similar to: buglet in dovecot.conf, 1.0alpha1

Help... Dovecot 1.0alpha1 hung up today with nothing working and the following spewing to my syslog: pipe() failed: Too many open files My setup: Solaris 9, just using imap, mbox format. The various "max" settings in dovecot.conf use the defaults. Suggestions on what to change in fovecot.conf to keep this from happening? Jeff Earickson Colby College

Hi, I have seen the following in my syslogs for 1.0alpha1: Corrupted index cache file [filename]: record points outside file and : indexid changed : record list is circular : Cache record offset 9772 points outside file : invalid record size Are these errors minor? Major? Self-healing? Should the cache files be blown away to fix these errors? Jeff Earickson Colby College

Help! I just noticed this error this morning, we have been getting a lot of these in rc28 and rc29, not in rc27 or before: Apr 6 07:22:52 karst dovecot: [ID 107833 mail.error] IMAP(jaearick): mbox file /var/mail/j/jaearick was modified while we were syncing, check your locking settings Pine coughed up an internal error, no core dump, and dovecot basically doubled the size of my mailbox by

Hi, I am stuck at rc13 because rc14 and rc15 have both failed to keep up with imap connections on my (slow old) system. I have the problem where if a lot of people check their email at once (like at the top of the hour) then my load goes way up and things slow to a crawl. My setup: Solaris 10 on an E220R, mbox format, imap and imaps only. I also use "login_process_per_connection =

Hi, I just upgraded from test80 to 1.0alpha1. I wanted to try something new on the lock_method, so I set: lock_method = fcntl, dotlock I got syslog complaints like: IMAP(username): Unknown lock_method: fcntl, dotlock Can one specify multiple file locking mechanisms? If so, what is the syntax (and can the syntax be shown in the comments of the .conf file)? Jeff Earickson Colby College

Hi, Today's core dump from 1.0alpha1 came from a syslog message of: IMAP(user): pool_data_stack_realloc(): stack frame changed gdb info on the resulting core dump attached. Question: how many people are building/using dovecot 1.0alpha1 with gcc 4.0.1 versus gcc 3.4.x? I am wondering if these issues come from the compiler instead of dovecot itself? Jeff Earickson Colby College

Timo, I see where at least one other person reported this, but here goes. I went from rc24 to rc25 this morning, and I got an assert and core from my own mailbox withing five minutes: Mar 2 06:52:26 karst dovecot: [ID 107833 mail.error] IMAP(jaearick): file mbox-sync-rewrite.c: line 408: assertion failed: (need_space == (uoff_t)-mails[idx].space) Mar 2 06:52:26 karst dovecot: [ID 107833

Gang, For those who asked about making a Solaris system a Kerberos client to Active Directory, the magic document to have is: http://www.connectathon.org/seam1.0/files/c0101.htm See the section "How to Configure a SEAM Client Using a Windows 2000 KDC". SEAM was Sun's Kerberos client stuff in Solaris 8; it is just there as part of Solaris 9 and 10. These instructions apply if

Help, new to dovecot... I want to get some output from my test sessions to confirm that SSL pop and imap are working. How does one get debug output from dovecot? How does one see that dovecot is using SSL in a transaction? I configured dovecot 0.99.13 with: CC=gcc CPPFLAGS=-I/opt/openssl/include LDFLAGS=-L/opt/openssl/lib \ ./configure --prefix=/opt/dovecot --with-ssl=openssl \

Hi, After searching gmane.mail.imap.dovecot on the topic, and wiki.dovecot.org, I'm still clueless on how to use RFC 2087 quotas in dovecot with the quota plugin. How come no wiki doc? My setup: Solaris 9, mbox format. Both the filesystem for INBOXes (/var/mail) and the user's home directories have filesystem-based quotas imposed. For example: % quota -v joeblow Disk quotas for

Timo, Output of gdb session on the core file attached. This one only produced the following syslog: IMAP(user): UIDs broken with partial sync in mbox file /var/mail/r/user with no assert. Setup: Solaris 9, gcc 4.0.1 for dovecot build. Jeff Earickson Colby College -------------- next part -------------- Script started on Thu Sep 08 10:05:53 2005 %gdb imap core.rtohara GNU gdb 6.3 Copyright

Timo, Attached is gdb information from core dumps related to the following assert in test-80: IMAP(username): file mbox-sync-update.c: line 442 (mbox_sync_update_header_from): assertion failed: (ctx->mail.uid == 0 || ctx->mail.uid_broken || ctx->mail.uid == mail->uid) My setup: Solaris 9, mbox format. test-80 compiled with gcc 4.0.1 using the following configure options: CC=gcc

Hi, My setup: Solaris 9, /var/mail is NFS mounted from another Solaris 9 box, home directories and dovecot index stuff is on local ufs filesystems. I upgraded from alpha4 to a5 about 36 hours ago. Dovecot is configured like so during the build: VERSION=1.0.alpha5 CC=gcc CFLAGS="-g -O" CPPFLAGS=-I/opt/openssl/include LDFLAGS=-L/opt/openssl/lib \ ./configure

If its too hard for them to setup the pop3 catch all, we can probably put some mapping in place to smtp froward mail to their server. This is more complex on our part so would prefer not to do it that way, but its an option and will require not setup on their part other than make sure their server does not accept mail for anyone other than our server. dovecot-request@dovecot.org wrote: >

Hi, I've been lurking on the list since Spring, planning to replace UW-IMAP with dovecot 1.0 sometime this summer. The recent thread on "1.0" has not given me a good feeling about seeing it before summer's end. So I wonder if I should use dovecot-1.0-test-xx code or dovecot-stable-[date] code (which seems to change every day, how stable is that???). This is confusing,

Help! Our students are back and they are pounding my dovecot server to death (ie "Login process died too early - shutting down" or "too many open files" type crash). I've had my third crash today. My setup: Solaris 10, imap and imaps, 3K users coming in via horde webmail. My current connection settings are: login_max_processes_count = 2048 login_max_connections = 4096

Hi, I find login_process_per_user referred to in the comments in dovecot.conf, but never set. Is this a typo and should it be login_process_per_connection or what? Thanks, Rob

Timo, After installing beta8 on Solaris 10 sparc and starting it for the first time, I noticed this in my syslog: May 14 19:35:45 emerald dovecot: [ID 475045 mail.info] Dovecot v1.0.beta8 starting up May 14 19:35:47 emerald dovecot: [ID 536776 mail.error] stat(~) failed: No such file or directory The only appearance of the tilde (~) in my dovecot.conf is here: default_mail_env =

Timo, This might be a bit vague, but I have noticed that rc19 seems to have broken the read/unread/reply flags in imap. My officemate has had problems with messages that are read and marked as read then mysteriously get remarked as unread the next time email is checked. I've had problems where I reply to a message and it does not get marked as "answered". We didn't see this

Hi, Could you say a few words about Dovecot and POP/IMAP to an NFS mounted filesystem? I thought I saw on the dovecot webpage that NFS is "no way" with 0.99 and "maybe" with 1.x. I was looking for the FAQ and couldn't find it. My setup is like so: My mail server is a Solaris 9 system, running sendmail/procmail, and qpopper POP. Mail is stored in mbox format. The