similar to: Auth process sometimes stop responding after upgrade

Hi all; I've upgraded a ring of dovecot directors from 2.2.15 to 2.2.36. After the upgrade I've got some instability: a few time per day per server, seemly at random, the auth process stop responding and the clients cannot authenticate any more: Sep 6 14:45:51 imap-front13 dovecot: pop3-login: Warning: Auth process not responding, delayed sending initial response (greeting):

In data venerd? 7 settembre 2018 10:06:00 CEST, Sami Ketola ha scritto: > > On 7 Sep 2018, at 11.00, Simone Lazzaris <s.lazzaris at interactive.eu> > > wrote: > > > > > > The only suspect thing is this: > > > > Sep 6 14:45:41 imap-front13 dovecot: director: doveadm: Host > > 192.168.1.142 > > vhost count changed from 100 to 0 >

In data venerd? 7 settembre 2018 09:39:23 CEST, Aki Tuomi ha scritto: > Do you see anything else in logs? > Just the usual proxy messages, like these: Sep 6 14:45:41 imap-front13 dovecot: imap-login: proxy(alessandro at soldinicarrelli.it): disconnecting 5.169.233.224 (Disconnected by server(0s idle, in=451, out=3556)): user=<alessandro at soldinicarrelli.it>, method=PLAIN,

In data venerd? 7 settembre 2018 11:20:49 CEST, Sami Ketola ha scritto: > > On 7 Sep 2018, at 11.25, Simone Lazzaris <simone.lazzaris at qcom.it> wrote: > > Actually, I have a poolmon script running that should drop vhost count for > > unresponsive backends; the strage thing is, the backends are NOT > > unresponsive, they are working as ususal. > If it's this

> On 7 Sep 2018, at 11.00, Simone Lazzaris <s.lazzaris at interactive.eu> wrote: > > > The only suspect thing is this: > > Sep 6 14:45:41 imap-front13 dovecot: director: doveadm: Host 192.168.1.142 > vhost count changed from 100 to 0 > Sep 6 14:45:41 imap-front13 dovecot: director: doveadm: Host 192.168.1.143 > vhost count changed from 100 to 0 > Sep 6

In data marted? 11 settembre 2018 10:46:30 CEST, Timo Sirainen ha scritto: > On 11 Sep 2018, at 10.57, Simone Lazzaris <s.lazzaris at interactive.eu> wrote: > > Sep 11 03:25:55 imap-front4 dovecot: director: Panic: file > > doveadm-connection.c: line 1097 (doveadm_connection_deinit): assertion > > failed: (conn->to_ring_sync_abort == NULL) Sep 11 03:25:55 imap-front4

Can you provide 'bt full' Aki On 18.09.2018 15:15, Simone Lazzaris wrote: > > In data marted? 18 settembre 2018 14:07:26 CEST, Aki Tuomi ha scritto: > > > If you are using systemd, create > > > /etc/systemd/system/dovecot.service.d/limits.conf and put > > > [Service]LimitCORE=infinity > > > > > > and run? > > > systemctl

Alas, I've set fs.suid_dumpable to 2 but the core is not dumped. So far I've checked: - ulimit -c unlimited, done - /proc/sys/kernel/core_pattern is set to /var/tmp/core.%p - /var/tmp is chmod 1777 - daemon is not chrooted - sysctl -w fs.suid_dumpable=2 - dir /var/tmp is empty and filesystem has 2GB free This is the logfile: Sep 18 13:54:22 imap-front4 dovecot: director: doveadm: Host

In data marted? 18 settembre 2018 14:07:26 CEST, Aki Tuomi ha scritto: > If you are using systemd, create > /etc/systemd/system/dovecot.service.d/limits.conf and put > [Service]LimitCORE=infinity > > and run > systemctl daemon-reloadsystemctl restart dovecot Nope, I'm on a debian 7, without systemd. Anyway, I've resolved the issue: I had to set fs.suid_dumpable BEFORE

