Merge lp:~sergei.glushchenko/percona-xtrabackup/2.2-xb-bug1214274 into lp:percona-xtrabackup/2.2

Proposed by Sergei Glushchenko on 2013-10-21
Status: Merged
Approved by: Alexey Kopytov on 2013-10-30
Approved revision: 4880
Merged at revision: 4882
Proposed branch: lp:~sergei.glushchenko/percona-xtrabackup/2.2-xb-bug1214274
Merge into: lp:percona-xtrabackup/2.2
Diff against target: 38 lines (+12/-2)
1 file modified
xtrabackup/test/t/kill_long_selects.sh (+12/-2)
To merge this branch: bzr merge lp:~sergei.glushchenko/percona-xtrabackup/2.2-xb-bug1214274
Reviewer Review Type Date Requested Status
Alexey Kopytov (community) 2013-10-21 Approve on 2013-10-30
Review via email: mp+191994@code.launchpad.net
To post a comment you must log in.
Alexey Kopytov (akopytov) :
review: Approve

Preview Diff

[H/L] Next/Prev Comment, [J/K] Next/Prev File, [N/P] Next/Prev Hunk
1=== modified file 'xtrabackup/test/t/kill_long_selects.sh'
2--- xtrabackup/test/t/kill_long_selects.sh 2013-05-26 07:18:14 +0000
3+++ xtrabackup/test/t/kill_long_selects.sh 2013-10-21 14:08:09 +0000
4@@ -68,7 +68,10 @@
5
6 function kill_all_queries()
7 {
8- run_cmd $MYSQL $MYSQL_ARGS test <<EOF
9+ # we really want to ignore errors
10+ # some connections can die by themselves
11+ # between SELECT and killall
12+ run_cmd $MYSQL $MYSQL_ARGS --force --batch test <<EOF
13 select concat('KILL ',id,';') from information_schema.processlist
14 where user='root' and time > 2 into outfile '$MYSQLD_TMPDIR/killall.sql';
15 source $MYSQLD_TMPDIR/killall.sql;
16@@ -90,7 +93,9 @@
17 bg_run bg_select_pid "mysql_select 3"
18 bg_run bg_update_pid "sleep 1" "mysql_update 3"
19
20-innobackupex $topdir/full --kill-long-queries-timeout=5 \
21+sleep 8
22+
23+innobackupex $topdir/full --kill-long-queries-timeout=10 \
24 --kill-long-query-type=all
25
26 bg_wait_ok $bg_select_pid
27@@ -99,6 +104,11 @@
28
29
30 # ==============================================================
31+# this case can succeed for two reasons
32+# 1. update was waited for by innobackupex an finished successfully
33+# by the time of kill_all_queries
34+# 2. update was stared late and did not block FTWRL but was locked by
35+# it. then it may be still alive by the time of kill_all_queries
36 vlog "===================== case 2 ====================="
37 bg_run bg_select_pid "mysql_select 200"
38 bg_run bg_update_pid "sleep 1" "mysql_update 5"

Subscribers

People subscribed via source and target branches