage_r=0x7fffffffec98, mail=0x800c1d848, destaddr=0x7fffffffef3e "user_xxxxx", mailbox=0x4c581c "INBOX") at cmusieve-plugin.c:81 #15 0x0000000000417d7f in main (argc=3, argv=0x7fffffffedc8) at deliver.c:1084 I log messages: deliver(user_xxxxx): Panic: file sieve-cmu.c: line 90 (unfold_header): assertion failed: (str[i] == ' ' || str[i] == '\t') -- Anton Yuzhaninov

2 2

[Dovecot] Effects of going read-only on dovecot & pop/imap clients
by Mark Zealey 19 Feb '09

19 Feb '09

Hi all, We need to set our filestore to read-only for quite a while, and I want to mitigate the effects for our clients. First issue; dovecot. If we just change the nfs mounts to read-only, dovecot complains: 2009-02-19T14:03:19+00:00 mail2 dovecot: pop3-login: Login: user=<test2(a)markandruth.co.uk>, method=PLAIN, rip=10.15.10.95, lip=10.15.10.102:11002 2009-02-19T14:03:19+00:00 mail2 dovecot: POP3(test2(a)markandruth.co.uk): open() failed with file /var/spool/mail/supplier_2/ma/rk/an/markandruth.co.uk/te/st/2_/test2/Mai ldir/dovecot.index.log: Read-only file system 2009-02-19T14:03:19+00:00 mail2 dovecot: POP3(test2(a)markandruth.co.uk): open(/var/spool/mail/supplier_2/ma/rk/an/markandruth.co.uk/te/st/2_/test 2/Maildir/dovecot-uidlist.lock) failed: Read-only file system 2009-02-19T14:03:19+00:00 mail2 dovecot: POP3(test2(a)markandruth.co.uk): file_dotlock_create(/var/spool/mail/supplier_2/ma/rk/an/markandruth.co.u k/te/st/2_/test2/Maildir/dovecot-uidlist) failed: Read-only file system 2009-02-19T14:03:19+00:00 mail2 dovecot: POP3(test2(a)markandruth.co.uk): Couldn't init INBOX: Internal error occurred. Refer to server log for more information. [2009-02-19 14:03:19] 2009-02-19T14:03:19+00:00 mail2 dovecot: POP3(test2(a)markandruth.co.uk): Mailbox init failed bytes=0/95, top=0/0, retr=0/0, del=0/0, size=0 Even when I use :INDEX=MEMORY, it's still complaining about creating .lock's or uidlist files and just issuing internal errors. Is there any way to make dovecot just do as much as it can under the circumstances ie allow read access to mails? Also, has anyone got experience of how clients handle mailboxes being read-only? My guess is that pop clients would probably break and download the same messages over and over? Thanks, Mark -- Mark Zealey -- Shared Hosting Team Leader Product Development * Webfusion 123-reg.co.uk, webfusion.co.uk, donhost.co.uk, supanames.co.uk This mail is subject to http://www.gxn.net/disclaimer

2 3

[Dovecot] UNSUBSCRIBE
by Eric B. Schorvitz, Ph.D. 19 Feb '09

19 Feb '09

2 1

[Dovecot] limiting authentication failures
by Eric B. Schorvitz, Ph.D. 19 Feb '09

19 Feb '09

In my log files I occasionally get a huge number of Dovecot authentication failures (see clip below). I wanted to know if there's a way to limit the number of times an IP address can attempt to authenticate, if there's a way to have a timeout between attempted authentications, or if there is a way to limit authentication attempts by a specific username within a certain period of time. My current solution is to permanently block the specific IP, an IP range, or an entire country from accessing my server AFTER I notice the huge number of authentication failures. This is too ad-hoc a process and was hoping dovecot has something more proactive built in. Thank you in advance for spending time considering this inquiry, Eric --------------------- pam_unix Begin ------------------------ dovecot: Authentication Failures: rhost=::ffff:200.111.39.219 : 764 Time(s) root: 25 Time(s) mysql: 6 Time(s) smmsp: 6 Time(s) --SNIP-- Unknown Entries: check pass; user unknown: 764 Time(s) ---------------------- pam_unix End ------------------------- --------------------- Connections (secure-log) Begin ------------------------ **Unmatched Entries** dovecot-auth: pam_succeed_if(dovecot:auth): error retrieving information about user info dovecot-auth: pam_succeed_if(dovecot:auth): error retrieving information about user info dovecot-auth: pam_succeed_if(dovecot:auth): error retrieving information about user info dovecot-auth: pam_succeed_if(dovecot:auth): error retrieving information about user info --SNIP--

2 1

[Dovecot] sieve filters
by Ian P. Christian 19 Feb '09

19 Feb '09

Hi all, I'm trying to figure out a way of filtering mail from a commercial anti-spam platform. Unfortunately you need to compare part of one header against part of another header. I figured I could solve this using sieve variables and pattern matching, however I'm having little luck and don't know how to debug sieve issues. # example headers: must compare 0.00799 against 1:0000 (the 2nd number in the 2nd header) # X-pstn-levels: (S: 0.00799/98.42547 CV:99.9000 FC:95.5390 LC:95.5390 R:95.9108 P:95.9108 M:97.0282 C:98.6951 ) # X-pstn-settings: 3 (1.0000:1.0000) s cv gt3 gt2 gt1 r p m c if header :matches "X-pstn-levels" "(S: */*" { set "score" "${1}"; if header :matches "X-pstn-settings" " ? (*:*) *" { fileinto "somewhere"; } } The first if appears to match, as if I move the fileinto up there it works - however it doesn't work as it shown here. Can anyone suggest why? Can I achieve what I'm trying to do with sieve? How can I go about debugging scripts? Thanks, Ian

2 2

[Dovecot] kerberos trying to obtain credentials for wrong machine
by Nikolay Shopik 19 Feb '09

19 Feb '09

Hi, I'm currently trying to configure Dovecot to use kerberos. My KDC is Windows 2003 and I successful generated keytab file for Dovecot machine. Problem is when I'm trying to use GSSAPI it told me Obtaining credentials for imap@debian5 - and of course this fails because debian5 isn't KDC, it should look for imap/debian5.inblock.local(a)INBLOCK.LOCAL. What I'm missing?

