You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@spamassassin.apache.org by bu...@bugzilla.spamassassin.org on 2009/06/19 23:03:48 UTC
[Bug 6138] New: "Hudson build is still unstable" messages
https://issues.apache.org/SpamAssassin/show_bug.cgi?id=6138
Summary: "Hudson build is still unstable" messages
Product: Spamassassin
Version: SVN Trunk (Latest Devel Version)
Platform: All
OS/Version: All
Status: NEW
Severity: major
Priority: P5
Component: Building & Packaging
AssignedTo: dev@spamassassin.apache.org
ReportedBy: sidney@sidney.com
Created an attachment (id=4466)
--> (https://issues.apache.org/SpamAssassin/attachment.cgi?id=4466)
Proposed patch
The Hudson builds started producing error messages reporting the build as
"unstable" for every build starting around June 12, 2009. For example see
http://hudson.zones.apache.org/hudson/job/SpamAssassin-trunk/3291/changes
There has been quite a bit of discussion on sa-dev mailing list. My analysis is
that it is caused by the loadplugin in v330.pre that uses the optional filename
syntax of loadplugin, which specifies to load the file from the current
directory. SATest.pm copies all the *.pre files from t/log/test_rules_copy/ to
t/log/localrules.tmp/. That causes the loadplugin to fail because the file,
EmailBL.pm is not also copied there. This all results from using a test rule
(EmailBL.PM) that is a plugin, something that I don't see having been done
before. Looking at old logs from Hudson, I see that the error in the build
started happening when EmailBL was first added back around May 11 or 12.
Something changed in Hudson that made it noisier and reporting the already
existing error on June 12.
I think that the correct fix is to copy *.pm at the same time that *.pre is
copied to t/log/localrules.tmp/ because any such loadplugin in a pre file
requires that the associated pm file be in the same directory.
--
Configure bugmail: https://issues.apache.org/SpamAssassin/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
[Bug 6138] "Hudson build is still unstable" messages
Posted by bu...@bugzilla.spamassassin.org.
https://issues.apache.org/SpamAssassin/show_bug.cgi?id=6138
Sidney Markowitz <si...@sidney.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |RESOLVED
Resolution| |FIXED
--- Comment #1 from Sidney Markowitz <si...@sidney.com> 2009-06-19 14:06:39 PST ---
Committed revision 786665 ( https://svn.apache.org/viewcvs.cgi?view=rev&rev=786665 ).
--
Configure bugmail: https://issues.apache.org/SpamAssassin/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
[Bug 6138] "Hudson build is still unstable" messages
Posted by bu...@bugzilla.spamassassin.org.
https://issues.apache.org/SpamAssassin/show_bug.cgi?id=6138
Sidney Markowitz <si...@sidney.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|REOPENED |RESOLVED
Resolution| |DUPLICATE
--- Comment #3 from Sidney Markowitz <si...@sidney.com> 2009-06-19 15:41:00 PST ---
I was confused by the warnings.
This is a duplicate of bug 6132
In general, warnings such as the failed loadplugin do not cause a test to fail
and are not reported by Hudson. t/recreate.t has an explicit test (currently
test 9) that generates a fail if there are any warnings from the previous tests
other that the loadplugin failures. It is the failure of test 9 in t/recreate.t
that is what Hudson is complaining about. So this is the problem being
discussed in bug 6132 comment 7
*** This bug has been marked as a duplicate of bug 6132 ***
--
Configure bugmail: https://issues.apache.org/SpamAssassin/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
[Bug 6138] "Hudson build is still unstable" messages
Posted by bu...@bugzilla.spamassassin.org.
https://issues.apache.org/SpamAssassin/show_bug.cgi?id=6138
Sidney Markowitz <si...@sidney.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|RESOLVED |REOPENED
Resolution|FIXED |
--- Comment #2 from Sidney Markowitz <si...@sidney.com> 2009-06-19 15:13:20 PST ---
I was too quick on this. The build still reports as unstable with the error
appearing in the test t/recreate.t, with the same type of loadplugin failure.
This could explain what happened on June 12, if Hudson doesn't care about the
loadplugin failure in most of the tests, but does in t/recreate.t.
--
Configure bugmail: https://issues.apache.org/SpamAssassin/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.