[2.3.8] possible replication issue

Carsten Rosenberg

18 Oct 2019 18 Oct '19

2:52 p.m.

Hi,

some of our customers have discovered a replication issue after upgraded from 2.3.7.2 to 2.3.8.

Running 2.3.8 several replication connections are hanging until defined timeout. So after some seconds there are $replication_max_conns hanging connections. Other replications are running fast and successful.

Also running a doveadm sync tcp:... is working fine for all users.

I can't see exactly, but I haven't seen mailboxes timeouting again and again. So I would assume it's not related to the mailbox.

...

From the logs:

server1: Oct 16 08:29:25 server1 dovecot[5715]: dsync-local(username1@domain.com)<FXnVDW22pl0tGAAA1cwDxA>: Error: dsync(172.16.0.1): I/O has stalled, no activity for 600 seconds (version not received) Oct 16 08:29:25 server1 dovecot[5715]: dsync-local(username1@domain.com)<FXnVDW22pl0tGAAA1cwDxA>: Error: Timeout during state=master_recv_handshake

server2:

Oct 16 08:29:25 server2 dovecot[8113]: doveadm: Error: read(server1) failed: EOF (last sent=handshake, last recv=handshake)

There aren't any additional logs regarding the replication.

I have tried increasing vsz_limit or reducing replication_max_conns. Nothing changed.

Both customers have 10k+ users. Currently I couldn't reproduce this on smaller test systems.

Both installation were downgraded to 2.3.7.2 to fix the issue for now

I've attached a tcpdump showing the client showing the client stops sending any data after the mailbox_guid table headers.

Any idea what could be wrong here or the debug this issue?

Thanks.

Carsten Rosenberg

Attachments:

doveconf-n.txt (text/plain — 3.1 KB)
repl-dump.txt (text/plain — 2.7 KB)

Show replies by date

Wenger Florian

18 Oct 18 Oct

3:57 p.m.

I have the same Problem here. All systems are running Debian 9 amd64.

My dovecot director servers are running 2.3.8, but the Mailbox Servers having sync / replication problems with 2.3.8. So i have downgraded the Mailbox Servers to 2.3.7 and now everything works fine again...

Am 18. Oktober 2019 13:52:37 MESZ schrieb Carsten Rosenberg via dovecot dovecot@dovecot.org:

...

Hi,

some of our customers have discovered a replication issue after upgraded from 2.3.7.2 to 2.3.8.

Running 2.3.8 several replication connections are hanging until defined timeout. So after some seconds there are $replication_max_conns hanging connections. Other replications are running fast and successful.

Also running a doveadm sync tcp:... is working fine for all users.

I can't see exactly, but I haven't seen mailboxes timeouting again and again. So I would assume it's not related to the mailbox.

From the logs:

server1: Oct 16 08:29:25 server1 dovecot[5715]: dsync-local(username1@domain.com)<FXnVDW22pl0tGAAA1cwDxA>: Error: dsync(172.16.0.1): I/O has stalled, no activity for 600 seconds (version not received) Oct 16 08:29:25 server1 dovecot[5715]: dsync-local(username1@domain.com)<FXnVDW22pl0tGAAA1cwDxA>: Error: Timeout during state=master_recv_handshake

server2:

Oct 16 08:29:25 server2 dovecot[8113]: doveadm: Error: read(server1) failed: EOF (last sent=handshake, last recv=handshake)

There aren't any additional logs regarding the replication.

I have tried increasing vsz_limit or reducing replication_max_conns. Nothing changed.

--

Both customers have 10k+ users. Currently I couldn't reproduce this on smaller test systems.

Both installation were downgraded to 2.3.7.2 to fix the issue for now

--

I've attached a tcpdump showing the client showing the client stops sending any data after the mailbox_guid table headers.

Any idea what could be wrong here or the debug this issue?

Thanks.

Carsten Rosenberg

-- Diese Nachricht wurde von meinem Android-Gerät mit K-9 Mail gesendet.

