Re: Replicator bug report

Yeah. As I said it's a bug, it should never allocate that much data stack memory. I don't know any workarond immediately for this. Maybe try setting `service_count=1` for the replicator service, it might or not help. Aki > On 12/12/2021 14:31 Jörg Faudin Schulz wrote: > > > No ... it just

Re: Replicator bug report

No ... it just crashes a bit later: Dec 12 13:28:26 replicator: Panic: data stack: Out of memory when allocating 68719476776 bytes Am 12.12.21 um 08:23 schrieb Daniel Miller: It appears when I set vsz_limit=0 it works without crashing. So the problem appears when setting an explicit maximum.

Re[9]: Replicator bug report

It appears when I set vsz_limit=0 it works without crashing. So the problem appears when setting an explicit maximum. -- Daniel On 12/7/2021 9:57:29 PM, "Aki Tuomi" wrote: On 7 December 2021 23.10.50 UTC, Daniel Miller wrote: On 12/7/2021 12:29:49 PM, "Daniel Miller" wrote: service re

Re: Re[7]: Replicator bug report

On 7 December 2021 23.10.50 UTC, Daniel Miller wrote: >On 12/7/2021 12:29:49 PM, "Daniel Miller" wrote: > >>>service replicator { >>>vsz_limit = 2G >>>} >>> >>>Aki >>> >>Tried that - got another one. > > >I just tried setting >service replicator { > vcsz_limit = 5G >} >and I still get: >

Re[7]: Replicator bug report

On 12/7/2021 12:29:49 PM, "Daniel Miller" wrote: service replicator { vsz_limit = 2G } Aki Tried that - got another one. I just tried setting service replicator { vcsz_limit = 5G } and I still get: Dec 7 15:08:25 bubba dovecot: replicator: Panic: data stack: Out of memory when allo

Re[6]: Replicator bug report

service replicator { vsz_limit = 2G } Aki Tried that - got another one. [New LWP 14072] Core was generated by `dovecot/replicator'. Program terminated with signal SIGABRT, Aborted. ---Type to continue, or q to quit--- #0 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c

Re: Re[4]: Replicator bug report

On 7 December 2021 19.27.20 UTC, Daniel Miller - CLOUD wrote: >>Use >> >>gdb /path/to/replicator /path/to/core >>bt full >> >>Aki >> service replicator { vsz_limit = 2G } Aki

Re[4]: Replicator bug report

Use gdb /path/to/replicator /path/to/core bt full Aki root@bubba:/var/core# gdb /usr/lib/dovecot/replicator /var/core/11199.replicator GNU gdb (Ubuntu 8.1.1-0ubuntu1) 8.1.1 Copyright (C) 2018 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later

Re: Re[2]: Replicator bug report

On 7 December 2021 19.11.18 UTC, Daniel Miller wrote: >-- Original Message -- >> >>Hi! >> >>Can you instead submit gdb bt full output and doveconf -n? >> >>Aki >> > >Certainly - but I need to know how. The problem is during TCP >replication. > > >-- >Daniel > > Use gdb /path/to/rep

Re[2]: Replicator bug report

-- Original Message -- Hi! Can you instead submit gdb bt full output and doveconf -n? Aki Certainly - but I need to know how. The problem is during TCP replication. Here is dovecot -n: # 2.3.17.1 (476cd46418): /etc/dovecot/dovecot.conf # Pigeonhole version 0.5.17.1 (a1a0b892) #

Re: Replicator bug report

On 7 December 2021 18.55.40 UTC, Daniel Miller wrote: >I've run dovecot-sysreport -o and generated a file - but it's a >few gigs in size. Am I generating the core dump incorrectly? Should I do >something different? > >-- >Daniel Hi! Can you instead submit gdb bt full output and doveconf -n

Replicator bug report

I've run dovecot-sysreport -o and generated a file - but it's a few gigs in size. Am I generating the core dump incorrectly? Should I do something different? -- Daniel