2 1

[Dovecot] Time moved backwards ....
by Harry Lachanas 18 Feb '09

18 Feb '09

OK.. So I synced the clock.... and got .... dovecot: Time just moved backwards by 1 seconds. I'll sleep now until we're back in present. http://wiki.dovecot.org/TimeMovedBackwards ( The first time I did this the clock moved backwards 2 hours after a timezone change and dovecot suicided ) I think I understand the concept ... However a mail server should probably be synchronized to the local time ..... Suggestions ...??? Harry.

8 10

[Dovecot] mbox snarf plugin + idle
by Jonathan Siegle 18 Feb '09

18 Feb '09

I'm having a problem with mbox snarf not looking at /var/spool/mail/ when in idle mode thus never giving me a RECENT line even though there are new messages in /var/spool/mail/ . Here are the imap commands to reproduce the problem: 1 login userid password 2 select inbox 3 idle When I run "select inbox" it does see my messages in /var/spool/mail/ and moves them over fine. When I truss the process, I see it only running stat calls on my "mbox- snarf" file. To get new messages I issue DONE, CLOSE, and SELECT INBOX. I'm not sure why it is reporting alpha5 two lines below. I did an hg pull just the other day and see 1.2.beta1 in the output of hg tags. # /usr/ladmin2/sbin/dovecot -n # 1.2.alpha5: /usr/ladmin2/etc/dovecot.conf Warning: fd limit 2000 is lower than what Dovecot can use under full load (more than 4224). Either grow the limit or change login_max_processes_count and max_mail_processes settings # OS: AIX 3 0001112AD300 syslog_facility: local0 protocols: imap listen: *:2222 ssl: no disable_plaintext_auth: no login_dir: /usr/ladmin2/var/run/dovecot/login login_executable: /usr/ladmin2/libexec/dovecot/imap-login login_greeting: Dovecot ready. login_processes_count: 30 max_mail_processes: 4096 mail_location: mbox:%h/new:INBOX=/var/spool/mail/14/%u mmap_disable: yes dotlock_use_excl: no mbox_write_locks: fcntl mbox_lazy_writes: no mail_plugins: mbox_snarf imap_client_workarounds: delay-newmail auth default: mechanisms: plain gssapi krb5_keytab: /etc/krb5/dovecot.keytab gssapi_hostname: $ALL verbose: yes debug: yes passdb: driver: pam userdb: driver: passwd plugin: mbox_snarf: /gpfs/inbox/14/%u Thanks, Jonathan

2 2

Re: [Dovecot] from postfix deliver to dovecot deliver
by dovecot＠corwyn.net 18 Feb '09

18 Feb '09

you're using maildir? (I only ask because I am). In that case, no making the change doesn't require changes in the mailboxes. I made the appropriate changes (just a few days ago) to main.cf and I was good to go. Rick At 08:01 AM 2/18/2009, Johan Hendriks wrote: >Hello all. >I use dovecot in combination with postfix, postfixadmin for virtual users. >Deliver to the mailbox now is done by postfix >Now i want to enable quota's and therefor i need dovecot deliver. >Can i just switch to dovecot deliver? >Or must i first do some prrperations in the mailboxes. > >regards >-- >_______________________ >*Johan Hendriks* >*Schavemaker Transport* > >Tel: +31 (0)251 229098 >Fax: +31 (0)251 212016 >email: j.hendriks(a)schavemaker.com <mailto:j.hendriks@schavemaker.com> >web: http://www.schavemaker.com >_______________________ >/Confidentiality Notice: The information in this document may be >confidential. It is intended only for the use of the named recipient. >If you are not the intended recipient, please notify me immediately >and then delete this document. Do not disclose the contents of this document >to any other person, nor take any copies. Violation of this notice >may be unlawful. / >_______________________

1 0

[Dovecot] assertion failed
by Ian P. Christian 18 Feb '09

18 Feb '09

Does anyone know what might cause this? # dovecot --version 1.1.7 Feb 17 18:21:51 imap-proxy-temp dovecot: Panic: auth(default): file passdb-cache.c: line 121 (passdb_cache_lookup_credentials): assertion failed: (*scheme_r != NULL || *password_r == NULL) Feb 17 18:21:51 imap-proxy-temp dovecot: auth(default): Raw backtrace: dovecot-auth [0x42c0ea] -> dovecot-auth [0x42c143] -> dovecot-auth [0x42b7f6] -> dovecot-auth [0x41acb6] -> dovecot-auth(auth_request_lookup_credentials+0x6f) [0x411cdf] -> dovecot-auth [0x416caf] -> dovecot-auth(auth_request_handler_auth_continue+0xc9) [0x4128e9] -> dovecot-auth [0x40e40a] -> dovecot-auth(io_loop_handler_run+0xf8) [0x42fb48] -> dovecot-auth(io_loop_run+0x1d) [0x42ea4d] -> dovecot-auth(main+0x2f7) [0x416367] -> /lib/libc.so.6(__libc_start_main+0xda) [0x7f83990e54ca] -> dovecot-auth [0x40cdba] Feb 17 18:21:51 imap-proxy-temp dovecot: child 20579 (auth) killed with signal 6

2 4

[Dovecot] Maildir hierarchy sep bug ?
by Thomas Hummel 17 Feb '09

17 Feb '09

Hello Timo, I'm experiencing the following different behaviors with dovecot-1.1.8/Maildir. My Maildir hierarchy separator is the "." (dot) character. My namespaces, when I use them, separator is "/". When I run the server without any namespaces, I can create mailboxes named something like first.name which, of course, are shown to the client as first/ (non selectionnable) first/name For instance in Mutt, if I save a message (the "s" key) from first.name(a)some.domain dovecot happilly creates the .first.name/ Maildir But if I create a public Namespace, then I have to create a private default namespace as well and the server doesn't behave the same way : They're set up like this : namespace private { separator = / prefix = location = maildir:/courriel/boites/%u:CONTROL=/courriel/meta/%u:INDEX=/var/dovecot/indexes/%1u/%u inbox = yes hidden = no list = yes subscriptions = yes } namespace public { separator = / prefix = '#Foo/' location = maildir:/courriel/boites/public/foo inbox = no hidden = yes list = no subscriptions = no } Then users cannot create mailboxes with a dot in it : Character not allowed in mailbox name: '.' which seems normal but can be confusing since, a user used to saving his mail in first.name (in Mutt for instance) wouldn't be able to do so anymore if the server goes from the first to the second setting described above. Granted, a user which does so, probably doesn't do what he thinks he does (he probably thinks he save his message in a folder named "first.name" while instead he actually saves it in a subfolder of "first" named "name"). Still it can be confusing. Is that difference of behavior normal ? Which is the correct behavior ? And is there still no way (except to patch the source) to allow dots in mailbox names in Maildir ? Thanks. -- Thomas Hummel | Institut Pasteur <hummel(a)pasteur.fr> | Pôle informatique - systèmes et réseau