Piper Andreas

5 Dec 5 Dec

8:34 a.m.

Hello,

upgrading to 2.3.9 unfortunately does *not* solve this issue:

I upgraded one of my replicators from 2.3.7.2 to 2.3.9 and after some seconds replication stopped. The other replicator remained with 2.3.7.2. After downgrading to 2.3.7.2 replication is again working fine.

I did not try to upgrade both replicators up to now, as this is a live production system. Is there a chance, that upgrading both replicators will solve the problem?

The machines are running Ubuntu 18.04

Any help is appreciated.

Thanks, Andreas

Am 18.10.19 um 13:52 schrieb Carsten Rosenberg via dovecot:

...

Hi,

some of our customers have discovered a replication issue after upgraded from 2.3.7.2 to 2.3.8.

Running 2.3.8 several replication connections are hanging until defined timeout. So after some seconds there are $replication_max_conns hanging connections. Other replications are running fast and successful.

Also running a doveadm sync tcp:... is working fine for all users.

I can't see exactly, but I haven't seen mailboxes timeouting again and again. So I would assume it's not related to the mailbox.

From the logs:

server1: Oct 16 08:29:25 server1 dovecot[5715]: dsync-local(username1@domain.com)<FXnVDW22pl0tGAAA1cwDxA>: Error: dsync(172.16.0.1): I/O has stalled, no activity for 600 seconds (version not received) Oct 16 08:29:25 server1 dovecot[5715]: dsync-local(username1@domain.com)<FXnVDW22pl0tGAAA1cwDxA>: Error: Timeout during state=master_recv_handshake

server2:

Oct 16 08:29:25 server2 dovecot[8113]: doveadm: Error: read(server1) failed: EOF (last sent=handshake, last recv=handshake)

There aren't any additional logs regarding the replication.

I have tried increasing vsz_limit or reducing replication_max_conns. Nothing changed.

--

Both customers have 10k+ users. Currently I couldn't reproduce this on smaller test systems.

Both installation were downgraded to 2.3.7.2 to fix the issue for now

--

I've attached a tcpdump showing the client showing the client stops sending any data after the mailbox_guid table headers.

Any idea what could be wrong here or the debug this issue?

Thanks.

Carsten Rosenberg

Dr. Andreas Piper, Hochschulrechenzentrum der Philipps-Univ. Marburg Hans-Meerwein-Straße 6, 35032 Marburg, Germany Phone: +49 6421 28-23521 Fax: -26994 E-Mail: piper@HRZ.Uni-Marburg.DE

Timo Sirainen

2:15 p.m.

I think there's a good chance that upgrading both will fix it. The bug already existed in old versions, it just wasn't normally triggered. Since v2.3.8 this situation is triggered on one dsync side, so the v2.3.9 fix needs to be on the other side.

...

On 5. Dec 2019, at 8.34, Piper Andreas via dovecot dovecot@dovecot.org wrote:

Hello,

upgrading to 2.3.9 unfortunately does *not* solve this issue:

I upgraded one of my replicators from 2.3.7.2 to 2.3.9 and after some seconds replication stopped. The other replicator remained with 2.3.7.2. After downgrading to 2.3.7.2 replication is again working fine.

I did not try to upgrade both replicators up to now, as this is a live production system. Is there a chance, that upgrading both replicators will solve the problem?

The machines are running Ubuntu 18.04

Any help is appreciated.

Thanks, Andreas

Am 18.10.19 um 13:52 schrieb Carsten Rosenberg via dovecot:

