[Dovecot] Dovecot 1.2 ACL problem.
Sir, I have migrate from courier-imap to dovecot without any problem but on courier in have shared folder which access on the basis on rights. I have manage these folders in dovecot but i have one problem all the shared folder is accessible by all user whether they have rights or not i have place dovecot-acl in shared folder and assign user my dovecot-acl is user=raj...@teamsg.in tr. But all other user which is not listing on dovecot-acl also able to access these folder. Here is the output of dovecot -n. # 1.2.17: /usr/local/etc/dovecot.conf # OS: Linux 2.6.18-194.3.1.el5 i686 CentOS release 5.5 (Final) base_dir: /var/run/dovecot/ log_path: /var/log/dovecot.log ssl_listen: 192.168.0.251:993 ssl_cert_file: /var/qmail/control/servercert.pem ssl_key_file: /var/qmail/control/servercert.pem login_dir: /var/run/dovecot//login login_executable: /usr/local/libexec/dovecot/imap-login first_valid_uid: 508 last_valid_uid: 508 mail_location: maildir:~/Maildir namespace: type: private separator: . prefix: INBOX. inbox: yes list: yes subscriptions: yes namespace: type: shared separator: . prefix: INBOX.Shared. location: maildir:/home/vpopmail/domains/sgcricket.com/Maildir/Out:INDEX=~/Maildir/Shared list: children auth default: user: vpopmail passdb: driver: vpopmail userdb: driver: vpopmail plugin: acl: vfile Please help whats wrong i am doing. Rajeev Sharma.
Re: [Dovecot] Seen flag getting lost
On Fri, 2011-11-25 at 13:49 +0100, Edgar Fuß wrote: > > Then the 'S' flag is added to the current Maildir filename without > > losing any other changes. > And this is supposed to work even over NFS? Great. Absolutely, always has worked over NFS signature.asc Description: This is a digitally signed message part
Re: [Dovecot] Seen flag getting lost
On 25.11.2011, at 14.49, Edgar Fuß wrote: >> That shouldn't happen because of Dovecot's indexes. > Hm, also if the indexes are local? Fine. Yep. It just means that indexes then aren't fully up to date. Hmm. I guess possibly one potential problem is if 1. dovecot.index.log says that "set flags to ()" but then it doesn't actually finish that 2. Next time indexes are accessed from second server, and flags are set to (\Seen) 3. And then finally user logs in back to original server, which sees that "aha, flags should be set to ()" and does that. But the problem is the first step. It shouldn't normally do it only halfway through, unless it crashes or something.. Also typically flags aren't replaced, they're added/removed, but this completely depends on the client and I don't remember what Apple Mail does. >> Then the 'S' flag is added to the current Maildir filename without >> losing any other changes. > And this is supposed to work even over NFS? Great. > > So, what can I do to track down the problem as, according to what I read from > your answer, it's probably a Dovecot bug? You could try enabling mail_log plugin with flag_change event to see when it actually changes. http://wiki.dovecot.org/Plugins/MailLog
Re: [Dovecot] Seen flag getting lost
On 25.11.2011 17:59, wrote Ed W: On 25/11/2011 12:49, Edgar Fuß wrote: You don't have any Thunderbird clients accessing this box do you? I have some wierd issue where our TB with the option "don't mark message read", still triggers messages to be marked read... Wierdly it only does it on some messages and all those from specific senders - can't correlate it with anything obvious in the message itself though... I had the same issue a while ago with Thunderbird. On one installation it worked on another it didn't. After comparing their config I found the culprit. Changing mail.server.default.mime_parts_on_demand to **false** fixed this issue for me. This setting is discussed in... http://kb.mozillazine.org/Entire_message_fetched_when_opening_a_IMAP_message ... but not mentioning the "don't mark message as read" issue. Regards, e-frog * * Englisch * Deutsch * Englisch * Deutsch
Re: [Dovecot] Seen flag getting lost
> You don't have any Thunderbird clients accessing this box do you? Yes, I have. But I also experienced the problem myself only using mutt and Apple Mail. > I have some w[ei]rd issue where our TB with the option > "don't mark message read", still triggers messages to be marked read It's the other way round. People read their messages, they are flagged as read and all of a sudden revert to unread.
