xtrabackup's --parallel option asserts / crashes with a value of -1

Reported by Patrick Crews on 2011-11-01
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Percona XtraBackup
Low
Alexey Kopytov
1.6
Low
Alexey Kopytov
2.0
Low
Alexey Kopytov

Bug Description

Passing a value of -1 to --parallel results in a nasty little crash. Admittedly this is a bit of a gotcha, but I had to test it ; )
innobackupex: Starting ibbackup with command: /percona-xtrabackup/mysql-5.5/storage/innobase/xtrabackup/xtrabackup_innodb55 --defaults-file="/dbqp/workdir/bot0/s0/var/my.cnf" --backup --suspend-at-end --target-dir=/dbqp/workdir/bot0/s0/var/_xtrabackup --parallel=-1
20111101-091107 innobackupex: Waiting for ibbackup (pid=30593) to suspend
20111101-091107 innobackupex: Suspend file '/dbqp/workdir/bot0/s0/var/_xtrabackup/xtrabackup_suspended'
20111101-091107
20111101-091107 Warning: option 'parallel': unsigned value 18446744073709551615 adjusted to 4294967295
20111101-091107 xtrabackup: uses posix_fadvise().
20111101-091107 xtrabackup: cd to /dbqp/workdir/bot0/s0/var/master-data
20111101-091107 xtrabackup: Target instance is assumed as followings.
20111101-091107 xtrabackup: innodb_data_home_dir = ./
20111101-091107 xtrabackup: innodb_data_file_path = ibdata1:10M:autoextend
20111101-091107 xtrabackup: innodb_log_group_home_dir = ./
20111101-091107 xtrabackup: innodb_log_files_in_group = 2
20111101-091107 xtrabackup: innodb_log_file_size = 5242880
20111101-091107 111101 9:11:04 InnoDB: Using Linux native AIO
20111101-091107 >> log scanned up to (1595675)
20111101-091107 111101 9:11:04 InnoDB: Assertion failure in thread 140139762509600 in file /percona-xtrabackup/mysql-5.5/storage/innobase/ut/ut0mem.c line 107
20111101-091107 InnoDB: Failing assertion: ret || !assert_on_error
20111101-091107 InnoDB: We intentionally generate a memory trap.
20111101-091107 InnoDB: Submit a detailed bug report to http://bugs.mysql.com.
20111101-091107 InnoDB: If you get repeated assertion failures or crashes, even
20111101-091107 InnoDB: immediately after the mysqld startup, there may be
20111101-091107 InnoDB: corruption in the InnoDB tablespace. Please refer to
20111101-091107 InnoDB: http://dev.mysql.com/doc/refman/5.5/en/forcing-innodb-recovery.html
20111101-091107 InnoDB: about forcing recovery.
20111101-091107 /percona-xtrabackup/mysql-5.5/storage/innobase/xtrabackup/xtrabackup_innodb55 version 1.6.3 for MySQL server 5.5.10 Linux (x86_64) (revision id: undefined)
20111101-091107 xtrabackup: Starting 4294967295 threads for parallel data files transfer
20111101-091107 Aborted
20111101-091107 innobackupex: Error: ibbackup child process has died at /percona-xtrabackup/innobackupex line 354.

Patrick Crews (patrick-crews) wrote :

To repeat via dbqp:
 ./dbqp.py --default-server-type=mysql --basedir=/mysql-5.5 --xtrabackup-path=/percona-xtrabackup/mysql-5.5/storage/innobase/xtrabackup/xtrabackup_innodb55 --innobackupex-path=/percona-xtrabackup/innobackupex --suite=xtrabackup_basic bug884737_test

Patrick Crews (patrick-crews) wrote :

--parallel=0 works a-ok.
When it is 0, the option seems to be dropped from the xtrabackup call.
We might wish to document this, even though nobody should ever *really* try this (except devious QA guys...) : )

Stewart Smith (stewart) on 2011-11-15
Changed in percona-xtrabackup:
status: New → Fix Committed
assignee: nobody → Alexey Kopytov (akopytov)
importance: Undecided → Low
milestone: none → 1.6.4
Changed in percona-xtrabackup:
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