...
Hi, some of our customers have discovered a replication issue after upgraded from 2.3.7.2 to 2.3.8. Running 2.3.8 several replication connections are hanging until defined timeout. So after some seconds there are $replication_max_conns hanging connections. Other replications are running fast and successful. Also running a doveadm sync tcp:... is working fine for all users. I can't see exactly, but I haven't seen mailboxes timeouting again and again. So I would assume it's not related to the mailbox. From the logs: server1: Oct 16 08:29:25 server1 dovecot[5715]: dsync-local(username1@domain.com)<FXnVDW22pl0tGAAA1cwDxA>: Error: dsync(172.16.0.1): I/O has stalled, no activity for 600 seconds (version not received) Oct 16 08:29:25 server1 dovecot[5715]: dsync-local(username1@domain.com)<FXnVDW22pl0tGAAA1cwDxA>: Error: Timeout during state=master_recv_handshake server2: Oct 16 08:29:25 server2 dovecot[8113]: doveadm: Error: read(server1) failed: EOF (last sent=handshake, last recv=handshake) There aren't any additional logs regarding the replication. I have tried increasing vsz_limit or reducing replication_max_conns. Nothing changed.

Both customers have 10k+ users. Currently I couldn't reproduce this on smaller test systems. Both installation were downgraded to 2.3.7.2 to fix the issue for now

I've attached a tcpdump showing the client showing the client stops sending any data after the mailbox_guid table headers. Any idea what could be wrong here or the debug this issue? Thanks. Carsten Rosenberg

--

Dr. Andreas Piper, Hochschulrechenzentrum der Philipps-Univ. Marburg Hans-Meerwein-Straße 6, 35032 Marburg, Germany Phone: +49 6421 28-23521 Fax: -26994 E-Mail: piper@HRZ.Uni-Marburg.DE mailto:piper@HRZ.Uni-Marburg.DE

Piper Andreas

6 Dec 6 Dec

8:09 a.m.

Hello Timo,

upgrading both replicators did the job! Both replicators now run v2.3.9 and replication works fine, all sync-jobs which queued up during the upgrading have been processed successfully.

Thanks for the reassurement and all your great work with dovecot,

Andreas

Am 05.12.19 um 13:15 schrieb Timo Sirainen via dovecot:

...

I think there's a good chance that upgrading both will fix it. The bug already existed in old versions, it just wasn't normally triggered. Since v2.3.8 this situation is triggered on one dsync side, so the v2.3.9 fix needs to be on the other side.

...
On 5. Dec 2019, at 8.34, Piper Andreas via dovecot mailto:dovecot@dovecot.org> wrote:

Hello,

upgrading to 2.3.9 unfortunately does *not* solve this issue:

I upgraded one of my replicators from 2.3.7.2 to 2.3.9 and after some seconds replication stopped. The other replicator remained with 2.3.7.2. After downgrading to 2.3.7.2 replication is again working fine.

I did not try to upgrade both replicators up to now, as this is a live production system. Is there a chance, that upgrading both replicators will solve the problem?

The machines are running Ubuntu 18.04

Any help is appreciated.

Thanks, Andreas

Am 18.10.19 um 13:52 schrieb Carsten Rosenberg via dovecot:

...
Hi, some of our customers have discovered a replication issue after upgraded from 2.3.7.2 to 2.3.8. Running 2.3.8 several replication connections are hanging until defined timeout. So after some seconds there are $replication_max_conns hanging connections. Other replications are running fast and successful. Also running a doveadm sync tcp:... is working fine for all users. I can't see exactly, but I haven't seen mailboxes timeouting again and again. So I would assume it's not related to the mailbox. From the logs: server1: Oct 16 08:29:25 server1 dovecot[5715]: dsync-local(username1@domain.com mailto:username1@domain.com)<FXnVDW22pl0tGAAA1cwDxA>: Error: dsync(172.16.0.1): I/O has stalled, no activity for 600 seconds (version not received) Oct 16 08:29:25 server1 dovecot[5715]: dsync-local(username1@domain.com mailto:username1@domain.com)<FXnVDW22pl0tGAAA1cwDxA>: Error: Timeout during state=master_recv_handshake server2: Oct 16 08:29:25 server2 dovecot[8113]: doveadm: Error: read(server1) failed: EOF (last sent=handshake, last recv=handshake) There aren't any additional logs regarding the replication. I have tried increasing vsz_limit or reducing replication_max_conns. Nothing changed.

