rpms/mysql8.4
7
0
Fork 0
Code Issues Pull Requests Releases Activity

Add code to cleanup leftover mysql daemon processes --- should make

Browse Source 
brew/beehive builds more robust when retrying after a failure.
This commit is contained in:
Tom Lane 2006-06-10 17:43:55 +00:00 committed by Michal Schorm
parent 6d57e80587
commit 96ebf13bb4
2 changed files with 30 additions and 26 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

50
mysql-testing.patch
View File

@ -2,7 +2,10 @@ This patch is needed to improve the probability that the MySQL tests pass
in Red Hat's build environment. The 32-bit and 64-bit variants of a single
architecture may be built on the same machine at the same time, so we have
to ensure that the tests use different port numbers on each arch, else they
will interfere with each other.
will interfere with each other. Also, it's been observed that sometimes
the build environment fails to kill a test mysql daemon when a build is
stopped early due to build failure on another machine. To ensure subsequent
tests can succeed, kill off any such daemons.
We also have to hack the top-level Makefile to enable the openssl regression
tests. (Why doesn't this happen automatically given the configure option??)
@ -16,7 +19,7 @@ would need to be changed to use either one.
diff -Naur mysql-5.0.22.orig/Makefile.am mysql-5.0.22/Makefile.am
--- mysql-5.0.22.orig/Makefile.am 2006-05-25 04:56:41.000000000 -0400
+++ mysql-5.0.22/Makefile.am 2006-06-09 00:03:21.000000000 -0400
+++ mysql-5.0.22/Makefile.am 2006-06-10 13:13:34.000000000 -0400
@@ -106,8 +106,8 @@
test:
@ -30,39 +33,36 @@ diff -Naur mysql-5.0.22.orig/Makefile.am mysql-5.0.22/Makefile.am
cd mysql-test; \
diff -Naur mysql-5.0.22.orig/mysql-test/mysql-test-run.sh mysql-5.0.22/mysql-test/mysql-test-run.sh
--- mysql-5.0.22.orig/mysql-test/mysql-test-run.sh 2006-05-25 04:56:49.000000000 -0400
+++ mysql-5.0.22/mysql-test/mysql-test-run.sh 2006-06-09 00:06:22.000000000 -0400
@@ -252,6 +252,34 @@
+++ mysql-5.0.22/mysql-test/mysql-test-run.sh 2006-06-10 13:36:32.000000000 -0400
@@ -252,6 +252,16 @@
# 5.1 test run, even if different MTR_BUILD_THREAD is used. This means
# all port numbers might not be used in this version of the script.
#
+# hack to allow 32- and 64-bit tests to run concurrently on same build machine
+case `uname -m` in
+ i?86)
+ MTR_BUILD_THREAD=0
+ ;;
+ x86_64)
+ MTR_BUILD_THREAD=1
+ ;;
+ ia64)
+ MTR_BUILD_THREAD=2
+ ;;
+ ppc)
+ MTR_BUILD_THREAD=3
+ ;;
+ ppc64)
+ MTR_BUILD_THREAD=4
+ ;;
+ s390)
+ MTR_BUILD_THREAD=5
+ ;;
+ s390x)
+ MTR_BUILD_THREAD=6
+ ppc64 | s390x | x86_64)
+ MTR_BUILD_THREAD=7
+ ;;
+ *)
+ MTR_BUILD_THREAD=7
+ MTR_BUILD_THREAD=11
+ ;;
+esac
+
if [ -n "$MTR_BUILD_THREAD" ] ; then
MASTER_MYPORT=`expr $MTR_BUILD_THREAD '*' 10 + 10000`
MYSQL_MANAGER_PORT=`expr $MASTER_MYPORT + 2`
@@ -265,6 +275,14 @@
echo "Using NDBCLUSTER_PORT = $NDBCLUSTER_PORT"
fi
+#
+# If a previous build attempt failed it's possible that mysql daemons
+# launched by previous run of this script are still hanging around.
+# Kill 'em so we can re-use the ports.
+#
+ps auxww | grep "/sql/mysqld .*port=$MASTER_MYPORT" | awk '{print $2}' | xargs kill
+ps auxww | grep "/sql/mysqld .*port=$SLAVE_MYPORT" | awk '{print $2}' | xargs kill
+
NO_SLAVE=0
USER_TEST=
FAILED_CASES=

6
mysql.spec
View File

@ -1,6 +1,6 @@
Name: mysql
Version: 5.0.22
Release: 1
Release: 2
Summary: MySQL client programs and shared libraries.
License: GPL
Group: Applications/Databases
@ -414,6 +414,10 @@ fi
%attr(-,mysql,mysql) %{_datadir}/mysql-test
%changelog
* Sat Jun 10 2006 Tom Lane <tgl@redhat.com> 5.0.22-2
- Work around brew's tendency not to clean up failed builds completely,
by adding code in mysql-testing.patch to kill leftover mysql daemons.
* Thu Jun 8 2006 Tom Lane <tgl@redhat.com> 5.0.22-1
- Update to MySQL 5.0.22 (fixes CVE-2006-2753)
- Install temporary workaround for gcc bug on s390x (bz #193912)