Ok a little more help : vsz_limit = 0 --> means unlimited ram for allocation, change this/try 2g etc pending avaliable ram.

I try with 524M, 1G, 2G, 4G and 8G but in any case repclicator proces got crash.

I think it is because of the domain, obviously you should only be running a website.

-----Original Message----- From: dovecot dovecot-bounces@dovecot.org On Behalf Of hi@zakaria.website Sent: Saturday, 4 June 2022 11:15 To: Dovecot dovecot@dovecot.org Subject: Re: Replicator: Panic: data stack: Out of memory

On 2022-06-04 02:46, Ivan Jurišić wrote:

...

...

Ok a little more help : vsz_limit = 0 --> means unlimited ram for allocation, change this/try 2g etc pending avaliable ram.

I try with 524M, 1G, 2G, 4G and 8G but in any case repclicator proces got crash.

Maybe there is another service process causing OOM? e.g. check clamd, antivirus DBs tend to be quite big and in updating for sometime becomes double the size due to reloading.

Also, somtimes httpd service when using event worker, and its not tuned properly, it will cause the OOM crash to other service along itself.

:) You should not be responding to my post, as it was sort of off topic and only focused on the text '.website'

just an fyi domain is registered and appears to be active so there should not be any issues with the domain

.website is an actual domain (like .com, .ca etc)

however i did note

ZAKARIA.WEBSITE. 14400 IN MX 10 ZAKARIA.WEBSITE.

usually the mx record points to an actual sub domain like mail. or whatever

if you are running everything on one server then this is ok

its just usually better to seperate the mx record incase you want to goto a different server down the road.

Domain Name: ZAKARIA.WEBSITE Registry Domain ID: D198561373-CNIC Registrar WHOIS Server: whois.ionos.com Registrar URL: https://ionos.com Updated Date: 2021-11-02T01:42:25.0Z Creation Date: 2020-08-29T09:28:59.0Z Registry Expiry Date: 2022-08-29T23:59:59.0Z Registrar: IONOS SE Registrar IANA ID: 83 Domain Status: clientTransferProhibited https://icann.org/epp#clientTransferProhibited Registrant Organization: 1&1 Internet Limited Registrant State/Province: GLS Registrant Country: GB Registrant Email: Please query the RDDS service of the Registrar of Record identified in this output for information on how to contact the Registrant, Admin, or Tech contact of the queried domain name. Admin Email: Please query the RDDS service of the Registrar of Record identified in this output for information on how to contact the Registrant, Admin, or Tech contact of the queried domain name. Tech Email: Please query the RDDS service of the Registrar of Record identified in this output for information on how to contact the Registrant, Admin, or Tech contact of the queried domain name. Name Server: NS1.ZAKARIA.WEBSITE Name Server: NS2.ZAKARIA.WEBSITE

## nslookup ZAKARIA.WEBSITE Server: 10.220.0.2 Address: 10.220.0.2#53

Non-authoritative answer: Name: ZAKARIA.WEBSITE Address: 213.171.210.111 Name: ZAKARIA.WEBSITE Address: 2a00:da00:1800:834c::1

## dig mx ZAKARIA.WEBSITE

; <<>> DiG 9.14.3 <<>> mx ZAKARIA.WEBSITE ;; global options: +cmd ;; Got answer: ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 32110 ;; flags: qr rd ra ad; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 3

;; OPT PSEUDOSECTION: ; EDNS: version: 0, flags:; udp: 4096 ; COOKIE: ba2f2ec47dfcc90f4581111d629b4d2855567ad8dfa57bf8 (good) ;; QUESTION SECTION: ;ZAKARIA.WEBSITE. IN MX

;; ANSWER SECTION: ZAKARIA.WEBSITE. 14400 IN MX 10 ZAKARIA.WEBSITE.