2 7

[Dovecot] Error: IMAP(user): nfs_flush_file_handle_cache_dir: rmdir(/var/mail) failed: Device busy
by Grandy Fu 17 Feb '09

17 Feb '09

Hi, I am running dovecot 1.1.10 and 1.1.11 on Solaris 10, mailboxes are in mbox format and served by Solaris 10 NFSv3, the dovecot cache is on local disk. Some of my users always had above nfs error and their mailbox usually very big, over 100M, 400M, or even 800M. Grandy

2 1

[Dovecot] dbox redesign
by Timo Sirainen 17 Feb '09

17 Feb '09

This is about how to implement multiple msgs/file dbox format. The current v1.1's one msg/file design would stay pretty much the same and it would be compatible with this new design. dbox directories with multiple msgs/file would be like: ~/dbox/storage/ has the actual mail data for all mailboxes ~/dbox/mailboxes/ has subdirectories containing mailboxes and their indexes Also since dbox supports already the single msg per file, those files would be stored in the mailboxes/ directory. So the idea would be that either you use multiple msgs per file using a global storage, or you use single msg per file without a global storage (or it's also possible to be in a mixed setup with some mails in storage/ and some in mailboxes/, mainly to allow migration between those configurations). The storage/ directory would have a new "map index" which is a regular dovecot index (dovecot.index and dovecot.index.log). So the mailbox index would point to mails using an intermediary "map UID". This way if mails are moved to another file only the map index needs to be updated. GUID would be a globally unique 128 bit ID for messages. So if map indexes get corrupted for any reason it's possible to rebuild it by finding the mails using GUIDs. v1.1 dbox has this "dbox.index" file which I was originally planning on using with multiple msgs/file. It had complex file range locking stuff. Now I'm thinking that it's pretty much useless. The only reason for its existence with the new design is for listing metadata for files converted from Maildir. Map index record would contain: - 32 bit map UID - 8 bit flags (MAIL_DELETED flag = message marked as expunged) - 8 bit unused wasted space - 16 bit refcount - 32 bit file sequence - 32 bit file offset --> total 128 bits/msg Mailbox index: - IMAP UID, flags, keywords, etc. - 32 bit map UID - 128 bit GUID dbox file metadata: - 128 bit GUID - size, vsize, received time, saved time, etc. - initial mailbox name (if all indexes get trashed, we can still figure out at least one mailbox where to put the mail. copies would get lost though.) (- no map UID, no imap UID) How to save a message with multiple msgs/file: 1. Find dbox file where to append to: 1.1. Look up the last message from map index 1.2. Is the file "too old"? (or doesn't exist at all) - Yes -> Create new dbox file 1.3. Is the file "too large"? - Yes -> Look at the previous file (one sequence less) and goto 1.2. 1.4. Try to lock the file. - Fail -> Look at prev file and goto 1.2. Now we have a locked/new dbox file where we can write to. Because 1.4. step only tries to lock the file, there's no waiting on locks. This also means that if e.g. two processes are writing new messages rapidly they may be appending actively to two different files. I don't think that's a problem, better than waiting for locks. 2a) We're using an existing file and we need to find the append offset. Since we found the file by finding the last msg in the file, we also know the last message's offset. I wasn't really planning on saving the message sizes in the index file, so to get the append offset I guess it needs to do an extra read on the last msg's header to find the size and skip over it. Hmm. Or would it be less disk I/O to store the size on the index so it could be found directly? I'm not really sure.. In any case, after we find the append offset, check to see if it's at EOF. If not, that means that either another process just saved a new message there or a process crashed previously and left garbage lying around. Refresh the map index to see if this file+offset exists in it. If not, truncate the file and just continue writing there. If it exists, figure out the new append offset and see again if the file limit would be reached. If the file would become too large, unlock the file and goto step 1. 2b) We're writing to a new file. No need to worry about anything in 2a) 3. Write the message and its metadata to dbox file (including generated 128 bit GUID). 4. Assign map UIDs for the written mails and write APPEND records to map index's transaction log. The record would contain the map UID, file seq, offset, refcount=1. The transaction is saved with a "weak" flag (wonder if there's a better name for this) and its offset is remembered. - If we're creating a new dbox file, it's assigned the file seq and rename()d to the final file name while the map index is locked. 5. Write APPEND record to mailbox index's transaction log with IMAP UID, map UID and GUID (and flags, keywords, etc). 6. Write "commit offset=x" record where x is the offset remembered in step 4. This marks the 4's weak transaction as being fully finished. 7. dbox file is unlocked (if we weren't creating a new file). When reading the index and we see a weak transaction without a commit record, call a resolve() function in dbox code. It finds the dbox file in the weak transaction and tries to lock it. If it can't lock it, it (probably) means that there's still a process that's going to finish the save. If a locking succeeded, run a fixup code that goes through all files that are referenced in weak transactions and makes sure all the refcounts and such are valid. Then it writes commit records for each such transaction. Copying can be done by locking source dbox file, writing a weak "refcount++" transaction and continuing from saving step 5. The only step I can think of where there's a small problem is if data is written to dbox file, then process dies, but no saving is attempted to the same file later. The data is left there as garbage until maybe some day later when a message is expunged from that file and cleanup code notices that there's extra data. I thought about reordering steps so that the records are written to map index first, but that's less efficient when writing multiple messages, because each one would need to be saved separately to the map index (where each save would be lock + fstat (+ read) + write + unlock). This is probably a rare problem which doesn't really matter.. Expunging: The "expunge" and "cleanup to remove deleted space" are separate steps. Preferably the cleanup would be done when the server is "less busy", but I guess it could also be configured to happen either immediately or during logout (or once an hour or so). 1. Write "want to expunge" record to mailbox's transaction log. 2. Write a weak "refcount--" to map index's transaction log. 3. Write "expunged" record to mailbox's transaction log. 4. Write "commit" for the weak refcount--. (The two-stage want-to-expunge(1)/expunged(3) is already done for maildir/mbox also.) I was also thinking maybe this could be just 3 steps without a weak transaction. Especially since without a fsync() between 2 and 3 there's really no guarantee about the order in which 2 and 3 happen. And fsync()s are kind of annoying. But then again if that isn't done and it crashes between 2 and 3 the expunge is tried again later which now decrements the refcount again. If the message's original refcount was 1 it would now be negative. That's not really a problem. More problematic is if the message had been copied to another mailbox and refcount=2, now the refcount would be 0 and the message in the other mailbox would get unexpectedly lost. Not good. Space deletion cleanup: 1. Check from map index header to see there are any refcount=0 mails. Although having this step probably requires some annoying dbox-specific code in index handling code. I guess I'll have to see later if there's a non-horribly-ugly way to do this. 2. Go through all records in the map index. Get a list of files that have refcount=0 records. Also make a copy of file seq+offsets and sort them (so at step 3 we can easily find what mails exist). 3. For each file that has refcount=0 records: 3.1. Try to lock the file. -> Fail: Skip to next file. Either someone's appending to it or cleaning it up already. 3.2. Refresh map index. Add any newly added records to the sorted records list (rare). 3.2. Find the first refcount=0 mail. 3.3. Copy all mails after that with refcount>0 to a new file. 3.4. Lock map index, assign new file seq for the new file, rename() it, for each mail we just copied, using a weak transaction write the seq +offset updates to map index's transaction log. Unlock map index. 3.5. Depending on if first refcount=0 mail was the first mail, either unlink() the file or ftruncate() it. 3.6. Write "commit" record to finish the weak transaction. That's a bit inefficient when cleaning up multiple dbox files, but I'm not sure if there's a much better way unless the map index is locked for the entire time. Since the copying might take a while, a long map index lock is bad since it prevents messages from being saved. Reading mails can be done without locking. But since a file can be truncated unexpectedly, the code must be able to find the new file and offset in that situation and continue reading from there. If the message we were reading was just expunged+cleaned up, the client will have to be just disconnected. Since the cleanup shouldn't normally happen immediately after expunge, this should happen rarely/never. Oh and about all these new temp files that are created. When dbox is opened the dir is stat()ed and if its atime is older than 1 week (currently hardcoded) it readdir()s through the dir to see if there are any old temp* files and then unlinks them. With the new design flags/keywords would never be written to actual dbox files. This helps with backups and also avoids the horribleness of having to increase metadata area. To avoid complete data loss due to index corruption (which really shouldn't be happening nowadays), the indexes are also once in a while backed up to e.g. dovecot.index.backup file. If Dovecot detects missing/broken index, it uses this backup copy to rebuild the index as well as it can. I'm also wondering if it's better for each mailbox to have its separate dovecot.index.cache file or if there should be one cache file for the map index. The upside with one index is that if a mail is copied there's no need to duplicate its cached data. The downside is that I'm worried it would cause more disk I/O, since mailboxes are typically accessed separately, so it might be better that the cache data for one mailbox is stored together rather than all around one bigger cache file.