In data gioved? 19 marzo 2020 09:10:06 CET, Aki Tuomi ha scritto: > If you are using director, you can issue `doveadm quota recalc -A` on > director, which will then direct it to correct backend. > > For this to work, it requires that you setup doveadm server on the backend, > and configure doveadm_port and doveadm_password on director. It also either > needs iteratable user

In data mercoled? 18 marzo 2020 23:08:27 CET, Alessio Cecchi ha scritto: > Ciao Simone, > > why you want each backend to recalc quota only for its managed users and > not run "doveadm quota recalc -A" only one time from a backend tha > recalc quota for all users? > Because the indexes are local to the backends. If I make the recalc from one backend only, that one

Hello dovecot! I administer a dovecot installation with 30k users. I've got 4 dovecot directors as frontend and 10 backends. The mailbox are now in maildir format, with maildir++ quota, on a shared netapp filer. Indexes are local on each backend. I'm reconfiguring the quota plugin: as a first step, I want to use the clone plugin to keep a copy of the quota on a redis database. Next,

Hi all; I've just compiled version 2.2.36.1 on a new debian buster (debian 10) and it seems that this patch https://github.com/dovecot/core/commit/ 63a74b9e8e0604486a15a879e7f1a27257322400.patch is needed to make it work. Without the patch, the authentication segfaults. *Simone Lazzaris* *Qcom S.p.A.* simone.lazzaris at qcom.it[1] | www.qcom.it[2] * LinkedIn[3]* | *Facebook*[4] [5]

Ciao Simone, why you want each backend to recalc quota only for its managed users and not run "doveadm quota recalc -A" only one time from a backend tha recalc quota for all users? Il 10/03/20 11:18, Simone Lazzaris ha scritto: > Hello dovecot! > > I administer a dovecot installation with 30k users. I've got 4 dovecot > directors as frontend and 10 backends. >

> Hi all, again; > > I've enabled the core dumps and let it go for some day waiting for the issue > to reoccur. > > Meantime I've also upgraded the poolmon script, as Sami suggested. > > It seems that the upgrade has scared the issue away, because it no longer > occurred. > > Maybe the problem is related to the way the old poolmon talked to the >

Some more information: the issue has just occurred, again on an instance without the "service_count = 0" configuration directive on pop3-login. I've observed that while the issue is occurring, the director process goes 100% CPU. I've straced the process. It is seemingly looping: ... ... epoll_ctl(13, EPOLL_CTL_ADD, 78, {EPOLLIN|EPOLLPRI|EPOLLERR|EPOLLHUP, {u32=149035320,

In data luned? 10 settembre 2018 09:58:50 CEST, Timo Sirainen ha scritto: > On 8 Sep 2018, at 15.18, Simone Lazzaris <simone.lazzaris at qcom.it> wrote: > > Timo, unfortunately the patch doesn't compile; I've moved the declaration > > of "conn" one line up to make it work. > > Oops, I guess I was too much in a hurry to even compile it. Here's a

On 19 Sep 2018, at 11.30, Timo Sirainen <tss at iki.fi> wrote: > >> >> On 19 Sep 2018, at 11.11, Simone Lazzaris <s.lazzaris at interactive.eu <mailto:s.lazzaris at interactive.eu>> wrote: >> >> In data mercoled? 19 settembre 2018 09:30:47 CEST, Timo Sirainen ha scritto: >> > On 18 Sep 2018, at 15.15, Simone Lazzaris <s.lazzaris at

> On 7 Sep 2018, at 11.25, Simone Lazzaris <simone.lazzaris at qcom.it> wrote: > Actually, I have a poolmon script running that should drop vhost count for unresponsive backends; the strage thing is, the backends are NOT unresponsive, they are working as ususal. > If it's this one https://github.com/brandond/poolmon/blob/master/poolmon

On 8 Sep 2018, at 15.18, Simone Lazzaris <simone.lazzaris at qcom.it> wrote: > > Timo, unfortunately the patch doesn't compile; I've moved the declaration of > "conn" one line up to make it work. Oops, I guess I was too much in a hurry to even compile it. Here's a new patch that compiles and passes our director CI tests. -------------- next part