;; ADDITIONAL SECTION: zakaria.website. 14372 IN A 213.171.210.111 zakaria.website. 14372 IN AAAA 2a00:da00:1800:834c::1

;; Query time: 87 msec ;; SERVER: 10.220.0.2#53(10.220.0.2) ;; WHEN: Sat Jun 04 08:16:40 EDT 2022 ;; MSG SIZE rcvd: 147

Happy Saturday !!! Thanks - paul

Paul Kudla

Scom.ca Internet Services http://www.scom.ca 004-1009 Byron Street South Whitby, Ontario - Canada L1N 4S3

Toronto 416.642.7266 Main 1.866.411.7266 Fax 1.888.892.7266 Email paul@scom.ca

On 6/4/2022 6:07 AM, Marc wrote:

...

I think it is because of the domain, obviously you should only be

running a website.

...

...

-----Original Message----- From: dovecot dovecot-bounces@dovecot.org On Behalf Of hi@zakaria.website Sent: Saturday, 4 June 2022 11:15 To: Dovecot dovecot@dovecot.org Subject: Re: Replicator: Panic: data stack: Out of memory

On 2022-06-04 02:46, Ivan Jurišić wrote:

...

...

Ok a little more help : vsz_limit = 0 --> means unlimited ram for allocation, change this/try 2g etc pending avaliable ram.

I try with 524M, 1G, 2G, 4G and 8G but in any case repclicator

proces

...

...

...

got crash.

Maybe there is another service process causing OOM? e.g. check clamd, antivirus DBs tend to be quite big and in updating for sometime becomes double the size due to reloading.

Also, somtimes httpd service when using event worker, and its not tuned properly, it will cause the OOM crash to other service along itself.

actually suggestion below is a good idea

run

ps -axww (or top)

to list active processes this will give you some hints

top is better for overall memory

i also have a perl script that will show actual memory useage, free etc

utilities like this are handy to have

also i found i had to set in dovecot.conf

default_process_limit = 16384

also are you running debug ?

auth_debug = no auth_debug_passwords = no

mail_debug = no

ie set debug to = yes?

might give more detail if this is really a dovecot issue.

other background processes can eat memory

I run mailscanner for example and someone every one in a while tries to crash it!

it recovers but lord knows

mem outputs :

# mem

SYSTEM MEMORY SUMMARY: mem_used: 16GB [ 12%] Logically used memory mem_avail: + 111GB [ 87%] Logically available memory

mem_total: = 128GB [100%] Logically total memory

SYSTEM MEMORY INFORMATION: mem_wire: 13GB [ 10%] Wired: disabled for paging out mem_active: + 0GB [ 0%] Active: recently referenced mem_inactive:+ 71GB [ 57%] Inactive: recently not referenced mem_cache: + 0GB [ 0%] Cached: almost avail. for allocation mem_free: + 40GB [ 32%] Free: fully available for allocation mem_gap_vm: + 0GB [ 0%] Memory gap: UNKNOWN

mem_all: = 124GB [100%] Total real memory managed mem_gap_sys: + 3GB Memory gap: Kernel?!

mem_phys: = 127GB Total real memory available mem_gap_hw: + 0GB Memory gap: Segment Mappings?!

mem_hw: = 128GB Total real memory installed

# cat /programs/common/mem #!/usr/local/bin/perl ## ## freebsd-memory -- List Total System Memory Usage ## Copyright (c) 2003-2004 Ralf S. Engelschall rse@engelschall.com ## ## Redistribution and use in source and binary forms, with or without ## modification, are permitted provided that the following conditions ## are met: ## 1. Redistributions of source code must retain the above copyright ## notice, this list of conditions and the following disclaimer. ## 2. Redistributions in binary form must reproduce the above copyright ## notice, this list of conditions and the following disclaimer in the ## documentation and/or other materials provided with the distribution. ## ## THIS SOFTWARE IS PROVIDED BY AUTHOR AND CONTRIBUTORS ``AS IS'' AND ## ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE ## IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ## ARE DISCLAIMED. IN NO EVENT SHALL AUTHOR OR CONTRIBUTORS BE LIABLE ## FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL ## DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS ## OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) ## HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT ## LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY ## OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF ## SUCH DAMAGE. ##