5 7

[Dovecot] Changing level an error logs at
by Thomas M Goerger 17 Feb '09

17 Feb '09

Hi, I'd like to change the log levels of a few of the errors that Dovecot throws. Our servers throw errors at the console above the 'Error' level, and there are a few that I'd like to change to 'Warn' so that they don't do so. Has anyone out there attempted this and can give me a few pointers? Or is there a doc somewhere about how to do this? Thanks, ************************************************************************* * Tom Goerger - Email/Unix System Administrator * * * * University of Minnesota Email: tmg(a)umn.edu * * Operations, Infrastructure and Architecture Phone: 4-5804 * * Internet Services Office: 626J WBOB * * * *************************************************************************

3 3

[Dovecot] Replication status
by Asheesh Laroia 16 Feb '09

16 Feb '09

I remember last year we discussed Dovecot replication a bit on this list, e.g. http://www.dovecot.org/list/dovecot/2008-May/030446.html . Has there been any development related to this, and can I help at all? The more I think about the hacks I'm employing to synchronize mail, the more I realize I want this. So hopefully I can put either some money or some time where my mouth is. -- Asheesh. -- Its name is Public Opinion. It is held in reverence. It settles everything. Some think it is the voice of God. -- Mark Twain

7 10

[Dovecot] Access denied when client is connecting
by Florian Wörter 16 Feb '09

16 Feb '09

Hello! I'm having a problem with dovecot on my gentoo linux box. it would be great if you could help me. i'm planing to use dovecot as imap server. the incoming mails for the different useres are fetched by getmail. for every user a script should be called, which fetches the e-mails of this user from a pop3 server and puts them in ~/mailbox/new. this allready works. now i installed dovecot but everytime a client connects, i find these lines in my dovecot logfile (/var/log/dovecot.log) dovecot: Feb 16 09:08:09 Fatal: chdir(/home/woefl81) failed with uid 1001: Permission denied dovecot: Feb 16 09:08:09 Error: child 16211 (imap) returned error 89 (Fatal failure) the username of the connected user is woefl81, his user id is 1001 and his home directory is /home/woefl81/. i can't figure out why this user has no permission for his own files. well, for testing i now adjusted the directory permissions: in /: drwxr-xr-x 5 root root 152 Feb 10 09:19 home in /home: drwxr-xr-x 6 woefl81 users 296 Feb 11 08:27 woefl81 in /home/woefl81: drwxrwxrwx 5 woefl81 users 120 Feb 9 13:09 mailbox in /home/woefl81/mailbox: drwxrwxrwx 5 woefl81 users 120 Feb 9 13:09 . drwxr-xr-x 6 woefl81 users 296 Feb 11 08:27 .. drwxrwxrwx 2 woefl81 users 48 Feb 9 13:09 cur drwxrwxrwx 2 woefl81 users 120 Feb 9 13:13 new drwxrwxrwx 2 woefl81 users 48 Feb 9 13:13 tmp and finally in /home/woefl81/mailbox/new i find the email which was retrieved by getmail: drwxrwxrwx 2 woefl81 users 120 Feb 9 13:13 . drwxrwxrwx 5 woefl81 users 120 Feb 9 13:09 .. -rwxrwxrwx 1 woefl81 users 4.6K Feb 9 13:13 1234181629.M605602P32438Q0R5b6a645fa35278a9.proxystu as i said before i adjusted the rights for the mailbox subdir for testing to 0777, so everyone really should be able to do everything... my dovecot version is 1.1.11 my config: # 1.1.11: /etc/dovecot/dovecot.conf # OS: Linux 2.6.27-gentoo-r7 i686 Gentoo Base System release 1.12.11.1 log_path: /var/log/dovecot.log info_log_path: /var/log/dovecot_info.log protocols: imap ssl_cert_file: /etc/ssl/dovecot/server.pem ssl_key_file: /etc/ssl/dovecot/server.key disable_plaintext_auth: no login_dir: /var/run/dovecot/login login_executable: /usr/libexec/dovecot/imap-login login_chroot: no verbose_proctitle: yes mail_location: maildir:~/mailbox auth default: passdb: driver: pam args: * userdb: driver: passwd the user woefl81 is a "normal" system user (from /etc/passwd): woefl81:x:1001:100::/home/woefl81:/bin/bash his main group is users. on the client i installed mozilla tunderbird.. the user connects to the ip of the linux box with his username (woefl81) and the same password as his account pwd on the linux box. it would be great if somebody had an idea what else i should try to get dovecot to work as imap server. thank you very much in advance! best regards, Florian

