[MDEV-31074] Galera SST fails when datadir / logdir are symlinks Created: 2023-04-18  Updated: 2023-10-13  Resolved: 2023-10-13

Status: Closed
Project: MariaDB Server
Component/s: Galera, Galera SST
Affects Version/s: 10.11.2
Fix Version/s: 10.4.32, 10.5.23, 10.6.16, 10.10.7, 10.11.6, 11.0.4, 11.1.3, 11.2.2, 11.3.1

Type: Bug Priority: Critical
Reporter: Rick Tuk Assignee: Julius Goryavsky
Resolution: Fixed Votes: 1
Labels: None
Environment:

Ubuntu 20.04 LTS


Issue Links:
Duplicate
duplicates MDEV-29893 SST fails when having datadir set to ... Closed
Relates
relates to MDEV-28669 Cannot join second node to new cluste... Closed

 Description   

On our servers the datadir is configured as /var/lib/mysql, the log is configured as /var/log/mysql, both directories are symlinks:

root@node02:~# ls -al /var/lib/mysql
lrwxrwxrwx 1 root root 15 Apr 4 16:31 /var/lib/mysql -> /database/mysql

root@node02:~# ls -al /var/log/mysql
lrwxrwxrwx 1 root root 13 Apr 4 09:19 /var/log/mysql -> /database/log

when adding a second node to the cluster the SST fails:

Apr 4 16:35:14 node02 mariadbd[6833]: 2023-04-04 16:35:14 0 [ERROR] WSREP: Failed to read uuid:seqno and wsrep_gtid_domain_id from joiner script.
Apr 4 16:35:14 node02 mariadbd[6833]: 2023-04-04 16:35:14 0 [ERROR] WSREP: Process completed with error: wsrep_sst_mariabackup --role 'joiner' --address '10.97.148.12' --datadir '/var/lib/mysql/' --parent 6833 --progress 0 --binlog '/var/log/mysql/mysql-bin' --binlog-index '/var/log/mysql/mariadb-bin.index' --mysqld-args --wsrep_start_position=00000000-0000-0000-0000-000000000000:-1: 2 (No such file or directory)
Apr 4 16:35:14 node02 mariadbd[6833]: 2023-04-04 16:35:14 3 [Note] WSREP: SST received
Apr 4 16:35:14 node02 mariadbd[6833]: 2023-04-04 16:35:14 3 [Note] WSREP: SST received: 00000000-0000-0000-0000-000000000000:-1
Apr 4 16:35:14 node02 mariadbd[6833]: 2023-04-04 16:35:14 3 [Note] WSREP: SST succeeded for position 00000000-0000-0000-0000-000000000000:-1
Apr 4 16:35:14 node02 mariadbd[6833]: 2023-04-04 16:35:14 0 [Note] WSREP: Joiner monitor thread ended with total time 15 sec
Apr 4 16:35:14 node02 mariadbd[6833]: 2023-04-04 16:35:14 2 [ERROR] WSREP: Application received wrong state:
Apr 4 16:35:14 node02 mariadbd[6833]: #011Received: 00000000-0000-0000-0000-000000000000
Apr 4 16:35:14 node02 mariadbd[6833]: #011Required: b9e3e449-d2da-11ed-ae66-72a327b8e5ab
Apr 4 16:35:14 node02 mariadbd[6833]: 2023-04-04 16:35:14 2 [ERROR] WSREP: Application state transfer failed. This is unrecoverable condition, restart required.

This issue was already reported as MDEV-28669 and has been marked as fixed.
After testing the fixed version I've added a comment with the above results 2 weeks ago, since the issue is closed, my comment may have not been seen, hence the registration of a new issue.



 Comments   
Comment by Khai Ping [ 2023-06-09 ]

it happens to our clusters in 10.6.12 as well.

Comment by Julius Goryavsky [ 2023-10-13 ]

Fixed, duplicate of https://jira.mariadb.org/browse/MDEV-29893

Generated at Thu Feb 08 10:21:05 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.