Details
-
Type:
Bug
-
Status: Open
-
Priority:
Major
-
Resolution: Unresolved
-
Affects Version/s: 5.5.41-galera
-
Fix Version/s: None
-
Component/s: Galera
-
Labels:None
Description
Hello,
I am running two MariaDB servers, clustered using Galera. After upgrading to `5.5.41-MariaDB-1~wheezy-wsrep-log` I am having the following issue:
- The database nodes will become unreachable with the error `WSREP has not yet prepared node for application use`.
- This will occur approximately every 3 - 4 hours despite very low load on the servers.
- Sometimes they come back online after 10 - 15 minutes, sometimes they crash until I reboot both nodes in the cluster.
In the syslog at the exact time the server becomes unavailable, I see this error:
May 22 22:54:56 dbserver-uk mysqld: }150522 22:54:56 [ERROR] WSREP: exception from gcomm, backend must be restarted: f0af4e14-0053-11e5-9aec-2a044cd57fd7Install message self state does not match, message state: prim=0,un=0,last_seq=2,last_prim=view_id(PRIM,f0af4e14-0053-11e5-9aec-2a044cd57fd7,94),to_seq=375297,weight=1,segment=0, local state: prim=0,un=1,last_seq=2,last_prim=view_id(PRIM,f0af4e14-0053-11e5-9aec-2a044cd57fd7,94),to_seq=375297,weight=1,segment=0 (FATAL)
I am running both nodes on Debian 7 with the latest packages. Please let me know if there is any more information I can apply.
Gliffy Diagrams
Attachments
Activity
- All
- Comments
- Work Log
- History
- Activity
- Transitions
5.5.41-galera uses a relatively older galera library version (3.6). The later versions have 3.9. Can you try upgrading MariaDB and see if you still experience this issue?