# query the system through the generic sysctl(8) interface # (this does not require special priviledges) my $sysctl = {}; my $sysctl_output = /sbin/sysctl -a; foreach my $line (split(/\n/, $sysctl_output)) { if ($line =~ m/^([^:]+):\s+(.+)\s*$/s) { $sysctl->{$1} = $2; } }

# round the physical memory size to the next power of two which is # reasonable for memory cards. We do this by first determining the # guessed memory card size under the assumption that usual computer # hardware has an average of a maximally eight memory cards installed # and those are usually of equal size. sub mem_rounded { my ($mem_size) = @_; my $chip_size = 1; my $chip_guess = ($mem_size / 8) - 1; while ($chip_guess != 0) { $chip_guess >>= 1; $chip_size <<= 1; } my $mem_round = (int($mem_size / $chip_size) + 1) * $chip_size; return $mem_round; }

# determine the individual known information # NOTICE: forget hw.usermem, it is just (hw.physmem - vm.stats.vm.v_wire_count). # NOTICE: forget vm.stats.misc.zero_page_count, it is just the subset of # vm.stats.vm.v_free_count which is already pre-zeroed. my $mem_hw = &mem_rounded($sysctl->{"hw.physmem"}); my $mem_phys = $sysctl->{"hw.physmem"}; my $mem_all = $sysctl->{"vm.stats.vm.v_page_count"} * $sysctl->{"hw.pagesize"}; my $mem_wire = $sysctl->{"vm.stats.vm.v_wire_count"} * $sysctl->{"hw.pagesize"}; my $mem_active = $sysctl->{"vm.stats.vm.v_active_count"} * $sysctl->{"hw.pagesize"}; my $mem_inactive = $sysctl->{"vm.stats.vm.v_inactive_count"} * $sysctl->{"hw.pagesize"}; my $mem_cache = $sysctl->{"vm.stats.vm.v_cache_count"} * $sysctl->{"hw.pagesize"}; my $mem_free = $sysctl->{"vm.stats.vm.v_free_count"} * $sysctl->{"hw.pagesize"};

# determine the individual unknown information my $mem_gap_vm = $mem_all - ($mem_wire + $mem_active + $mem_inactive

• $mem_cache + $mem_free); my $mem_gap_sys = $mem_phys - $mem_all; my $mem_gap_hw = $mem_hw - $mem_phys;

determine logical summary information

my $mem_total = $mem_hw; my $mem_avail = $mem_inactive + $mem_cache + $mem_free; my $mem_used = $mem_total - $mem_avail;

# information annotations my $info = { "mem_wire" => 'Wired: disabled for paging out', "mem_active" => 'Active: recently referenced', "mem_inactive" => 'Inactive: recently not referenced', "mem_cache" => 'Cached: almost avail. for allocation', "mem_free" => 'Free: fully available for allocation', "mem_gap_vm" => 'Memory gap: UNKNOWN', "mem_all" => 'Total real memory managed', "mem_gap_sys" => 'Memory gap: Kernel?!', "mem_phys" => 'Total real memory available', "mem_gap_hw" => 'Memory gap: Segment Mappings?!', "mem_hw" => 'Total real memory installed', "mem_used" => 'Logically used memory', "mem_avail" => 'Logically available memory', "mem_total" => 'Logically total memory', };

# print system results printf("\n");

printf("SYSTEM MEMORY SUMMARY:\n"); printf("mem_used: %7dGB [%3d%%] %s\n", $mem_used / (1024*1024*1024), ($mem_used / $mem_total) * 100, $info->{"mem_used"}); printf("mem_avail: + %7dGB [%3d%%] %s\n", $mem_avail / (1024*1024*1024), ($mem_avail / $mem_total) * 100, $info->{"mem_avail"}); printf("-------------- ------------ ----------- ------\n"); printf("mem_total: = %7dGB [100%%] %s\n", $mem_total / (1024*1024*1024), $info->{"mem_total"});

printf("\n");

