Details
-
Bug
-
Status: Closed (View Workflow)
-
Major
-
Resolution: Not a Bug
-
None
-
None
-
None
-
9.0-RELEASE-p4 amd64
Description
After upgrade from 5.3 to 5.5 MariaDB starts crash wish signal 10 about every hour, max uptime 20h. It was on 5.5.29 and nothing changed with 5.5.30.
Backtrace always the same in gdb:
(gdb) bt
|
#0 0x00000008022fb81e in _malloc_prefork () from /lib/libc.so.7
|
#1 0x0000000802300bd3 in free () from /lib/libc.so.7
|
#2 0x00000000009b2206 in free_root ()
|
#3 0x000000000056c9cf in dispatch_command ()
|
#4 0x000000000056dfe3 in do_command ()
|
#5 0x0000000000613072 in do_handle_one_connection ()
|
#6 0x00000000006131aa in handle_one_connection ()
|
#7 0x0000000802066274 in pthread_getprio () from /lib/libthr.so.3
|
#8 0x00007fffeb044000 in ?? ()
|
General log finishes with different queries before crash, Init DB or simple select.
We have 2 servers with Maria for HA with master-master, and this problem on both. Also have another server with another data and with the same problem.
For a first time we have used to use Aria SE, then converted to InnoDB/ExtaDB, but no effect on the problem.
What additional helpful info can I provide for you?