Both customers have 10k+ users. Currently I couldn't reproduce this on smaller test systems. Both installation were downgraded to 2.3.7.2 to fix the issue for now

I've attached a tcpdump showing the client showing the client stops sending any data after the mailbox_guid table headers. Any idea what could be wrong here or the debug this issue? Thanks. Carsten Rosenberg

KOCIK Fabien (Acoss)

9:25 a.m.

Hello all,

Just tested this morning : I can confirm that issue seems to be resolved for me after upgrading both servers from 2.3.7.2 to 2.3.9.

Refs :

No more "I/O has stalled" error messages and replication works fine now. Thanks very much to the Dovecot team.

Have a nice day. Fabien

-----Message d'origine----- De : dovecot dovecot-bounces@dovecot.org De la part de Piper Andreas via dovecot Envoyé : vendredi 6 décembre 2019 07:10 À : dovecot@dovecot.org Objet : Re: [2.3.8] possible replication issue

Hello Timo,

upgrading both replicators did the job! Both replicators now run v2.3.9 and replication works fine, all sync-jobs which queued up during the upgrading have been processed successfully.

Thanks for the reassurement and all your great work with dovecot,

Andreas

Am 05.12.19 um 13:15 schrieb Timo Sirainen via dovecot:

...

I think there's a good chance that upgrading both will fix it. The bug already existed in old versions, it just wasn't normally triggered. Since v2.3.8 this situation is triggered on one dsync side, so the v2.3.9 fix needs to be on the other side.

...
On 5. Dec 2019, at 8.34, Piper Andreas via dovecot mailto:dovecot@dovecot.org> wrote:

Hello,

upgrading to 2.3.9 unfortunately does *not* solve this issue:

I upgraded one of my replicators from 2.3.7.2 to 2.3.9 and after some seconds replication stopped. The other replicator remained with 2.3.7.2. After downgrading to 2.3.7.2 replication is again working fine.

I did not try to upgrade both replicators up to now, as this is a live production system. Is there a chance, that upgrading both replicators will solve the problem?

The machines are running Ubuntu 18.04

Any help is appreciated.

Thanks, Andreas

Am 18.10.19 um 13:52 schrieb Carsten Rosenberg via dovecot:

...
Hi, some of our customers have discovered a replication issue after upgraded from 2.3.7.2 to 2.3.8. Running 2.3.8 several replication connections are hanging until defined timeout. So after some seconds there are $replication_max_conns hanging connections. Other replications are running fast and successful. Also running a doveadm sync tcp:... is working fine for all users. I can't see exactly, but I haven't seen mailboxes timeouting again and again. So I would assume it's not related to the mailbox. From the logs: server1: Oct 16 08:29:25 server1 dovecot[5715]: dsync-local(username1@domain.com mailto:username1@domain.com)<FXnVDW22pl0tGAAA1cwDxA>: Error: dsync(172.16.0.1): I/O has stalled, no activity for 600 seconds (version not received) Oct 16 08:29:25 server1 dovecot[5715]: dsync-local(username1@domain.com mailto:username1@domain.com)<FXnVDW22pl0tGAAA1cwDxA>: Error: Timeout during state=master_recv_handshake server2: Oct 16 08:29:25 server2 dovecot[8113]: doveadm: Error: read(server1) failed: EOF (last sent=handshake, last recv=handshake) There aren't any additional logs regarding the replication. I have tried increasing vsz_limit or reducing replication_max_conns. Nothing changed.

Both customers have 10k+ users. Currently I couldn't reproduce this on smaller test systems. Both installation were downgraded to 2.3.7.2 to fix the issue for now

I've attached a tcpdump showing the client showing the client stops sending any data after the mailbox_guid table headers. Any idea what could be wrong here or the debug this issue? Thanks. Carsten Rosenberg

1808

Age (days ago)

1857

Last active (days ago)

List overview

5 comments

5 participants

participants (5)

Carsten Rosenberg
KOCIK Fabien (Acoss)
Piper Andreas
Timo Sirainen
Wenger Florian