printf("SYSTEM MEMORY INFORMATION:\n"); printf("mem_wire: %7dGB [%3d%%] %s\n", $mem_wire / (1024*1024*1024), ($mem_wire / $mem_all) * 100, $info->{"mem_wire"}); printf("mem_active: + %7dGB [%3d%%] %s\n", $mem_active / (1024*1024*1024), ($mem_active / $mem_all) * 100, $info->{"mem_active"}); printf("mem_inactive:+ %7dGB [%3d%%] %s\n", $mem_inactive / (1024*1024*1024), ($mem_inactive / $mem_all) * 100, $info->{"mem_inactive"}); printf("mem_cache: + %7dGB [%3d%%] %s\n", $mem_cache / (1024*1024*1024), ($mem_cache / $mem_all) * 100, $info->{"mem_cache"}); printf("mem_free: + %7dGB [%3d%%] %s\n", $mem_free / (1024*1024*1024), ($mem_free / $mem_all) * 100, $info->{"mem_free"}); printf("mem_gap_vm: + %7dGB [%3d%%] %s\n", $mem_gap_vm / (1024*1024*1024), ($mem_gap_vm / $mem_all) * 100, $info->{"mem_gap_vm"}); printf("-------------- ------------ ----------- ------\n"); printf("mem_all: = %7dGB [100%%] %s\n", $mem_all / (1024*1024*1024), $info->{"mem_all"}); printf("mem_gap_sys: + %7dGB %s\n", $mem_gap_sys / (1024*1024*1024), $info->{"mem_gap_sys"}); printf("-------------- ------------ -----------\n"); printf("mem_phys: = %7dGB %s\n", $mem_phys / (1024*1024*1024), $info->{"mem_phys"}); printf("mem_gap_hw: + %7dGB %s\n", $mem_gap_hw / (1024*1024*1024), $info->{"mem_gap_hw"}); printf("-------------- ------------ -----------\n"); printf("mem_hw: = %7dGB %s\n", $mem_hw / (1024*1024*1024), $info->{"mem_hw"});

# print logical results

top will display something like this ?

last pid: 85373; load averages: 0.71, 0.48, 0.38

    up 72+04:51:41  08:22:11

207 processes: 1 running, 206 sleeping CPU: 1.5% user, 0.0% nice, 0.4% system, 0.0% interrupt, 98.0% idle Mem: 336M Active, 71G Inact, 139M Laundry, 13G Wired, 770M Buf, 40G Free ARC: 4319M Total, 1346M MFU, 501M MRU, 2368K Anon, 55M Header, 2414M Other 383M Compressed, 1469M Uncompressed, 3.83:1 Ratio Swap: 16G Total, 16G Free

PID USERNAME THR PRI NICE SIZE RES STATE C TIME WCPU COMMAND 1986 pgsql 1 26 0 195M 46M select 14 426:27 11.47% postgres 83810 pgsql 1 27 0 200M 171M select 13 3:15 9.19% postgres 1882 root 128 20 0 11M 3732K rpcsvc 15 29.7H 2.26% nfsd 1987 pgsql 1 20 0 195M 47M select 5 33:21 1.84% postgres 1985 root 34 21 0 141M 88M sigwai 14 72:22 1.32% named 1937 root 1 20 0 27M 15M select 15 491:36 0.90% python3.8 99555 root 1 20 0 28M 18M select 10 634:23 0.88% python3.8 1939 root 1 20 0 27M 15M select 1 939:47 0.87% python3.8 1988 pgsql 1 20 0 195M 47M select 7 6:58 0.28% postgres 1989 pgsql 1 20 0 195M 47M select 8 2:14 0.17% postgres 1964 pgsql 1 20 0 194M 164M select 9 10:02 0.08% postgres 85373 root 1 20 0 14M 3644K CPU0 0 0:00 0.07% top 3150 pgsql 1 20 0 195M 42M select 6 39:21 0.06% postgres