2 1

[Dovecot] LDAP, MD5-CRYPT, invalid credentials, BUG or config issue?
by Guenther Falk 16 Feb '09

16 Feb '09

Hi, I read a lot of howto's and I got problems with LDAP and Dovecot to work together. I'm using: Gentoo Linux 2008.0 hardened Dovecot 1.1.7 Kernel 2.6.26 OpenLDAP 2.3.43 My dovecot-ldap.conf is: uris = ldaps://auth.mydomain.com:636 auth_bind = yes auth_bind_userdn = uid=%u,ou=People,dc=mydomain,dc=com ldap_version = 3 base = ou=People,dc=mydomain,dc=com deref = never scope = subtree user_attrs = homeDirectory=home,uidNumber=uid,gidNumber=gid user_filter = (&(objectClass=posixAccount)(uid=%u)) pass_attrs = uid=user,userPassword=password pass_filter = (&(objectClass=posixAccount)(uid=%u)) default_pass_scheme = CRYPT I also tried default_pass_scheme = MD5-CRYPT In the logs I find smething like this: Feb 16 12:20:49 mail dovecot: Dovecot v1.1.7 starting up Feb 16 12:20:50 mail dovecot: auth(default): new auth connection: pid=30582 Feb 16 12:20:50 mail dovecot: auth(default): new auth connection: pid=30583 Feb 16 12:20:50 mail dovecot: auth(default): new auth connection: pid=30584 Feb 16 12:20:53 mail dovecot: auth(default): new auth connection: pid=30585 Feb 16 12:20:58 mail dovecot: auth(default): client in: AUTH 1 PLAIN service=imap secured lip=172.30.0.10 rip=172.30.0.254 lport=993 rport=51269 Feb 16 12:20:58 mail dovecot: auth(default): client out: CONT 1 Feb 16 12:20:58 mail dovecot: auth(default): client in: CONT 1 AGd1ZW50aGVyADE*********= Feb 16 12:20:58 mail dovecot: auth(default): ldap(guenther,172.30.0.254): invalid credentials Feb 16 12:21:00 mail dovecot: auth(default): client out: FAIL 1 user=guenther Feb 16 12:21:00 mail dovecot: auth(default): client in: AUTH 2 PLAIN service=imap secured lip=172.30.0.10 rip=172.30.0.254 lport=993 rport=51269 resp=AGd1ZW50aGVyADE*********= Feb 16 12:21:00 mail dovecot: auth(default): ldap(guenther,172.30.0.254): invalid credentials Feb 16 12:21:02 mail dovecot: auth(default): client out: FAIL 2 user=guenther It seems dovecot ignores the CRYPT password scheme. The password seems to be encrypted in SSHA. What am I doing wrong please?? I am sure, the password is correct as I can login into the machine via nss_ldap and pam_ldap. Regards Günther

2 1

[Dovecot] LDAP, MD5-CRYPT, invalid credentials, BUG or config issue?
by guenther 16 Feb '09

16 Feb '09

Hi, I read a lot of howto's and I got problems with LDAP and Dovecot to work together. I'm using: Gentoo Linux 2008.0 hardened Dovecot 1.1.7 Kernel 2.6.26 OpenLDAP 2.3.43 My dovecot-ldap.conf is: uris = ldaps://auth.mydomain.com:636 auth_bind = yes auth_bind_userdn = uid=%u,ou=People,dc=mydomain,dc=com ldap_version = 3 base = ou=People,dc=mydomain,dc=com deref = never scope = subtree user_attrs = homeDirectory=home,uidNumber=uid,gidNumber=gid user_filter = (&(objectClass=posixAccount)(uid=%u)) pass_attrs = uid=user,userPassword=password pass_filter = (&(objectClass=posixAccount)(uid=%u)) default_pass_scheme = CRYPT I also tried default_pass_scheme = MD5-CRYPT In the logs I find smething like this: Feb 16 12:20:49 mail dovecot: Dovecot v1.1.7 starting up Feb 16 12:20:50 mail dovecot: auth(default): new auth connection: pid=30582 Feb 16 12:20:50 mail dovecot: auth(default): new auth connection: pid=30583 Feb 16 12:20:50 mail dovecot: auth(default): new auth connection: pid=30584 Feb 16 12:20:53 mail dovecot: auth(default): new auth connection: pid=30585 Feb 16 12:20:58 mail dovecot: auth(default): client in: AUTH 1 PLAIN service=imap secured lip=172.30.0.10 rip=172.30.0.254 lport=993 rport=51269 Feb 16 12:20:58 mail dovecot: auth(default): client out: CONT 1 Feb 16 12:20:58 mail dovecot: auth(default): client in: CONT 1 AGd1ZW50aGVyADE*********= Feb 16 12:20:58 mail dovecot: auth(default): ldap(guenther,172.30.0.254): invalid credentials Feb 16 12:21:00 mail dovecot: auth(default): client out: FAIL 1 user=guenther Feb 16 12:21:00 mail dovecot: auth(default): client in: AUTH 2 PLAIN service=imap secured lip=172.30.0.10 rip=172.30.0.254 lport=993 rport=51269 resp=AGd1ZW50aGVyADE*********= Feb 16 12:21:00 mail dovecot: auth(default): ldap(guenther,172.30.0.254): invalid credentials Feb 16 12:21:02 mail dovecot: auth(default): client out: FAIL 2 user=guenther It seems dovecot ignores the CRYPT password scheme. The password seems to be encrypted in SSHA. What am I doing wrong please?? I am sure, the password is correct as I can login into the machine via nss_ldap and pam_ldap. Regards Günther

