similar to: quick question about fs quota overhead in plugin

In the past I dabbled with the imap quota plugin with the fs backend because I wanted to report usage to my users (not limit them). At the time, the quota plugin would make dovecot crash when trying to write to a folder (I can bring up this report if needed). However, in a later beta of dovecot 1.1 I tried quota again but I cannot get it to report any results. I've been wanting to use

I noticed the following this week. No idea how long its been happening because I have not tested deleting folders hardly at all. Steps to reproduce: 1. Create a folder "NewFolder", click on it in thunderbird so an index dir is created: ls -ldi Maildir/.NewFolder/ 1504062 drwx------ 5 mcdouga9 egrstaff 4096 Nov 16 19:44 Maildir/.NewFolder/ ls -ldi

Early next week I need to upload over 100,000 emails to an IMAP server as quickly as possible from an Outlook client. I am looking for any methods I can use to (temporarily?) speed up the rate at which dovecot can accept and store IMAP uploads, whether it be storing on local disk, ram disk, etc. I can setup a temporary server on a laptop for example and once the upload has finished I can use

I have up to 4 servers that will run dovecot behind a load balancer, which means the same user might be accessing the same mailbox from multiple servers, and it seems like dovecot doesn't like multiple access to the dovecot indexes for the one user since I currently have them stored in a nfs home directory. Is this a bad thing? Must I keep a seperate index location per server? Just today

When I was initially testing dovecot 1.1b2,3 I had ACLs turned on and encountered this problem below. I had them turned off until now, I'll need to have ACLs working before I can widen testing. I'm not sure how to make "env MAIL=maildir:~/Maildir gdb /tmp/imap" load the ACL plugin so I assume that is why it does not crash; not getting any log either from that, maybe I made a

ACL plugin still disabled. The folders listed below definitely do have Children. telnet session problem example in 1.1b4: * OK Dovecot ready. ? login mcdouga9 password ? OK Logged in. 9 LIST "" "#shared/decs/%" * LIST (\HasNoChildren) "/" "#shared/decs/support" * LIST (\HasNoChildren) "/" "#shared/decs/receipts" * LIST (\HasNoChildren)

I think this is where I left off last weekend. Instead of this happening at the base of one of my shared mail namespaces, it happens when I try to select a top level folder inside the namespace, or try to directly select an entire path to a folder. This only happens when ACL is enabled. Let me know if I need to provide more. Thanks. Oct 27 16:03:27 gribble dovecot: IMAP(mcdouga9): file

First of all, I think dovecot is really fantastic and I have thanks for all the hard work. I think it will be the best fit for my ~5000 users when I have it setup completely. We normally have approx 500 concurrent IMAP connections during the day. I am trying to convert from courier-imap to dovecot, but I have an issue with public namespace folders that are not writable by certain users.

I was trying to debug this with gdb, but I'm not sure how to make env MAIL=maildir:~/Maildir MAIL_PLUGINS=acl ACL=vfile gdb /tmp/imap understand the #shared/decs namespace from below so I can SELECT it. I have a number of maildirs under #ahared/decs, I tried opening at least 2 of them but dovecot gives sig 11 without a direct error about the problem. I don't think this was exactly the

I have some folders with a good amount of spam. While demonstrating full text search to a user, we found a folder that would crash dovecot while doing a fts. While splitting up the mailbox to narrow it down, I found there are a good number of messages in that folder that all made it crash in the same manner. I figured I'd narrow down a single one, get that fixed, then retest the rest. I

I noticed these happen when one of my users searches his Trash folder which he doesn't empty. He uses thunderbird and it is reproducable. Feb 5 22:47:39 boomhauer dovecot: IMAP(username): file message-header-parser.c: line 350 (message_parse_header_next): assertion failed: +(IS_LWSP(line->value[0])) Feb 5 22:47:41 boomhauer dovecot: child 8022 (imap) killed with signal 6 Feb 5

Item 1: I store the indexes and control in their own directory. I noticed in 1.1: 1192 ./private/indexes/.INBOX 32 ./private/indexes/.Trash 1228 ./private/indexes 4 ./private/control/.INBOX/.INBOX 84 ./private/control/.INBOX 4 ./private/control/.Trash/.Trash 12 ./private/control/.Trash 100 ./private/control 1332 ./private .INBOX/.INBOX and .Trash/.Trash

I noticed this today, I had a user outside of our department test out dovecot. They were using squirrelmail and I noticed that dovecot thinks this user is subscribed to ALL public folders even though a dovecot ACL prevents all access. I'm pretty sure access is still denied. I was able to reproduce this with a guest account I added: l lsub "" "#shared/decs/%" * LSUB

Two nights ago I took a leap and extended my testing of dovecot 1.1 by replacing 1.0 for the approx 15 users I had on 1.0. At that time I also for the first time tried dovecot 1.1 in a load balanced 2 server configuration with indexes on NFS. I was hoping I did this right, using the mail_nfs params and 1.1 so fchown etc would flush the access cache, but I am getting a number of messages and

I have some public mailboxes setup in 1.0.0, location = maildir:/egr/mail/shared/org:CONTROL=%h/Maildir/dovecot/public/control/org:INDEX=%h/Maildir/dovecot/public/indexes/org I also have a dovecot-shared file in the maildir folders so messages that are added to the inbox get a decent set of default permissions. However, I think dovecot is also trying to fchown the index files (stored in a user

I use CONTROL=%h/Maildir/dovecot/public/control/[namespace] for each of my public namespaces. A few nights ago, I noticed having a new user subscribe to public folders for the first time will silently fail, becuase the path up to the subscriptions file will not be created. I am pretty sure dovecot used to create it for me, but maybe that changed in a recent version, or maybe I did something to

For quite some time I have seen dovecot imap exit with a sig11 without any other error message, but usually with the frequency of between once per week and once per day (per server). A few weeks ago I finally enabled core dumping with debug symbols and it seemed to happen less often. I finally caught a coredump that wasn't overwritten. I have no idea what on the client end may cause it,

Hello, I would like to suggest a change in handling of 'Time moved backwards' problem. Right now dovecot just dies. So, the scenario: 1) Colocation server is shut down for some reason. The internal time drifts. 2) Server is started again. 3) Dovecot starts successfully. 4) In about a minute, NTP daemon feels confident about adjusting the system time. 5) Dovecot sees the changed time and

A user alerted me to a problem when he renamed an upper level mail folder and lost access to a lower level folder under it. I tried with and without listescape, it didn't seem to make a difference, and I saw no errors in the dovecot log. Symptoms are if a client has a maildir folder tree .a and .a.b and renames a to c in their client, Thunderbird shows c/b but b is inaccessible since it

I haven't seen this before 1.1b16, it happened to two users today while they were searching using fts. Feb 18 16:41:36 hill dovecot: IMAP(username): file buffer.c: line 288 (buffer_set_used_size): assertion failed: (used_size <= buf->alloc) Feb 18 16:41:36 hill dovecot: child 53560 (imap) killed with signal 6 I can probably narrow it down to an example mail if needed. GNU gdb 6.1.1