Mailing List Archive: SpamAssassin: devel

Hudson build is still unstable: SpamAssassin-trunk #3283

Index | Next | Previous | View Threaded

hudson at hudson

Jun 17, 2009, 4:51 PM

Post #1 of 9 (553 views)
Permalink

Hudson build is still unstable: SpamAssassin-trunk #3283

See http://hudson.zones.apache.org/hudson/job/SpamAssassin-trunk/3283/

michael.monnerie at is

Jun 17, 2009, 10:46 PM

Post #2 of 9 (516 views)
Permalink

Re: Hudson build is still unstable: SpamAssassin-trunk #3283 [In reply to]

On Donnerstag 18 Juni 2009 Apache Hudson Server wrote:
> See
> http://hudson.zones.apache.org/hudson/job/SpamAssassin-trunk/3283/

Poor Hudson is whining since a long time. Anybody taking care already?
Or does everybody think "somebody should take care about it"? :-)

mfg zmi
--
// Michael Monnerie, Ing.BSc ----- http://it-management.at
// Tel: 0660 / 415 65 31 .network.your.ideas.
// PGP Key: "curl -s http://zmi.at/zmi.asc | gpg --import"
// Fingerprint: AC19 F9D5 36ED CD8A EF38 500E CE14 91F7 1C12 09B4
// Keyserver: wwwkeys.eu.pgp.net Key-ID: 1C1209B4

jm at jmason

Jun 18, 2009, 12:53 AM

Post #3 of 9 (518 views)
Permalink

Re: Hudson build is still unstable: SpamAssassin-trunk #3283 [In reply to]

it's an issue with a recently-added test plugin; there's a comment on
the ticket about it. hopefully someone is looking into it ;)

--j.

On Thu, Jun 18, 2009 at 06:46, Michael
Monnerie<michael.monnerie[at]is.it-management.at> wrote:
> On Donnerstag 18 Juni 2009 Apache Hudson Server wrote:
>> See
>> http://hudson.zones.apache.org/hudson/job/SpamAssassin-trunk/3283/
>
> Poor Hudson is whining since a long time. Anybody taking care already?
> Or does everybody think "somebody should take care about it"? :-)
>
> mfg zmi
> --
> // Michael Monnerie, Ing.BSc ----- http://it-management.at
> // Tel: 0660 / 415 65 31 .network.your.ideas.
> // PGP Key: "curl -s http://zmi.at/zmi.asc | gpg --import"
> // Fingerprint: AC19 F9D5 36ED CD8A EF38 500E CE14 91F7 1C12 09B4
> // Keyserver: wwwkeys.eu.pgp.net Key-ID: 1C1209B4
>
>

sidney at sidney

Jun 18, 2009, 2:09 AM

Post #4 of 9 (513 views)
Permalink

Re: Hudson build is still unstable: SpamAssassin-trunk #3283 [In reply to]

Justin Mason wrote, On 18/6/09 7:53 PM:
> it's an issue with a recently-added test plugin; there's a comment on
> the ticket about it. hopefully someone is looking into it ;)

I just looked at it and I'm not making any sense out of it.

The failures started after the checkin of revision 783933. That checkin
was a change in masses/rule-qa/report_from_logs and I don't see how that
could effect make test. The previous checkin was 783698, and it did
trigger a successful Hudson build.

I can make the errors go away in t/SATest.pm by forcing the copy of
EmailBL.pm from trunk/rules/ to t/log/localrules.tmp/ at the same time
that SATest.pm copies *.pre from there to there.

I can also make the errors go away by commenting out the Loadplugin of
EmailBL that is in v330.pre

However, all the Hudson builds since that loadplugin line was added to
v330.pre last month were successful up until the failures started after
revision 783933 checkin on June 12.

Does anyone make sense out of this?

-- sidney

jm at jmason

Jun 18, 2009, 3:25 AM

Post #5 of 9 (511 views)
Permalink

Re: Hudson build is still unstable: SpamAssassin-trunk #3283 [In reply to]

On Thu, Jun 18, 2009 at 10:09, Sidney Markowitz<sidney[at]sidney.com> wrote:
> Justin Mason wrote, On 18/6/09 7:53 PM:
>>
>> it's an issue with a recently-added test plugin; there's a comment on
>> the ticket about it. hopefully someone is looking into it ;)
>
> I just looked at it and I'm not making any sense out of it.
>
> The failures started after the checkin of revision 783933. That checkin was
> a change in masses/rule-qa/report_from_logs and I don't see how that could
> effect make test. The previous checkin was 783698, and it did trigger a
> successful Hudson build.
>
> I can make the errors go away in t/SATest.pm by forcing the copy of
> EmailBL.pm from trunk/rules/ to t/log/localrules.tmp/ at the same time that
> SATest.pm copies *.pre from there to there.
>
> I can also make the errors go away by commenting out the Loadplugin of
> EmailBL that is in v330.pre
>
> However, all the Hudson builds since that loadplugin line was added to
> v330.pre last month were successful up until the failures started after
> revision 783933 checkin on June 12.
>
> Does anyone make sense out of this?

hmm. that sounds like the process of "compiling" it from the rulesrc
sandbox into rules, and then into t/log/localrules.tmp , is going
awry.

