You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@sling.apache.org by Radu Cotescu <ra...@apache.org> on 2016/03/14 10:51:26 UTC
launchpad random failures during Sightly IT
Hello,
I've noticed that the launchpad randomly fails to start for the Sightly IT
module [0][1]. The first logged errors are:
14.03.2016 10:32:49.148 *ERROR* [qtp368404860-211]
org.apache.sling.engine.impl.SlingHttpContext handleSecurity:
AuthenticationSupport service missing. Cannot authenticate request.
14.03.2016 10:32:49.148 *ERROR* [qtp368404860-211]
org.apache.sling.engine.impl.SlingHttpContext handleSecurity: Possible
reason is missing Repository service. Check AuthenticationSupport
dependencies.
Any idea what might be wrong? We do set a custom Resource Resolver Factory
configuration at [2] but SLING-5561 [3] and SLING-5580 [4] were fixed three
days ago and this problem can still be reproduced on a freshly built
launchpad.
Thanks,
Radu
[0] -
https://github.com/apache/sling/blob/trunk/bundles/scripting/sightly/testing/pom.xml
[1] -
https://builds.apache.org/job/sling-trunk-1.7/3581/org.apache.sling$org.apache.sling.scripting.sightly.testing/consoleFull
[2] -
https://github.com/apache/sling/blob/trunk/bundles/scripting/sightly/testing-content/src/main/resources/SLING-INF/apps/sightlytck/install.json
[3] - https://issues.apache.org/jira/browse/SLING-5561
[4] - https://issues.apache.org/jira/browse/SLING-5580
Re: launchpad random failures during Sightly IT
Posted by Radu Cotescu <ra...@apache.org>.
I guess the source of the failures is
https://issues.apache.org/jira/browse/SLING-5602 since the Sightly tests
were providing a RRF configuration through a sling:OsgiConfig mode.
After moving the RRF config to the tests' provisioning model I couldn't
reproduce the random test failures any more.
On Mon, 14 Mar 2016 at 13:26 Bertrand Delacretaz <bd...@apache.org>
wrote:
> On Mon, Mar 14, 2016 at 12:27 PM, Radu Cotescu <ra...@cotescu.com> wrote:
> > ...The tests should start after the Launchpad is ready - this is the
> reason
> > for which there's a LaunchpadReadyIT test run first...
>
> Ok, yes getting 200 on /index.html should indicate that the repository is
> up.
>
> I don't have more clues right now, your log says "405 Method POST not
> supported", I don't know where that comes from, next step is probably
> to find a way to debug what's happening.
>
>
> -Bertrand
>
Re: launchpad random failures during Sightly IT
Posted by Bertrand Delacretaz <bd...@apache.org>.
On Mon, Mar 14, 2016 at 12:27 PM, Radu Cotescu <ra...@cotescu.com> wrote:
> ...The tests should start after the Launchpad is ready - this is the reason
> for which there's a LaunchpadReadyIT test run first...
Ok, yes getting 200 on /index.html should indicate that the repository is up.
I don't have more clues right now, your log says "405 Method POST not
supported", I don't know where that comes from, next step is probably
to find a way to debug what's happening.
-Bertrand
Re: launchpad random failures during Sightly IT
Posted by Radu Cotescu <ra...@cotescu.com>.
The tests should start after the Launchpad is ready - this is the reason
for which there's a LaunchpadReadyIT test run first -
https://github.com/apache/sling/blob/trunk/bundles/scripting/sightly/testing/src/test/java/LaunchpadReadyIT.java
.
Oak seems to be up and running:
14.03.2016 10:32:46.116 *INFO* [OsgiInstallerImpl]
org.apache.jackrabbit.oak.plugins.segment.file.FileStore TarMK opened:
sling/repository/segmentstore (mmap=true)
14.03.2016 10:32:46.119 *INFO* [OsgiInstallerImpl]
org.apache.jackrabbit.oak.plugins.segment.file.FileStore Compaction
strategy set to: CompactionStrategy{paused=true, cloneBinaries=false,
cleanupType=CLEAN_OLD, olderThan=36000000, memoryThreshold=5,
persistedCompactionMap=true, retryCount=5, forceAfterFail=false,
compactionStart=1457947966119, offlineCompaction=false}
Any other clues?
On Mon, 14 Mar 2016 at 11:43 Bertrand Delacretaz <bd...@apache.org>
wrote:
> Hi,
>
> On Mon, Mar 14, 2016 at 10:51 AM, Radu Cotescu <ra...@apache.org> wrote:
> > ...org.apache.sling.engine.impl.SlingHttpContext handleSecurity: Possible
> > reason is missing Repository service....
>
> This message is usually correct, did you check logs for any messages
> from the content repository startup modules?
>
> Or maybe your tests are starting too soon and Sling isn't ready yet?
>
> -Bertrand
>
Re: launchpad random failures during Sightly IT
Posted by Bertrand Delacretaz <bd...@apache.org>.
Hi,
On Mon, Mar 14, 2016 at 10:51 AM, Radu Cotescu <ra...@apache.org> wrote:
> ...org.apache.sling.engine.impl.SlingHttpContext handleSecurity: Possible
> reason is missing Repository service....
This message is usually correct, did you check logs for any messages
from the content repository startup modules?
Or maybe your tests are starting too soon and Sling isn't ready yet?
-Bertrand