ps -axw or ps -axww or freebsd

# ps -axww PID TT STAT TIME COMMAND 0 - DLs 3788:48.94 [kernel] 1 - ILs 0:05.38 /sbin/init -- 2 - DL 0:00.00 [crypto] 3 - DL 0:00.00 [crypto returns 0] 4 - DL 0:00.00 [crypto returns 1] 5 - DL 0:00.00 [crypto returns 2] 6 - DL 0:00.00 [crypto returns 3] 7 - DL 0:00.00 [crypto returns 4] 8 - DL 0:00.00 [crypto returns 5] 9 - DL 0:00.00 [crypto returns 6] 10 - DL 0:00.00 [audit] 11 - RNL 1629112:33.34 [idle] 12 - WL 180:00.70 [intr] 13 - DL 123:57.70 [geom] 14 - DL 0:00.00 [crypto returns 7] 15 - DL 0:00.00 [crypto returns 8] 16 - DL 0:00.00 [crypto returns 9] 17 - DL 0:00.00 [crypto returns 10] 18 - DL 0:00.00 [crypto returns 11] 19 - DL 0:00.00 [crypto returns 12] 20 - DL 0:00.00 [crypto returns 13] 21 - DL 0:00.00 [crypto returns 14] 22 - DL 0:00.00 [crypto returns 15] 23 - DL 0:00.00 [sequencer 00] 24 - DL 0:00.00 [cam] 25 - DL 5:42.32 [usb] 26 - DL 0:00.47 [soaiod1] 27 - DL 0:00.47 [soaiod2] 28 - DL 0:00.46 [soaiod3] 29 - DL 0:00.47 [soaiod4] 30 - DL 1714:58.15 [zfskern] 31 - DL 0:00.00 [sctp_iterator] 32 - DL 12:50.77 [pf purge] 33 - DL 2:16.82 [rand_harvestq] 34 - DL 29:00.62 [pagedaemon] 35 - DL 0:00.00 [vmdaemon] 36 - DL 5:25.68 [bufdaemon] 37 - DL 1:44.98 [vnlru] 38 - DL 2040:33.82 [syncer] 1657 - Is 0:01.21 /sbin/devd 1863 - Ss 0:03.44 /usr/sbin/rpcbind 1878 - Is 0:00.08 /usr/sbin/mountd -r -S 1880 - Is 0:00.27 nfsd: master (nfsd) 1882 - S 1780:23.16 nfsd: server (nfsd) 1907 - Ss 10:01.06 /usr/sbin/syslogd -s 1909 - Is 0:00.55 /usr/sbin/inetd -wW -C 50 -s 500 1911 - Is 0:00.25 /usr/sbin/sshd 1955 - Is 24:50.70 /usr/local/sbin/clamd 1964 - Ss 10:02.28 postmaster: checkpointer (postgres) 1965 - Ss 1:38.52 postmaster: background writer (postgres) 1966 - Ss 3:48.60 postmaster: walwriter (postgres) 1967 - Ss 2:03.84 postmaster: autovacuum launcher (postgres) 1968 - Ss 12:41.60 postmaster: stats collector (postgres) 1969 - Is 0:01.82 postmaster: logical replication launcher (postgres) 1974 - Ss 37:19.26 postmaster: walsender pgsql 10.221.0.16(30421) (postgres) 1976 - Ss 39:37.29 postmaster: walsender pgsql 10.221.0.10(64872) (postgres) 1985 - Is 72:21.96 /usr/local/sbin/named -d 0 -4 1986 - Ss 426:29.15 postmaster: pgsql scom_billing 10.221.0.18(52852) (postgres) 1987 - Ss 33:21.50 postmaster: pgsql scom_billing 10.221.0.18(60830) (postgres) 1988 - Ss 6:57.70 postmaster: pgsql scom_billing 10.221.0.18(34255) (postgres) 1989 - Ss 2:13.52 postmaster: pgsql scom_billing 10.221.0.18(17265) (postgres) 2073 - Ss 10:12.46 /usr/local/libexec/postfix/master -w 2076 - I 0:07.82 qmgr -l -t fifo -u 2166 - Is 1:53.61 /usr/local/libexec/postfix/master -w 2168 - I 0:55.23 qmgr -l -t fifo -u 2238 - Is 1:49.77 /usr/local/libexec/postfix/master -w 2240 - I 1:01.17 qmgr -l -t fifo -u 2253 - I 0:39.34 tlsmgr -l -t unix -u 2397 - Is 0:05.58 MailScanner: starting child (perl) 2513 - Is 0:20.43 /usr/sbin/cron -s 3150 - Rs 39:21.01 postmaster: walsender pgsql 10.221.0.6(10000) (postgres) 3175 - Is 0:00.35 postmaster: pgsql scom_billing 10.221.0.6(10017) (postgres) 3176 - Is 0:10.80 postmaster: pgsql scom_billing 10.221.0.6(10018) (postgres) 3177 - Ss 1:10.22 postmaster: pgsql scom_billing 10.221.0.6(10019) (postgres)