--j.

sidney at sidney

Jun 18, 2009, 4:18 AM

Post #6 of 9 (507 views)
Permalink

Re: Hudson build is still unstable: SpamAssassin-trunk #3283 [In reply to]

Justin Mason wrote, On 18/6/09 10:25 PM:
> hmm. that sounds like the process of "compiling" it from the rulesrc
> sandbox into rules, and then into t/log/localrules.tmp , is going
> awry.

I tried backing out to revision 783698 and it still failed the same way.
But I'm not convinced that the external pull of rulesrc honors the
revision number I specified by saying svn update -r 783698

In fact, I don't see how it could be affected by the -r option.

-- sidney

jm at jmason

Jun 18, 2009, 4:24 AM

Post #7 of 9 (507 views)
Permalink

Re: Hudson build is still unstable: SpamAssassin-trunk #3283 [In reply to]

On Thu, Jun 18, 2009 at 12:18, Sidney Markowitz<sidney[at]sidney.com> wrote:
> Justin Mason wrote, On 18/6/09 10:25 PM:
>>
>> hmm. that sounds like the process of "compiling" it from the rulesrc
>> sandbox into rules, and then into t/log/localrules.tmp , is going
>> awry.
>
> I tried backing out to revision 783698 and it still failed the same way. But
> I'm not convinced that the external pull of rulesrc honors the revision
> number I specified by saying svn update -r 783698
>
> In fact, I don't see how it could be affected by the -r option.

with externals you have to to a _separate_ svn update command on the
rulesrc dir.
I can't wait to get rid of that ;)

--j.

sidney at sidney

Jun 18, 2009, 11:57 AM

Post #8 of 9 (500 views)
Permalink

Re: Hudson build is still unstable: SpamAssassin-trunk #3283 [In reply to]

Justin Mason wrote, On 18/6/09 11:24 PM:
> On Thu, Jun 18, 2009 at 12:18, Sidney Markowitz<sidney[at]sidney.com> wrote:
>> Justin Mason wrote, On 18/6/09 10:25 PM:
>>> hmm. that sounds like the process of "compiling" it from the rulesrc
>>> sandbox into rules, and then into t/log/localrules.tmp , is going
>>> awry.

At what point is it supposed to end up in t/log/localrules.tmp?

>> I tried backing out to revision 783698 and it still failed the same way. But
>> I'm not convinced that the external pull of rulesrc honors the revision
>> number I specified by saying svn update -r 783698
>>
>> In fact, I don't see how it could be affected by the -r option.
>
> with externals you have to to a _separate_ svn update command on the
> rulesrc dir.
> I can't wait to get rid of that ;)

I tried doing an svn update -r 783698 in spamassassin/trunk, and then in
spamassassin/rules/trunk, and then in spamassassin/trunk/rulesrc, and
then I did make distclean; perl Makefile.PL; make; make test

I still got the failure.

-- sidney

sidney at sidney

Jun 18, 2009, 6:18 PM

Post #9 of 9 (490 views)
Permalink

Re: Hudson build is still unstable: SpamAssassin-trunk #3283 [In reply to]

Justin Mason wrote, On 18/6/09 11:24 PM:
> On Thu, Jun 18, 2009 at 12:18, Sidney Markowitz<sidney[at]sidney.com> wrote:
>> Justin Mason wrote, On 18/6/09 10:25 PM:
>>> hmm. that sounds like the process of "compiling" it from the rulesrc
>>> sandbox into rules, and then into t/log/localrules.tmp , is going
>>> awry.

I see where rules are copied into t/log/test_rules_copy and where *.pre
is copied into t/log/localrules.tmp, but I don't see any place that has
anything to do with copying rules into t/log/localrules.tmp even though
the directory name does make it sound like that's what it was made for.

v330.pre contains a loadplugin that specifies loading EmailBL.pm from
the same directory as the v330.pre file that is being run. All of the
*.pre and the rules *.pm files are in t/log/test_rules_copy already. If
I suppress the copying of *pre from there to t/log/localrules.tmp then
there are no files in t/log/localrules.tmp but all those *.pre files get
loaded and run from t/log/test_rules_copy and everything works ok.

So what is the point of t/log/localrules.tmp?

>> I tried backing out to revision 783698 and it still failed the same way

I went even further and used svn export -r 783698 into a new directory
to get rev 783698 trunk, then deleted trunk/ruleserc, then used svn
export -r 783698 to get that older revision of it. That failed the same
way. I don't understand how the build did not start failing when EmailBL
was first added.

-- sidney

Index | Next | Previous | View Threaded

Interested in having your list archived? Contact lists@gossamer-threads.com