- dovecot - dovecot.org

[Dovecot] Indexes
by Daniel L. Miller 13 Jan '08

13 Jan '08

When something "bad" happens to the indexes, my e-mail client (Thunderbird) reports an "unable to succeed" error on opening a mailbox. Leaving that mailbox and coming back works fine. Is this expected behaviour? -- Daniel

3 6

[Dovecot] Custom 'quota exceeded' message + Sieve script from database
by Przemyslaw Wegrzyn 13 Jan '08

13 Jan '08

Hi! I'm quite new to Dovecot. helping my friend to migrate from Courier-IMAP to something more efficient, we are evaluating Dovecot at the moment. I had to prepare 2 small patches for Dovecot to suit our needs, and I've decided to post it here. I'm not fully familiar with Dovecot architecture yet, I'd appreciate if someone could look if these changes make sense (i.e. doesn't break anything). And of course, maybe someone is interested as well. 1. Customized "quota exceeded" message. http://czajnick.sitenet.pl/dovecot/dovecot_1.0.9_quota_exceeded_message.diff - quick hack to let me define custom message instead of typical "Quota exceeded", simply by using plugins { quota_exceeded_message = ... }. Browsing this mailing list I've seen some posts about such feature being implemented in Dovecot. Is it done in a similar way? Which stable version is the first to include this functionality? 2. Sieve filter modification http://czajnick.sitenet.pl/dovecot/dovecot_sieve_1.0.2_envscript_patch.diff In our systems we currently use a modified maildrop MDA that picks the filter script from the database instead of the file. This has proven useful - end-user's administration panel doesn't need any special permissions (or networking capabilities), as it doesn't need to access the filesystem - it is reduced to a plain database web-app. The patch above makes the delivery agent check for the sieve script present in SIEVE_SCRIPT environment variable. If this is the case, env var content is used instead of the script stored in a usual file pointed to by SIEVE variable. This makes it trivial to pass the script from the database using base dovecot functionality. A minor drawback of such solution is that the script is re-compiled on each delivery, but I don't think it takes much time to compile a few lines of a script using (f)lex/bison-or-yacc parser. Yes, I've noticed the sieve-manage patch, but I don't want the complexity of managing sieve script over the socket connection (if my understanding of the script is wrong, please correct me). BR, Przemyslaw

2 2

[Dovecot] do not lose mail when dovecot is dead
by Daniel 13 Jan '08

13 Jan '08

Hi! Recently I've experienced problems with dovecot. Dovecot died silently, what means is that there are no core files, nor any indication in the log files, just the "kevent(): Invalid argument" messages. Dovecot just died, so deliver could not connect to the auth-master socket, which means that messages couldn't been delivered. The problem is that postfix thought that the delivery was successful, and didn't place the message to delay, but deliver(LDA) couldn't deliver the message, so it got lost. An example: postfix/pickup[29619]: 8A1C6F261: uid=1001 from=<id> postfix/cleanup[19148]: 8A1C6F261: message-id=<20080107192402.8A1C6F261(a)domain.hu> postfix/qmgr[29441]: 8A1C6F261: from=<id(a)domain.hu>, size=585, nrcpt=1 (queue active) deliver(leva): Loading modules from directory: /usr/local/lib/dovecot/lda deliver(leva): Module loaded: /usr/local/lib/dovecot/lda/lib10_quota_plugin.so deliver(leva): Module loaded: /usr/local/lib/dovecot/lda/lib90_cmusieve_plugin.so deliver(leva): Can't connect to auth server at /var/run/dovecot/auth-master: Connection refused postfix/pipe[8582]: 8A1C6F261: to=<leva(a)domain.hu>, orig_to=<id>, relay=dovecot-lda, delay=1.7, delays=1.1/0/0/0 .68, dsn=2.0.0, status=sent (delivered via dovecot-lda service) postfix/qmgr[29441]: 8A1C6F261: removed See? Postfix removed it from the queue, so basically all bets are off, and everything depends on deliver, but it just dropped the message. Is there anything that can be done regarding this issue? Thanks! Daniel

6 10

[Dovecot] randomly disconnected with imap, and latest hg
by Daniel 12 Jan '08

12 Jan '08

Hi! I'm using latest hg from the 1.1 branch, and noticed that since yesterday's updates, dovecot randomly disconnects clients when using imap, with these message in the log: "Disconnected for inactivity in reading our output bytes=153/41319" This happens randomly. Eg. I'm fetching messages, and suddenly, dovecot disconnects. Or my MUA just sits there, and then again, dovecot disconnects me with this message. I didn't even do anything with the client, so I can't think of anything which could've been read. What could be the problem? Daniel

2 6

[Dovecot] Please help me resolve why mail isn't being delivered to virtual users
by Andrew Falanga 12 Jan '08

