Hi,
It seems Jenkins build is back to normal now. I've triggered another build
just to make sure. Created a JIRA to track the registry exception handling
issue at [1]. Appreciate if someone can go through the commit [2] and
review the changes.
This will only make sure registry exceptions are prop
Great. That registry locks without releasing may cause this intermittent
issue. Will see with the new build. As there are no such a big functional
changes from 4.1.2 to 4.1.3, we can check this integration test against
4.1.2 release as we don't experience this issue with 4.1.2
Thanks,
Gayan
On We
Hi Gayan,
Yes, this was my mistake. Sorry for the inconvenience. I've fixed it now.
While going through Carbon registry transactions one by one, I noticed that
some transactions are not handled properly. There were cases in which
registry exception could permanently leave a lock without releasing
yeah.Seems we are catching the wrong exception.
StratosApiV41Utils.java
try {
autoscalerServiceClient.undeployApplication(applicationId,
force);
try {
clearMetadata(applicationId);
} catch (RegistryException e) {
I'm now seeing below error in 4.1.x branch:
[INFO]
[INFO] BUILD FAILURE
[INFO]
[INFO] Total time: 1:48.133s
[INFO] Finished at: Wed Sep 16 09:42:32 IST
Hi,
I'm still not able to reproduce this issue locally. I've run integration
tests multiple times and they are successful at every time. I compared the
p2-profile and and configuration files between Stratos 4.1.2 and Stratos
4.1.3-SNAPSHOT, and there are no differences in p2-profile. Same dependen
Hi Gayan,
lock timeout is already set to 60s which is not the problem I think. It
seems this is a known issue in Carbon registry [1]. There is a fix
suggested in wso2-dev mailing list which is to set the mvcc option. I'm
currently trying that option locally.
[1] https://wso2.org/jira/browse/REGIS
Seems this is a intermittent issue.Seems to be that another process has
locked the table, and I think we may need to try with increasing the lock
timeout and multi-version concurrency option[1][2]
[1]http://h2database.com/html/grammar.html#set_lock_timeout
[2]http://www.h2database.com/html/advance
It looks like this is the same issue which Udara has reported sometime
back [1].
[1] - Registry issue when persisting topology
On Tue, Sep 15, 2015 at 7:20 AM, Dinithi De Silva wrote:
> It looks like this is the same issue which Udara has reported sometime
> back [1].
>
> [1] - Registry issue
It looks like this is the same issue which Udara has reported sometime back
[1].
[1] - Registry issue when persisting topology
On Mon, Sep 14, 2015 at 6:38 PM, Imesh Gunaratne wrote:
> Hi Devs,
>
> I see the following issue in 4.1.x branch:
>
> INFO - [2015-09-14 06:39:49,522] INFO
> {org.apac
It looks like this is the same issue which Udara has reported sometime back
[1].
[1] - Registry issue when persisting topology
Thanks.
On Mon, Sep 14, 2015 at 6:38 PM, Imesh Gunaratne wrote:
> Hi Devs,
>
> I see the following issue in 4.1.x branch:
>
> INFO - [2015-09-14 06:39:49,522] INFO
>
Hi Devs,
I see the following issue in 4.1.x branch:
INFO - [2015-09-14 06:39:49,522] INFO
{org.apache.stratos.mock.iaas.services.impl.MockIaasServiceImpl} -
Terminating instance: [instance-id]
sample-applications-test-1.c1-1x0-sample-applications-test.c1-sample-applications-test.domaina2d8a0c8-
12 matches
Mail list logo