POT crash on handle_connections_sockets()/libc.poll()

Bug #791163 reported by Valentine Gostev on 2011-06-01
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
percona-projects-qa
Critical
Laurynas Biveinis

Bug Description

While running rpl.rpl_mix_insert_delayed 'mix' both mysqld instances crashed on same calls. Please find test log attached.

Valentine Gostev (longbow) wrote :
Valentine Gostev (longbow) wrote :

mysqld.1
(gdb) bt
#0 0x0000003fccc0bd02 in pthread_kill () from /lib64/libpthread.so.0
#1 0x00000000008c853f in my_write_core (sig=6) at /mnt/x25m/gval/r3487/mysys/stacktrace.c:423
#2 0x00000000005395a7 in handle_segfault (sig=6) at /mnt/x25m/gval/r3487/sql/mysqld.cc:2546
#3 <signal handler called>
#4 0x0000003fcc4cb186 in poll () from /lib64/libc.so.6
#5 0x0000000000539f1f in handle_connections_sockets () at /mnt/x25m/gval/r3487/sql/mysqld.cc:5198
#6 0x000000000053e936 in mysqld_main (argc=49, argv=0xe89fa80) at /mnt/x25m/gval/r3487/sql/mysqld.cc:4638
#7 0x0000000000533383 in main (argc=10, argv=0x7fff0e1a2588) at /mnt/x25m/gval/r3487/sql/main.cc:24

mysqld.2
(gdb) bt
#0 0x0000003fccc0bd02 in pthread_kill () from /lib64/libpthread.so.0
#1 0x00000000008c853f in my_write_core (sig=6) at /mnt/x25m/gval/r3487/mysys/stacktrace.c:423
#2 0x00000000005395a7 in handle_segfault (sig=6) at /mnt/x25m/gval/r3487/sql/mysqld.cc:2546
#3 <signal handler called>
#4 0x0000003fcc4cb186 in poll () from /lib64/libc.so.6
#5 0x0000000000539f1f in handle_connections_sockets () at /mnt/x25m/gval/r3487/sql/mysqld.cc:5198
#6 0x000000000053e936 in mysqld_main (argc=58, argv=0x2d34b50) at /mnt/x25m/gval/r3487/sql/mysqld.cc:4638
#7 0x0000000000533383 in main (argc=10, argv=0x7fff23e45838) at /mnt/x25m/gval/r3487/sql/main.cc:24

Changed in percona-projects-qa:
milestone: none → 5.5.13-eb
Valentine Gostev (longbow) wrote :

core.28906 - mysqld.1
core.28911 - mysqld.2

Changed in percona-projects-qa:
status: New → Confirmed
assignee: nobody → Laurynas Biveinis (laurynas-biveinis)
Changed in percona-projects-qa:
status: Confirmed → In Progress
importance: Undecided → Critical
Changed in percona-projects-qa:
status: In Progress → Fix Committed
Alexey Kopytov (akopytov) wrote :

"handle_segfault (sig=6)"

I think it was actually an assertion failure in another thread. So backtraces and poll() have nothing to do with the actual reason of the crash.

Just something to keep in mind in we see anything similar once again.

Changed in percona-projects-qa:
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers