[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: cond5.test spurious failure
From: |
Peter Rosin |
Subject: |
Re: cond5.test spurious failure |
Date: |
Fri, 06 Aug 2010 16:11:20 +0200 |
User-agent: |
Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.7) Gecko/20100713 Thunderbird/3.1.1 |
Den 2010-08-06 15:56 skrev Peter Rosin:
kill success:
=== Running test ./tmp.test
++ pwd
/home/peda/automake/git/automake/tests/tmp.dir
+ set -e
+ cat
+ cat
+ sh -c 'echo %aclocal pid: $$; sleep 2'
%aclocal pid: 12228
+ pid=13504
+ sh -c 'echo %automake pid: $$; sleep 4; echo variable OPT_SRC recursively defined
>&2'
+ try=1
+ test 1 -le 30
+ kill -0 13504
+ : process 13504 is still alive, wait and retry
+ sleep 1
%automake pid: 13504
++ expr 1 + 1
+ try=2
+ test 2 -le 30
+ kill -0 13504
+ : process 13504 is still alive, wait and retry
+ sleep 1
++ expr 2 + 1
+ try=3
+ test 3 -le 30
+ kill -0 13504
+ : process 13504 is still alive, wait and retry
+ sleep 1
++ expr 3 + 1
+ try=4
+ test 4 -le 30
+ kill -0 13504
+ : process 13504 is still alive, wait and retry
+ sleep 1
++ expr 4 + 1
+ try=5
+ test 5 -le 30
+ kill -0 13504
./tmp.test: line 54: kill: (13504) - No such process
+ cat stderr
variable OPT_SRC recursively defined
+ grep 'variable.*OPT_SRC.*recursively defined' stderr
variable OPT_SRC recursively defined
+ Exit 0
+ set +e
+ exit 0
+ exit 0
+ exit_status=0
+ set +e
+ cd /home/peda/automake/git/automake/tests
+ case $exit_status,$keep_testdirs in
+ find tmp.dir -type d '!' -perm -200 -exec chmod u+w '{}' ';'
+ rm -rf tmp.dir
+ test 0 '!=' 0
+ echo 'tmp: exit 0'
tmp: exit 0
+ exit 0
As seen above, kill seems to always work once it starts to work,
so I tried to insert a sleep before the first kill. That "fixes"
it. I have 10+ consecutive successes, no fails.
Please apply this on top of maint:
Cheers,
Peter
commit 146f01f899d677dbf02a41d55256a61d50a2b73c
Author: Peter Rosin <address@hidden>
Date: Fri Aug 6 16:03:52 2010 +0200
Make cond5.test more robust on MSYS.
* tests/cond5.test: Add delay before the first kill attempt to
cater for problems with MSYS bash.
diff --git a/ChangeLog b/ChangeLog
index 4f96039..de7124d 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,9 @@
+2010-08-06 Peter Rosin <address@hidden>
+
+ Make cond5.test more robust on MSYS.
+ * tests/cond5.test: Add delay before the first kill attempt to
+ cater for problems with MSYS bash.
+
2010-07-31 Ralf Wildenhues <address@hidden>
Add example git work flow; discuss merge --log in HACKING.
diff --git a/tests/cond5.test b/tests/cond5.test
index 84afdd0..88e2e42 100755
--- a/tests/cond5.test
+++ b/tests/cond5.test
@@ -48,6 +48,9 @@ $ACLOCAL
$AUTOMAKE 2>stderr &
pid=$!
+# MSYS bash seems to have a bug in kill, so don't try to kill too soon.
+sleep 2
+
# Make at most 30 tries, one every 10 seconds (= 300 seconds = 5 min).
try=1
while test $try -le 30; do
- Re: New auxiliary archive script, (continued)
- Re: New auxiliary archive script, Peter Rosin, 2010/08/05
- Re: New auxiliary archive script, Ralf Wildenhues, 2010/08/06
- Re: New auxiliary archive script, Peter Rosin, 2010/08/06
- cond5.test spurious failure (was: New auxiliary archive script), Stefano Lattarini, 2010/08/06
- Re: cond5.test spurious failure, Peter Rosin, 2010/08/06
- Re: cond5.test spurious failure, Stefano Lattarini, 2010/08/06
- Re: cond5.test spurious failure, Peter Rosin, 2010/08/06
- Re: cond5.test spurious failure,
Peter Rosin <=
- Re: cond5.test spurious failure, Stefano Lattarini, 2010/08/06
- Re: cond5.test spurious failure, Ralf Wildenhues, 2010/08/06
- Re: cond5.test spurious failure, Stefano Lattarini, 2010/08/06
- Re: cond5.test spurious failure, Peter Rosin, 2010/08/06
- Re: cond5.test spurious failure, Stefano Lattarini, 2010/08/06
- [PATCH] Remove a race condition from test cond5.test (was: cond5.test spurious failure), Stefano Lattarini, 2010/08/06
- Re: cond5.test spurious failure, Peter Rosin, 2010/08/06
- Re: cond5.test spurious failure, Peter Rosin, 2010/08/11
- Re: New auxiliary archive script, Peter Rosin, 2010/08/06
- Re: New auxiliary archive script, Ralf Wildenhues, 2010/08/07