Happy Saturday !!! Thanks - paul

Paul Kudla

Scom.ca Internet Services http://www.scom.ca 004-1009 Byron Street South Whitby, Ontario - Canada L1N 4S3

Toronto 416.642.7266 Main 1.866.411.7266 Fax 1.888.892.7266 Email paul@scom.ca

On 6/4/2022 5:15 AM, dovecot-bounces@dovecot.org wrote:

On 2022-06-04 02:46, Ivan Jurišić wrote:

...

...

Ok a little more help : vsz_limit = 0 --> means unlimited ram for allocation, change this/try 2g etc pending avaliable ram.

I try with 524M, 1G, 2G, 4G and 8G but in any case repclicator proces got crash.

Maybe there is another service process causing OOM? e.g. check clamd, antivirus DBs tend to be quite big and in updating for sometime becomes double the size due to reloading.

Also, somtimes httpd service when using event worker, and its not tuned properly, it will cause the OOM crash to other service along itself.

Good luck.

Zakaria.

Sry, I check my logs and this configuration of replication is not help.

Jun 06 12:05:40 doveadm(ivan@myserv.local)<2864367><O9xlAHTRnWLvtCsAeFl1HA>: Panic: data stack: Out of memory when allocating 134217768 bytes Jun 06 12:05:40 doveadm(ivan@myserv.local)<2864367><O9xlAHTRnWLvtCsAeFl1HA>: Error: Raw backtrace: /usr/lib/dovecot/libdovecot.so.0(backtrace_append+0x42) [0x7f722cdb6582] -> /usr/lib/dovecot/libdovecot.so.0(backtrace_get+0x1e) [0x7f722cdb669e] -> /usr/lib/dovecot/libdovecot.so.0(+0x1022fb) [0x7f722cdc32fb] -> /usr/lib/dovecot/libdovecot.so.0(+0x102391) [0x7f722cdc3391] -> /usr/lib/dovecot/libdovecot.so.0(+0x55589) [0x7f722cd16589] -> /usr/lib/dovecot/libdovecot.so.0(+0x54d52) [0x7f722cd15d52] -> /usr/lib/dovecot/libdovecot.so.0(+0xfb808) [0x7f722cdbc808] -> /usr/lib/dovecot/libdovecot.so.0(t_strdup_until+0x21) [0x7f722cdf95c1] -> dovecot/doveadm-server(dsync_mailbox_tree_find_delete+0x1a5) [0x5622f20c6e05] -> dovecot/doveadm-server(+0x4fe05) [0x5622f20bbe05] -> dovecot/doveadm-server(dsync_brain_recv_mailbox_tree_deletes+0xd8) [0x5622f20bcd68] -> dovecot/doveadm-server(dsync_brain_run+0x1ce) [0x5622f20b938e] -> dovecot/doveadm-server(+0x4da85) [0x5622f20b9a85] -> dovecot/doveadm-server(+0x5f3cf) [0x5622f20cb3cf] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_call_io+0x69) [0x7f722cdd9529] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_handler_run_internal+0x132) [0x7f722cddac12] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_handler_run+0x50) [0x7f722cdd95d0] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_run+0x40) [0x7f722cdd9790] -> dovecot/doveadm-server(+0x31195) [0x5622f209d195] -> dovecot/doveadm-server(+0x33a05) [0x5622f209fa05] -> dovecot/doveadm-server(doveadm_cmd_ver2_to_mail_cmd_wrapper+0x2ca) [0x5622f20a0aba] -> dovecot/doveadm-server(doveadm_cmd_run_ver2+0x501) [0x5622f20b1161] -> dovecot/doveadm-server(+0x495aa) [0x5622f20b55aa] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_call_io+0x69) [0x7f722cdd9529] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_handler_run_internal+0x132) [0x7f722cddac12] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_handler_run+0x50) [0x7f722cdd95d0] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_run+0x40) [0x7f722cdd9790] -> /usr/lib/dovecot/libdovecot.so.0(master_service_run+0x13) [0x7f722cd4c353] -> dovecot/doveadm-server(main+0x102) [0x5622f208ffc2] -> /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xea) [0x7f722c9ded0a]

