[MCOL-4311] MariaDB 10.5 with Columnstore keeps crashing with "buffer overflow detected" Created: 2020-09-16  Updated: 2023-11-25  Resolved: 2023-11-11

Status: Closed
Project: MariaDB ColumnStore
Component/s: MDB Plugin
Affects Version/s: 1.5.3
Fix Version/s: Icebox

Type: Bug Priority: Major
Reporter: David Hill (Inactive) Assignee: Unassigned
Resolution: Incomplete Votes: 1
Labels: crash
Environment:

Single node


 Description   

we installed the new MariaDB 10.5 with Columnstore to substitute old 1.2,
and since the 9th of september we already had 4 crashes, every time blocking our production.
I'll attach:

  • server conf file
  • error log
  • output of "show status"

Support report in case.

Request statement causing issue, customer didnt know. They are trying to find out.

Mysqld crash is in log in case

 Comments   
Comment by David Hill (Inactive) [ 2020-09-16 ]

It is possible that mysqld could use up to
key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 2336511 K bytes of memory
Hope that's ok; if not, decrease some variables in the equation.

Thread pointer: 0x7f2d040008d8
Attempting backtrace. You can use the following information to find out
where mysqld died. If you see no messages after this, something went
terribly wrong...
stack_bottom = 0x7f2dc409bc90 thread_stack 0x49000

      • buffer overflow detected ***: /usr/sbin/mariadbd terminated
        ======= Backtrace: =========
        /lib64/libc.so.6(__fortify_fail+0x37)[0x7f3292065577]
        /lib64/libc.so.6(+0x1166f2)[0x7f32920636f2]
        /lib64/libc.so.6(+0x1184d7)[0x7f32920654d7]
        /usr/sbin/mariadbd(my_addr_resolve+0xd9)[0x55fe6153ae09]
        /usr/sbin/mariadbd(my_print_stacktrace+0x1c2)[0x55fe61522542]
        /usr/sbin/mariadbd(handle_fatal_signal+0x307)[0x55fe60efc3e7]
        /lib64/libpthread.so.0(+0xf630)[0x7f3293eba630]
        /usr/lib64/mysql/plugin/ha_columnstore.so(_ZN11cal_impl_if12processWhereER13st_select_lexRNS_12gp_walk_infoERN5boost10shared_ptrIN8execplan26CalpontSelectExecutionPlanEEER4ListI4ItemERKSt6vectorIPSB_SaISF_EE+0x753)[0x7f3285c162d3]
        /usr/lib64/mysql/plugin/ha_columnstore.so(_ZN11cal_impl_if13getSelectPlanERNS_12gp_walk_infoER13st_select_lexRN5boost10shared_ptrIN8execplan26CalpontSelectExecutionPlanEEEbbRKSt6vectorIP4ItemSaISC_EE+0x1fb)[0x7f3285c292eb]
        /usr/lib64/mysql/plugin/ha_columnstore.so(+0x8238f)[0x7f3285bd538f]
        /usr/lib64/mysql/plugin/ha_columnstore.so(_Z20ha_mcs_impl_rnd_initP5TABLERKSt6vectorIP4ItemSaIS3_EE+0x756)[0x7f3285bdc766]
        /usr/lib64/mysql/plugin/ha_columnstore.so(_ZN6ha_mcs8rnd_initEb+0x29)[0x7f3285bc4729]
        /usr/sbin/mariadbd(_ZN7handler22ha_rnd_init_with_errorEb+0x17)[0x55fe60f03257]
        /usr/sbin/mariadbd(_Z16init_read_recordP11READ_RECORDP3THDP5TABLEP10SQL_SELECTP9SORT_INFOibb+0x599)[0x55fe61041119]
        /usr/sbin/mariadbd(_Z12mysql_deleteP3THDP10TABLE_LISTP4ItemP10SQL_I_ListI8st_orderEyyP13select_result+0xfb3)[0x55fe61058833]
        /usr/sbin/mariadbd(_Z21mysql_execute_commandP3THD+0x3285)[0x55fe60cfe665]
        /usr/sbin/mariadbd(_Z11mysql_parseP3THDPcjP12Parser_statebb+0x362)[0x55fe60d01c22]
        /usr/sbin/mariadbd(_Z16dispatch_command19enum_server_commandP3THDPcjbb+0x1bec)[0x55fe60d03fbc]
        /usr/sbin/mariadbd(_Z10do_commandP3THD+0x10b)[0x55fe60d0596b]
        /usr/sbin/mariadbd(_Z24do_handle_one_connectionP7CONNECTb+0x1d9)[0x55fe60df0109]
        /usr/sbin/mariadbd(handle_one_connection+0x34)[0x55fe60df03a4]
        /usr/sbin/mariadbd(+0xbdbc3d)[0x55fe6116ec3d]
        /lib64/libpthread.so.0(+0x7ea5)[0x7f3293eb2ea5]
        /lib64/libc.so.6(clone+0x6d)[0x7f329204b8dd]
Comment by Roman [ 2023-11-11 ]

No way to reproduce.
The crash trace tells me this has been fixed with 23.10.0.

Generated at Thu Feb 08 02:49:24 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.