Re: [Dovecot] Seen flag getting lost
On 25/11/2011 12:49, Edgar Fuß wrote: >> That shouldn't happen because of Dovecot's indexes. > Hm, also if the indexes are local? Fine. > >> Then the 'S' flag is added to the current Maildir filename without >> losing any other changes. > And this is supposed to work even over NFS? Great. > > So, what can I do to track down the problem as, according to what I read from > your answer, it's probably a Dovecot bug? You don't have any Thunderbird clients accessing this box do you? I have some wierd issue where our TB with the option "don't mark message read", still triggers messages to be marked read... Wierdly it only does it on some messages and all those from specific senders - can't correlate it with anything obvious in the message itself though... Regards Ed W
Re: [Dovecot] Which FTS to use!
On 25.11.2011, at 17.29, Henrique Santos Fernandes wrote:
> Hello, i am currently using squat dovecot, but some users that have lots os
> emails ( 1GB ) are complaining about speed.
> Iam using dovecot 2.0.13 on debian 6
>
> my squat conf is
> plugin {
> fts = squat
> fts_squat = partial=4 full=5
> }
>
> I am thinking in maybe start to use Solr, so any one can help me with an
> decision?
With v2.0 Solr is the only other choice.
> I am probably gonna update dovecot to 2.1 next year. and them maybe change
> the FTS plugin.
With v2.1 Lucene is another possibility, and it's somewhat easier to install
than Solr. Also v2.1's Solr backend is improved, so I wouldn't really recommend
spending a lot of time and disk I/O on v2.0's Solr since you'll need to rebuild
the index anyway in v2.1 (or keep using "solr_old" backend with some missing
featues).
> This year i will get 10k disk only to store the dovecot index, rigth now
> they are saved on the same storage that the mails are and also change the
> "full" parameter to something like 10 or 15 I know i will use more disk
> space for it, but it will make a lot of differences right? it will match
> the message without the need of opening it for any search smaller than 10
> character i guess.
>
>
> But this will be enough? or Solr or Lucene still better?
The main problem with Squat is its index updating performance, which sucks.
> Anyway, i am not current updating the indexes automatic, but i am planing
> on it, just not sure what is better, the command doveadm index or doveadm
> search... with one is better to do the task?
doveadm index
[Dovecot] Which FTS to use!
Hello, i am currently using squat dovecot, but some users that have lots os
emails ( 1GB ) are complaining about speed.
Iam using dovecot 2.0.13 on debian 6
my squat conf is
plugin {
fts = squat
fts_squat = partial=4 full=5
}
I am thinking in maybe start to use Solr, so any one can help me with an
decision?
I am probably gonna update dovecot to 2.1 next year. and them maybe change
the FTS plugin.
This year i will get 10k disk only to store the dovecot index, rigth now
they are saved on the same storage that the mails are and also change the
"full" parameter to something like 10 or 15 I know i will use more disk
space for it, but it will make a lot of differences right? it will match
the message without the need of opening it for any search smaller than 10
character i guess.
But this will be enough? or Solr or Lucene still better?
Anyway, i am not current updating the indexes automatic, but i am planing
on it, just not sure what is better, the command doveadm index or doveadm
search... with one is better to do the task?
Thanks!
[]'sf.rique
Re: [Dovecot] Seen flag getting lost
> That shouldn't happen because of Dovecot's indexes. Hm, also if the indexes are local? Fine. > Then the 'S' flag is added to the current Maildir filename without > losing any other changes. And this is supposed to work even over NFS? Great. So, what can I do to track down the problem as, according to what I read from your answer, it's probably a Dovecot bug?
Re: [Dovecot] v2.1.rc1 released
On Thu, Nov 24, 2011 at 09:25:47PM +0200, Timo Sirainen wrote: > I'm not sure if changing /etc/security/limits.conf helps. It's probably > only used by PAM when user logs in, so if Dovecot is started in system > bootup it's unlikely to have been even read yet. Also I think some OSes > override the limits in /etc/init.d/ scripts. Of course, I could be > completely wrong in all of the above, I haven't really tested any of > it :) The OP was running Centos, and the Red Hat way of increasing nofile for services is to add the ulimit setting to /etc/sysconfig/dovecot: ulimit -n 32768 This is what we use, and it survives dovecot upgrades (since this is tagged as a configfile in RPM, not as script that will be overwritten on upgrades). -jf
Re: [Dovecot] v2.1.rc1 released
El 24/11/11 21:35, Miguel Tormo escribió: El Jueves, 24 de Noviembre de 2011 20:25:47 Timo Sirainen escribió: I'm not sure if changing /etc/security/limits.conf helps. It's probably only used by PAM when user logs in, so if Dovecot is started in system bootup it's unlikely to have been even read yet. Also I think some OSes override the limits in /etc/init.d/ scripts. Of course, I could be completely wrong in all of the above, I haven't really tested any of it Well actually this is tricky, I'm not sure if it would work after a system boot, but I'm sure it works after a reboot of the service. It is also true what you say that some OSes override the limits in the /etc/init.d/ scripts, for example issuing an 'ulimit -n 4096' (if using bash) before dropping privileges (that should be inherited, but if privileges are dropped using "su" or something that uses PAM, then the /etc/security/limits.conf file comes into play again). I tested this "intensively" some time ago, and Timo is right -as usual :) -: modifications in limits.conf are only relevant after a login. This means that they are not enforced at system boot, but they are if you reboot the service by hand (because you have already logged in). The permanent solution is, as Noel Butler has pointed, modifying the startup script and using ulimit before actually starting dovecot. HTH
Re: [Dovecot] Questions about Proxy/Director
On Fri, 25 Nov 2011 09:17:15 +0100, Jan-Frode Myklebust wrote: On Thu, Nov 24, 2011 at 11:58:01PM +0100, Patrick Westenberg wrote: What is your frontend configuration? The frontend is two servers running dovecot director, and ucarp for managing failover ip-addresses. Two additional directors or the two directors from your posted config (.15/.17)? Will I have to configure external IPs for the backend servers or just internal for the connections to and from the directors? I'm not sure if the director "redirects" the connections to the backend servers or if they pass through on the internal network. How many users do you have that you need that many backend servers? I don't think I'm allowed to share that externally, sorry. This is the mail-system of an ISP, so we have quite a few users.. :-) Ok :-)
Re: [Dovecot] Questions about Proxy/Director
On Thu, Nov 24, 2011 at 11:58:01PM +0100, Patrick Westenberg wrote:
>
> What is your frontend configuration?
The frontend is two servers running dovecot director, and ucarp for
managing failover ip-addresses.
=
# 2.0.14: /etc/dovecot/dovecot.conf
# OS: x86_64 Red Hat Enterprise Linux Server release 5.x
director_mail_servers = 192.168.42.7 192.168.42.8 192.168.42.9 192.168.42.10
192.168.42.11 192.168.42.28 192.168.42.29
director_servers = 192.168.42.15 192.168.42.17
disable_plaintext_auth = no
listen = *
lmtp_proxy = yes
passdb {
args = proxy=y nopassword=y
driver = static
}
service anvil {
client_limit = 2051
}
service auth {
client_limit = 4196
unix_listener auth-userdb {
user = dovecot
}
}
service director {
fifo_listener login/proxy-notify {
mode = 0666
}
inet_listener {
port = 5515
}
unix_listener director-userdb {
mode = 0600
}
unix_listener login/director {
mode = 0666
}
}
service imap-login {
executable = imap-login director
process_limit = 1024
}
service lmtp {
inet_listener lmtp {
address = *
port = 24
}
}
service pop3-login {
executable = pop3-login director
process_limit = 1024
}
ssl_cert = How many users do you have that you need that many backend servers?
I don't think I'm allowed to share that externally, sorry. This is the
mail-system of an ISP, so we have quite a few users.. :-)
> Are your index stored together with the mails?
They're not stored together, because we made the error of not creating a
home directory for our users, but both maildirs and indexes are stored on
similar cluster fs' (GPFS).
-jf