1 0

[Dovecot] quota warning issue
by Alex 16 Feb '09

16 Feb '09

Hello to all, I'm using version 1.1.11 of dovecot. I set the quota using the following configuration: in /etc/dovecot/dovecot.conf quota = maildir:%u quota_warning = storage=75%% /root/quota-warning.sh 75 %u in /etc/dovecot/dovecot-sql.conf user_query = SELECT concat('/var/vmail/', maildir) as home, concat('maildir:/var/vmail/', maildir) as mail, 1000 AS uid, 12 AS gid, concat('*:storage=', ROUND(quota/1000)) AS quota_rule FROM mailbox WHERE username = '%u' AND active = '1' ## password_query = SELECT username as user, password, concat('/var/vmail/', maildir) as home, concat('maildir:/var/vmail/', maildir) as mail, 1000 AS uid, 12 AS gid, concat('*:storage=', ROUND(quota/1000)) AS quota_rule FROM mailbox WHERE username = '%u' AND active = '1' # 1.1.11: /etc/dovecot/dovecot.conf # OS: Linux 2.6.25-hardened-r11 x86_64 Gentoo base_dir: /var/run/dovecot/ log_path: /var/log/mail.log info_log_path: /var/log/mail.log protocols: imap imaps pop3 pop3s listen(default): *:143 listen(imap): *:143 listen(pop3): *:110 ssl_listen(default): *:993 ssl_listen(imap): *:993 ssl_listen(pop3): *:995 disable_plaintext_auth: no login_dir: /var/run/dovecot//login login_executable(default): /usr/libexec/dovecot/imap-login login_executable(imap): /usr/libexec/dovecot/imap-login login_executable(pop3): /usr/libexec/dovecot/pop3-login first_valid_uid: 1000 last_valid_uid: 1000 mail_location: maildir:/var/vmail/%d/%u mail_debug: yes mail_executable(default): /usr/libexec/dovecot/imap mail_executable(imap): /usr/libexec/dovecot/imap mail_executable(pop3): /usr/libexec/dovecot/pop3 mail_plugins(default): quota imap_quota mail_plugins(imap): quota imap_quota mail_plugins(pop3): quota mail_plugin_dir(default): /usr/lib64/dovecot/imap mail_plugin_dir(imap): /usr/lib64/dovecot/imap mail_plugin_dir(pop3): /usr/lib64/dovecot/pop3 imap_client_workarounds(default): outlook-idle delay-newmail imap_client_workarounds(imap): outlook-idle delay-newmail imap_client_workarounds(pop3): pop3_uidl_format(default): %08Xu%08Xv pop3_uidl_format(imap): %08Xu%08Xv pop3_uidl_format(pop3): %v.%u pop3_client_workarounds(default): pop3_client_workarounds(imap): pop3_client_workarounds(pop3): outlook-no-nuls oe-ns-eoh auth default: user: nobody verbose: yes passdb: driver: sql args: /etc/dovecot/dovecot-sql.conf userdb: driver: sql args: /etc/dovecot/dovecot-sql.conf userdb: driver: prefetch socket: type: listen client: path: /var/spool/postfix/private/auth mode: 432 user: postfix group: mail master: path: /var/run/dovecot/auth-master mode: 384 user: vmail group: mail plugin: quota: maildir:%u quota_warning: storage=75%% /root/quota-warning.sh 75 %u sieve: /var/vmail/%d/%n/.dovecot.sieve The warning is not delivered to users. In my logs the user quota's percentage is calculated right dovecot: Feb 15 23:42:33 Info: IMAP(alex(a)foo.org): Loading modules from directory: /usr/lib64/dovecot/imap dovecot: Feb 15 23:42:33 Info: IMAP(alex(a)foo.org): Module loaded: /usr/lib64/dovecot/imap/lib10_quota_plugin.so dovecot: Feb 15 23:42:33 Info: IMAP(alex(a)foo.org): Module loaded: /usr/lib64/dovecot/imap/lib11_imap_quota_plugin.so dovecot: Feb 15 23:42:33 Info: IMAP(alex(a)foo.org): Effective uid=1000, gid=12, home=/var/vmail/foo.org/alex/ dovecot: Feb 15 23:42:33 Info: IMAP(alex(a)foo.org): Quota root: name=alex(a)foo.org backend=maildir args= dovecot: Feb 15 23:42:33 Info: IMAP(alex(a)foo.org): Quota rule: root=alex(a)foo.org mailbox=* bytes=52428800 messages=0 dovecot: Feb 15 23:42:33 Info: imap-login: Login: user=<alex(a)foo.org>, method=PLAIN, rip=127.0.0.1, lip=127.0.0.1, secured dovecot: Feb 15 23:42:33 Info: IMAP(alex(a)foo.org): Quota warning: bytes=39321600 (75%) messages=0 command=/root/quota-warning.sh 75 alex(a)foo.org but also overcoming the space defined on quota warning, the command is not performed. The command does not contain errors because if performed by cli correctly sends the alert. Any ideas? Regards. Alex

3 7

[Dovecot] unkillable imap process(es) with high CPU-usage
by Arno Wald 15 Feb '09

15 Feb '09

Hello, I am having a problem with my dovecot-daemon. It is forking one or more (I saw up to perhaps 8 of them) imap processes under my user name. These processes are consuming a lot of CPU time and are not killable: > PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND > 8616 arno 20 0 2900 1600 1204 R 98 0.2 1196:38 imap Stopping dovecot does not quit these processes. Killing them (even "kill -9" as root) is not possible. The only solution to get rid of them is to reboot. I have found a mailing list post of one who seemed to have the same problem. He solved it by upgrading to version 1.1. But in my case this did not help. Is this a known problem? What could I check or do? My system: very up to date debian/sid (sidux) kernel 2.6.27-8.slh.1-sidux-686 CPU: Intel(R) Core(TM)2 CPU 4300 @ 1.80GHz Filesystem: ext3fs tried dovecot from sid: 1:1.0.15-2.3 tried dovecot from experimental: 1:1.1.2-3 My dovecot.conf is the original debian configuration with only one line changed into: protocols = imaps # dovecot -n # 1.1.2: /etc/dovecot/dovecot.conf log_timestamp: %Y-%m-%d %H:%M:%S protocols: imaps login_dir: /var/run/dovecot/login login_executable: /usr/lib/dovecot/imap-login mail_privileged_group: mail auth default: passdb: driver: pam userdb: driver: passwd I am using dovecot locally on my system with me as the only user. As client I am using thunderbird alias icedove 2.0.0.17-1. Icedove retrieves the Mails from another imap server and sorts them into Maildir-folders in dovecot. I do not know when the imap-process are going mad. It happens after the system (and the mail-client) is up for a while. Do you need more information? Thanks, Arno

