[MDEV-8820] After package update on Ubuntu 14.04, "mysqld 10.0.21-MariaDB-1~trusty-wsrep" crashes Created: 2015-09-20  Updated: 2018-03-12  Resolved: 2018-03-12

Status: Closed
Project: MariaDB Server
Component/s: Galera, Platform Debian
Affects Version/s: 10.0.21-galera
Fix Version/s: N/A

Type: Bug Priority: Major
Reporter: Christian Hoene Assignee: Sachin Setiya (Inactive)
Resolution: Incomplete Votes: 0
Labels: need_feedback
Environment:

Distributor ID: Ubuntu
Description: Ubuntu 14.04.3 LTS
Release: 14.04
Codename: trusty

Linux 3.13.0-63-generic #103-Ubuntu SMP Fri Aug 14 21:42:59 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux


Attachments: HTML File stacktrace     File term.log     Text File x.txt    

 Description   

after yesterdays "apt-get update", mysqld starts with a core dump.

Mariadb itself has not been updated, thus I assume it has problems with some newer dependencies (see attachment for the packages that have been installed).



 Comments   
Comment by Elena Stepanova [ 2015-09-20 ]

eneho

Could you please paste the stack trace that you are getting?

Comment by Christian Hoene [ 2015-09-20 ]

are the gdb output and the log file output enough?

Comment by Elena Stepanova [ 2015-09-20 ]

For now yes, thanks.
I've attached the gdb output you sent to me by email – don't know what JIRA did not like about the file, it would through the internal error for me as well. I've created a new one with the same contents, and then it worked.

Comment by Elena Stepanova [ 2015-09-20 ]

Thread 4 (Thread 0x7fffe87ff700 (LWP 11908)):
#0  0x00007ffff6391cc9 in __GI_raise (sig=sig@entry=6) at ../nptl/sysdeps/unix/sysv/linux/raise.c:56
#1  0x00007ffff63950d8 in __GI_abort () at abort.c:89
#2  0x00007ffff4be7547 in gu_abort () from /usr/lib/galera/libgalera_smm.so
#3  0x00007ffff4cf3a9b in gcs_core_recv(gcs_core*, gcs_act_rcvd*, long long) () from /usr/lib/galera/libgalera_smm.so
#4  0x00007ffff4cf7646 in ?? () from /usr/lib/galera/libgalera_smm.so
#5  0x00007ffff6d32182 in start_thread (arg=0x7fffe87ff700) at pthread_create.c:312
#6  0x00007ffff645547d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:111

More details in the attachment.

Comment by Nirbhay Choubey (Inactive) [ 2015-11-19 ]

I tried upgrading from 10.0.20 -> 10.0.21 on Trusty but did not see a similar crash.

Comment by Nirbhay Choubey (Inactive) [ 2015-11-19 ]

I have reported this to Codership for further analysis.

Comment by Nirbhay Choubey (Inactive) [ 2015-11-19 ]

https://github.com/codership/galera/issues/366

Comment by Nirbhay Choubey (Inactive) [ 2015-11-19 ]

eneho Looks like something interesting happened on the donor. Do you have donor's error log around the time of crash?

150920 12:06:34 [Note] WSREP: (43d6b2e3, 'tcp://0.0.0.0:4567') turning message relay requesting off
150920 12:08:39 [Warning] WSREP: 0.0 (tuebingen02): State transfer to 1.0 (cologne00.symonics.com) failed: -255 (Unknown error 255)
150920 12:08:39 [ERROR] WSREP: gcs/src/gcs_group.cpp:int gcs_group_handle_join_msg(gcs_group_t*, const gcs_recv_msg_t*)():731: Will never receive state. Need to abort.

Generated at Thu Feb 08 07:30:02 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.