12 Jan '08

Hi, I have to admit that I'm not quite sure why no one has responded to me about this one. If I'm not providing enough information or incorrect information please tell me. It's quite important that this e-mail system be working. I do not quite understand what it is that I'm missing now. I have the need to setup and configure various users as virtual users. To this end, I've configured them so that their home directories are "/virtusers/<userid>" and their INBOXes are "/var/mail/<userid>." I have created a system user ID called "vmail" which is referenced in my PostgreSQL database for UID/GID stuff. All virtual user home directories and INBOXes are made owner "vmail" and group is "users." Originally, I had my sendmail "/etc/mail/aliases" file setup to "map" all mail sent to these virtual users to the system user "vmail." However, none of the virtual users were getting any mail and vmail was just getting swamped. After doing more digging, I found that I just didn't understand the home directories and INBOX, mbox and such. That's what drove me to configuring what's in there now. Now, following directions on this WIKI page: http://wiki.dovecot.org/LDA/Sendmail?highlight=%28deliver%29 I'm still unable to get the e-mail working for them. Continually, sendmail bails with errors saying it, "Can't create output." Because I am using *.mc files, I made a new *.m4 file as directed in /usr/share/sendmail/cf/mailer. Dovecot does allow the virtual users to login. In fact, when I did, the first time, there appears a "mail" directory in their home directories after login. This would indicate to me that the login was successful. Further, I cut/pasted a message sent to one of the virtual users from the vmail INBOX (/var/mail/vmail) into the virtual user's INBOX (/var/mail/jdunkin). Then, logged in as this virtual user and ACTUALLY got the mail. I know this stuff works. What must be done to make sendmail do its thing for deliver? What is necessary to fix this situation? Andy

12 42

[Dovecot] Allow_nets
by Evaggelos Balaskas 11 Jan '08

11 Jan '08

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I use the allow_nets password extra field [0] for my users. Is there a way to use this functionality for ALL users, and not to edit my passwd-file every time a new user is added ? The alternative i am working for this is the TCP Wrappers. [0]: http://wiki.dovecot.org/PasswordDatabase/ExtraFields/AllowNets -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.8 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkd7ZzIACgkQWIK+Pe9twhrRDQCgjui9zyG67n6kA5I1Lf+0efvg cvIAoMe7+gniwfbZdFGh8UERtuw7IG0K =KKQk -----END PGP SIGNATURE-----

2 2

[Dovecot] Dovecot-imap, tls, gnus
by Richard G Riley 11 Jan '08

11 Jan '08

I have a connection to my dovecot server from gnus: ,---- | (add-to-list 'gnus-secondary-select-methods | '(nnimap "hermes" | (nnimap-stream starttls) | (nnimap-address "hermes") | (nnimap-authinfo-file "~/.authinfo"))) `---- If I enter a group and read my mail then wander off and come back after a few minutes I can not navigate out of the group because my nnimap session has ended. What config parameter do I need to change in Dovecot to keep it open for the life of the client? (I am assuming its a dovecot issue here since it didn't happen with courier-imap).

3 5

[Dovecot] Problem: "Missing IMAP reply key: F"
by Claus Atzenbeck 11 Jan '08

11 Jan '08

Hi all: I installed dovecot v1.0 (on Mac OS X) and want to connect to is on localhost via Pine or Alpine, but it fails. Pine's debug mode tells me the following: ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Debug output of the Pine program (debug=2 debug_imap=4). Version 4.64 (OSX) [...] Terminal type: screen About to open folder "INBOX" inbox is: "INBOX" IMAP 14:40:22 1/11 mm_log babble: Trying IP address [::1] IMAP 14:40:22 1/11 mm_log warning: Connection failed to localhost,143: Connection refused IMAP 14:40:22 1/11 mm_log babble: Trying IP address [127.0.0.1] IMAP DEBUG 14:40:22 1/11: F IMAP 14:40:22 1/11 mm_notify warning: {localhost}inbox: Missing IMAP reply key: F IMAP 14:40:22 1/11 mm_notify bye: {localhost}inbox: [CLOSED] IMAP connection broken (server response) IMAP 14:40:22 1/11 mm_log error: [CLOSED] IMAP connection broken (server response) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ I use the options "disable_plaintext_auth = no" and "ssl_disable = yes", because I don't need encryption. (I only access the server from localhost.) What does "Missing IMAP reply key: F" mean? Any easy way to solve the problem? Thanks for any hint! Claus

2 1

[Dovecot] Automatically subscribe to folder?
by Christian Flothmann 11 Jan '08

11 Jan '08

Hello everybody, i am filtering e-mails using the cmusieve-plugin. A global script filters all mails tagged by SpamAssasin and moves them into the folder Spam: if header :is "X-Spam-Flag" "YES" { fileinto "Spam"; } My problem is that users who did not had a folder Spam until the first time of filtering a spam mail are not automatically subscribed to this new created folder. They have to do this step manually in their mail client. Is it possible to automate this script and subscribe the account to this folder when it is created by the sieve script? Regards, Christian

7 14

[Dovecot] Test Environment Question
by Stewart Dean 11 Jan '08

11 Jan '08

I have my master IMAP server running DC V1.0.10. The homedir and INBOXdir are physically resident there and NFS exported (no caching) to 3 other machines. I have installed V1.1beta13 on one of them (which thus accesses the homedir/INBOXdir remotely) and plan to have a limited community test-drive it there. Are there any hazards or drawbacks in doing this? While the homedirs and INBOXdirs are thus shared, I have it so that each machine has its own local index directory and /var/run dir. Comments or dire warnings?

4 3