10 35

Re: [Dovecot] unkillable imap process(es) with high CPU-usage
by Robert Schetterer 15 Feb '09

15 Feb '09

just again to the list wasnt in reply Robert Schetterer schrieb: > Hi Gene , small tips depending suse > > Gene Steinberg schrieb: >> On Feb 14, 2009, at 9:57 AM, agent59624285 wrote: >> >>> I have this EXACT same problem after upgrading to SuSE 11.1, which >>> uses this >>> exact kernel version!! >>> >>> After reading this, I was excited to think that if I killed the nfsserver >>> daemon (which I had running for no good reason), that it would sort my >>> problem.... >>> >>> Sure enough, my computer - which up to now had been going unresponsive >>> every >>> 24 hours - was running fine for 72 hours and then BOOM... it happened >>> again. >>> >>> Just wanted to let people know that it seems that at the minute, the >>> dovecot >>> that ships with SuSE and the kernel they are using in 11.1 exhibit this >>> problem. ---my reply > > Suse kernels are not like vanilla kernels, > they are deeply patched an mostly have stuff from the next kernel > version allready in > you might use yast for upgrading latest suse kernel it might fixed allready > a short look gives > http://download.opensuse.org/repositories/home:/tristanhoffmann/openSUSE_11… > gives dovecot 1.1.11 which is latest > also loadable by yast or just download rpm and upgrade wtih rpm -u > > standart rep for mail stuff is > http://download.opensuse.org/repositories/server:/mail/openSUSE_11.1/ > but here dovecot is not at latest > but it shouldnt be any big problem downloading src and > prepare the spec with latest src dovecot > and compile by yourself > which is the best way and gives you the suse splitted packs > too and the advantage to see compile stuff > for sure you need some others libs etc cause of dependencies > befor but they are mostly downloadable also by yast > > be aware that suse sometimes splits up funktions > in own packs like > dovecot-autocreate11, dovecot-antispam11 > etc this musnt be the same by versions compile from home suse users > > also help full send a mail to the build service, > normally it takes only days and the newest version gets compiled > > > use http://software.opensuse.org/search > if looking for newer versions of software ----- > >>> Gino >>> >> This is sounding similar to the problem I have with my setup: >> >> 1. High CPU usage. >> >> 2. Can't kill IMAP. >> >> 3. Server becomes unresponsive. >> >> I'm using CentOS 5.2 64-bit version with the latest cPanel. >> >> So what am I missing, other than the problem nobody else is having is >> clearly something they ARE having? >> >> Peace, >> Gene > > -- Best Regards MfG Robert Schetterer Germany/Munich/Bavaria

1 0

[Dovecot] digest-md5 problem
by Nigel Kukard 15 Feb '09

15 Feb '09

Info: v1.1.11 OS = Linux 32bit, custom build CPU Arch = i686 Filesystem = NFSv4 Problem: Digest-md5 not working. Tested plain, login and cram-md5 with pop3, imap and smtp/sasl fine. Using digest-md5 doesn't authenticate with the below error.... *** Feb 15 12:46:45 s03-400-134 dovecot: auth(default): client in: AUTH^I1^IDIGEST-MD5^Iservice=pop3^Ilip=10.0.2.134^Irip=10.0.2.1^Ilport=110^Irport=41579^Iresp=<hidden> Feb 15 12:46:45 s03-400-134 dovecot: auth(default): client out: CONT^I1^IcmVhbG09IiIsbm9uY2U9IktvMkFoK0lyeVhHRWtmY1k5TjM1TXc9PSIscW9wPSJhdXRoIixjaGFyc2V0PSJ1dGYtOCIsYWxnb3JpdGhtPSJtZDUtc2VzcyI= Feb 15 12:46:45 s03-400-134 dovecot: auth(default): client in: CONT<hidden> Feb 15 12:46:47 s03-400-134 dovecot: auth(default): client out: FAIL^I1 Feb 15 12:46:47 s03-400-134 dovecot: pop3-login: Disconnected (auth failed, 1 attempts): method=DIGEST-MD5, rip=10.0.2.1, lip=10.0.2.134 *** The decoded reply is ... authzid="USERNAME_HERE",charset=utf-8,cnonce="32d669a1334164b3362bff5f809acf12",digest-uri="pop/SERVER_URL_HERE",nc=00000001,nonce="rDWjRuMzuvbr5TLsinsE0A==",qop=auth,realm="",response=5c7c2ff3284f350fc62c5a2a80d2be64,username="USERNAME_HERE" On the same database cyrus imap & sasl is working 100%. All passwords are in plaintext format. Maybe I"m doing something wrong? Below my dovecot config... # 1.1.11: /etc/dovecot/dovecot.conf # OS: Linux 2.6.27.10 i686 none protocols: imap imaps pop3 pop3s listen: 10.0.2.134 shutdown_clients: no login_dir: /var/run/dovecot/login login_executable(default): /usr/libexec/dovecot/imap-login login_executable(imap): /usr/libexec/dovecot/imap-login login_executable(pop3): /usr/libexec/dovecot/pop3-login login_greeting: s03-400-134 Secure server ready. max_mail_processes: 100 verbose_proctitle: yes first_valid_uid: 1000 first_valid_gid: 1000 mail_location: maildir:/var/vmail/%h mail_debug: yes mmap_disable: yes mail_nfs_storage: yes mail_nfs_index: yes mail_executable(default): /usr/libexec/dovecot/imap mail_executable(imap): /usr/libexec/dovecot/imap mail_executable(pop3): /usr/libexec/dovecot/pop3 mail_plugins(default): quota imap_quota trash mail_plugins(imap): quota imap_quota trash mail_plugins(pop3): mail_plugin_dir(default): /usr/lib/dovecot/imap mail_plugin_dir(imap): /usr/lib/dovecot/imap mail_plugin_dir(pop3): /usr/lib/dovecot/pop3 imap_client_workarounds(default): delay-newmail outlook-idle netscape-eoh imap_client_workarounds(imap): delay-newmail outlook-idle netscape-eoh imap_client_workarounds(pop3): pop3_enable_last(default): no pop3_enable_last(imap): no pop3_enable_last(pop3): yes pop3_lock_session(default): no pop3_lock_session(imap): no pop3_lock_session(pop3): yes pop3_client_workarounds(default): pop3_client_workarounds(imap): pop3_client_workarounds(pop3): outlook-no-nuls oe-ns-eoh namespace: type: private separator: . prefix: INBOX. inbox: yes list: yes subscriptions: yes auth default: mechanisms: digest-md5 cram-md5 login plain cache_size: 256 user: nobody master_user_separator: * verbose: yes debug: yes passdb: driver: sql args: /etc/dovecot/sql.conf userdb: driver: sql args: /etc/dovecot/sql.conf socket: type: listen client: path: /var/spool/postfix/private/auth mode: 432 user: dovecot group: mail master: path: /var/run/dovecot/auth-master mode: 432 user: dovecot group: vmail plugin: quota: maildir trash: /etc/dovecot/trash.conf