Dana 06.06.2022 10:03, Ivan Jurišić je napisao(la):

...

Dana 04.06.2022 15:34, Paul Kudla (SCOM.CA Internet Services Inc.) je napisao(la): ok thanks for the info from here you need to turn on full debugging and then filter the log by "replicat"

Now replication work when set vsz_limit in service aggregator and remove parametar replication_dsync_parameters and replication_full_sync_interval from my 90-replicator.conf. Now my configuration work for replication on another mail server.

Config file for replication /etc/dovecot/conf.d/90-replicator.conf

service aggregator { vsz_limit = 256M fifo_listener replication-notify-fifo { user = vmail } unix_listener replication-notify { user = vmail } }

service replicator { process_min_avail = 1 unix_listener replicator-doveadm { mode = 0600 user = vmail } }

service doveadm { inet_listener { port = 12345 ssl = no } }

replication_max_conns = 100 #replication_dsync_parameters = -d -N -l 30 -U #replication_full_sync_interval = 1 days

doveadm_port = 12345 doveadm_password = Jados82!

plugin { mail_replica = tcp:imap.myserv2.local:12345 }

Dana 06.06.2022 14:14, Paul Kudla (SCOM.CA Internet Services Inc.) je napisao(la):

did you get this figured out / working getting posts out of order - just wanted to make sure your ok?

Replication not work, now have problem with "data stack", error is:

Jun 06 15:49:32 doveadm(ivan@myserv.local)<79253>: Panic: data stack: Out of memory when allocating 536870952 bytes

I have lots free memory and not sure what is problem?

# free -m total used free shared buff/cache
available Mem: 7956 2685 4966 26 305
5151 Swap: 7629 0 7629

ok thanks for the info

from here you need to turn on full debugging and then filter the log by

"replicat"

It is starting to look like you have a bad email somewhere causing the crash

I have yet to see this however anything is possible, when you replicate anything it needs to update indexes & control files on the other side and vice versa

that being said its a good place to start looking?

At least see in the debug mail logs what it is trying to replicate and where it fails.

Im running a large email server and in 2.3.18 there was a bug that any account over 300+ folders would fail to replicate.

This appears to be fixed in 2.3.19

other then that TCP replication etc seems to work the best

if you look back on the mailing lists i have uploaded a sample config of my systems which seems to work ok.

Happy Saturday !!! Thanks - paul

Paul Kudla

Scom.ca Internet Services http://www.scom.ca 004-1009 Byron Street South Whitby, Ontario - Canada L1N 4S3

Toronto 416.642.7266 Main 1.866.411.7266 Fax 1.888.892.7266 Email paul@scom.ca

On 6/3/2022 10:46 PM, Ivan Jurišić wrote:

...

Ok a little more help : vsz_limit = 0 --> means unlimited ram for allocation, change this/try 2g etc pending avaliable ram.

I try with 524M, 1G, 2G, 4G and 8G but in any case repclicator proces got crash.

-- This message has been scanned for viruses and dangerous content by *MailScanner* http://www.mailscanner.info/, and is believed to be clean.