1 0

[Dovecot] great disappearing email mystery
by dhottinger＠harrisonburg.k12.va.us 14 Feb '09

14 Feb '09

Have there been any issues with dovecot and using outlook express (imap) as an email client? I have had a couple of users come up with random missing emails. Im trying to figure out if it is user error, or something wacky in my mailserver. I cant find anything telling in maillog files or my messages. Dovecot version is 1.0.3. Its ok to reply to me, I get list messages digest. thanks, ddh -- Dwayne Hottinger Network Administrator Harrisonburg City Public Schools "Everything should be made as simple as possible, but not simpler." -- Albert Einstein "The hottest places in Hell are reserved for those who, in times of moral crisis, preserved their neutrality." -- Dante

10 25

Re: [Dovecot] pop3_lock_session question
by Alan Ferrency 14 Feb '09

14 Feb '09

Rob Mangiafico <rmang(a)lexiconn.com> wrote: >On Wed, 4 Feb 2009, Timo Sirainen wrote: >> On Wed, 2009-02-04 at 11:17 -0700, Mark Costlow wrote: >>> Hello, I'm preparing to convert from qpopper + UW-IMAP to dovecot. >>> So far testing has gone very well. One problem we haven't figured >>> out is that long-running POP sessions keep the mailbox locked, so that >>> the MDA times out while trying to deliver. >> .. >> >> Switch to Maildir and the problem goes away. > > We see this as well with mbox and pop3 accesses where some pop3 clients do > not logout (iphone's are the worst offenders) for 30-90 minutes it seems. > Timeout settings in dovecot.conf do not seem to help. procmail backs up > waiting to get access to the inbox to deliver mail. We are having the same problem. It's not a problem with imap, only with pop3. We're using deliver as the MDA. We're also using flock() locking for the mbox: is that what everyone else is using as well? Migrating our servers to maildir globally is also not an option for us. In some cases, even convincing customers to use IMAP isn't going to work. One problem which might be making this worse than it needs to be, is the fact that mbox_lock_flock in mbox-lock.c is not using a blocking flock(); instead, it's polling for a non-blocking lock. This technique can cause lock starvation, if another process is dropping the lock and picking it back up again frequently: other processes will only see the lock as being available if they happen to poll for the lock at just the right instant. A better technique to use here, if it's adequately cross-platform, would be to set an alarm() for the max_wait_time, and use a blocking flock(). If the alarm times out and you don't have a lock, it's a timeout. In the meantime, you're guaranteed to eventually get the lock, if it is dropped. That said: I'm not sure whether this will solve our problem in practice. Why doesn't this happen with imap? Why can't we make pop3 do what imap does? Even if it's inefficient, it's better than hanging all incoming mail delivery while deliver eats up our local concurrency limits. Thanks! Alan Ferrency pair Networks, Inc. alan(a)pair.com

11 33

[Dovecot] Expire failing with error Remote Disconnected
by dovecot＠corwyn.net 14 Feb '09

14 Feb '09

I'm trying to follow the instructions from: http://wiki.dovecot.org/Plugins/Expire for the BDB backend. Everything is set up as indicated, but when I run: dovecot --exec-mail ext /usr/libexec/dovecot/expire-tool I get back: Error: read(/var/run/dovecot/dict-server) failed: Remote disconnected I'm sure it's something simple.... Thx rick dovecot -n: # 1.1.8: /etc/dovecot.conf # OS: Linux 2.6.18-92.el5 i686 Red Hat Enterprise Linux Server release 5.2 (Tikanga) ext3 ssl_cert_file: /etc/httpd/certs/_com.crt ssl_key_file: /etc/httpd/certs/.com.key.no.password login_dir: /var/run/dovecot/login login_executable(default): /usr/libexec/dovecot/imap-login login_executable(imap): /usr/libexec/dovecot/imap-login login_executable(pop3): /usr/libexec/dovecot/pop3-login mail_uid: 1015 mail_gid: 105 mail_location: maildir:/var/spool/mail/%d/%n/Maildir mail_executable(default): /usr/libexec/dovecot/imap mail_executable(imap): /usr/libexec/dovecot/imap mail_executable(pop3): /usr/libexec/dovecot/pop3 mail_plugins(default): quota imap_quota expire trash mail_plugins(imap): quota imap_quota expire trash mail_plugins(pop3): quota mail_plugin_dir(default): /usr/lib/dovecot/imap mail_plugin_dir(imap): /usr/lib/dovecot/imap mail_plugin_dir(pop3): /usr/lib/dovecot/pop3 imap_client_workarounds(default): delay-newmail outlook-idle imap_client_workarounds(imap): delay-newmail outlook-idle imap_client_workarounds(pop3): pop3_client_workarounds(default): pop3_client_workarounds(imap): pop3_client_workarounds(pop3): outlook-no-nuls oe-ns-eoh auth default: mechanisms: plain login passdb: driver: sql args: /etc/dovecot-sql.conf userdb: driver: passwd userdb: driver: sql args: /etc/dovecot-sql.conf userdb: driver: static socket: type: listen client: path: /var/spool/postfix/private/auth mode: 432 user: postfix group: postfix master: path: /var/run/dovecot/auth-master mode: 384 user: vuser group: vuser plugin: quota: maildir quota_rule: *:storage=102400000 quota_rule2: Trash:storage=100M trash: /etc/dovecot-trash.conf expire: Trash 14 Trash/* 14 Spam 14 expire_dict: proxy::expire dict: expire: db:/var/lib/dovecot/expire.db Rick Steeves http://www.sinister.net "The journey is the destination"

4 9