Re: Run jenkins test on a rhel6 host
On Thu, Jan 17, 2013 at 03:58:55PM +0800, Mark Wu wrote:
Hi guys,
I am a vdsm developer. Recently I submit a vdsm patch fix a break on fedora 18. I just verify it on fedora, but not on RHEL6 before merged. Unfortunately, my patch called a function which has a bug on rhel6. That bug was fixed in fedora, but not backported to rhel6. So I am thinking is it possible to also run the jenkins test on a rhel6 host for each patch? Then we can find the break in time. and the developer don't need to run the test cases on two systems for one patch.
I'd like to join Mark's plea. I believe that the only problem is of finding the resources for another Jenkins slave. Missing an el6 slave is a pain point for me - I've accepted Saggi's http://gerrit.ovirt.org/9986 and now Vdsm's master branch fails to build on EL6 on the following grounds: ERROR: testMethodBadParameters(reactorType='tcp') (jsonRpcTests.JsonRpcServerTests) ---------------------------------------------------------------------- Traceback (most recent call last): File "/home/mine/.rpm/BUILD/vdsm-4.10.3/tests/testrunner.py", line 65, in wrapper return f(self, *args) File "/home/mine/.rpm/BUILD/vdsm-4.10.3/tests/jsonRpcTests.py", line 273, in testMethodBadParameters JsonRpcInternalError().code) File "/usr/lib64/python2.6/contextlib.py", line 34, in __exit__ self.gen.throw(type, value, traceback) File "/home/mine/.rpm/BUILD/vdsm-4.10.3/tests/jsonRpcTests.py", line 97, in constructServer yield server, jsonClientFactory File "/home/mine/.rpm/BUILD/vdsm-4.10.3/tests/jsonRpcTests.py", line 269, in testMethodBadParameters with self.assertRaises(JsonRpcError) as cm: TypeError: failUnlessRaises() takes at least 3 arguments (2 given) ====================================================================== ERROR: testMethodMissingMethod(reactorType='tcp') (jsonRpcTests.JsonRpcServerTests) ---------------------------------------------------------------------- Traceback (most recent call last): File "/home/mine/.rpm/BUILD/vdsm-4.10.3/tests/testrunner.py", line 65, in wrapper return f(self, *args) File "/home/mine/.rpm/BUILD/vdsm-4.10.3/tests/jsonRpcTests.py", line 258, in testMethodMissingMethod JsonRpcMethodNotFoundError().code) File "/usr/lib64/python2.6/contextlib.py", line 34, in __exit__ self.gen.throw(type, value, traceback) File "/home/mine/.rpm/BUILD/vdsm-4.10.3/tests/jsonRpcTests.py", line 97, in constructServer yield server, jsonClientFactory File "/home/mine/.rpm/BUILD/vdsm-4.10.3/tests/jsonRpcTests.py", line 254, in testMethodMissingMethod with self.assertRaises(JsonRpcError) as cm: TypeError: failUnlessRaises() takes at least 3 arguments (2 given)
This is an OpenPGP/MIME signed message (RFC 4880 and 3156) ------enig2EWBJHROAWQKTEOQSPXPT Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable On 01/20/2013 06:34 AM, Dan Kenigsberg wrote:
Hi guys,
I am a vdsm developer. Recently I submit a vdsm patch fix a break on fedora 18. I just verify it on fedora, but not on RHEL6 before merged. Unfortunately, my patch called a function which has a bug on rhel6. That bug was fixed in fedora, but not backported to rhel6. So I am thinking is it possible to also run the jenkins test on a rhel6 host for each patch? Then we can find the break in time. and the developer don't need to run the test cases on two systems for one patch. =20 I'd like to join Mark's plea. I believe that the only problem is of fi= nding
On Thu, Jan 17, 2013 at 03:58:55PM +0800, Mark Wu wrote: the resources for another Jenkins slave. =20 Missing an el6 slave is a pain point for me - I've accepted Saggi's http://gerrit.ovirt.org/9986 and now Vdsm's master branch fails to build on EL6 on the following grounds:
OK, we're working on getting actual RHEL 6 entitlements for the new hosts at Rackspace, which has one server to be dedicated to multiple Jenkins slaves. It will be a week or more for those hosts to be available. Once I have the details, I'll forward them to the Jenkins admins so they can get started ASAP. (We don't need to wait for Puppet, we can puppetize the server later.) - Karsten --=20 Karsten 'quaid' Wade, Sr. Analyst - Community Growth http://TheOpenSourceWay.org .^\ http://community.redhat.com @quaid (identi.ca/twitter/IRC) \v' gpg: AD0E0C41 ------enig2EWBJHROAWQKTEOQSPXPT Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.13 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iD8DBQFQ/DLk2ZIOBq0ODEERAqV3AJoD913PhdylriLQNJAtAnU6NFgJ+wCgoSBM G228+dHtviR+95JrpaxbnY0= =O11q -----END PGP SIGNATURE----- ------enig2EWBJHROAWQKTEOQSPXPT--
On 01/20/2013 01:09 PM, Karsten 'quaid' Wade wrote:
On 01/20/2013 06:34 AM, Dan Kenigsberg wrote:
On Thu, Jan 17, 2013 at 03:58:55PM +0800, Mark Wu wrote:
Hi guys,
I am a vdsm developer. Recently I submit a vdsm patch fix a break on fedora 18. I just verify it on fedora, but not on RHEL6 before merged. Unfortunately, my patch called a function which has a bug on rhel6. That bug was fixed in fedora, but not backported to rhel6. So I am thinking is it possible to also run the jenkins test on a rhel6 host for each patch? Then we can find the break in time. and the developer don't need to run the test cases on two systems for one patch.
I'd like to join Mark's plea. I believe that the only problem is of finding the resources for another Jenkins slave.
Missing an el6 slave is a pain point for me - I've accepted Saggi's http://gerrit.ovirt.org/9986 and now Vdsm's master branch fails to build on EL6 on the following grounds:
OK, we're working on getting actual RHEL 6 entitlements for the new hosts at Rackspace, which has one server to be dedicated to multiple Jenkins slaves.
It will be a week or more for those hosts to be available. Once I have the details, I'll forward them to the Jenkins admins so they can get started ASAP. (We don't need to wait for Puppet, we can puppetize the server later.)
- Karsten
_______________________________________________ Infra mailing list Infra@ovirt.org http://lists.ovirt.org/mailman/listinfo/infra
we have two rhel 6 jenkins slaves on ec2 today - is there an issue with them?
----- Original Message -----
From: "Itamar Heim" <iheim@redhat.com> To: "Karsten 'quaid' Wade" <kwade@redhat.com> Cc: infra@ovirt.org Sent: Monday, January 21, 2013 6:54:09 AM Subject: Re: Run jenkins test on a rhel6 host
On 01/20/2013 01:09 PM, Karsten 'quaid' Wade wrote:
On 01/20/2013 06:34 AM, Dan Kenigsberg wrote:
On Thu, Jan 17, 2013 at 03:58:55PM +0800, Mark Wu wrote:
Hi guys,
I am a vdsm developer. Recently I submit a vdsm patch fix a break on fedora 18. I just verify it on fedora, but not on RHEL6 before merged. Unfortunately, my patch called a function which has a bug on rhel6. That bug was fixed in fedora, but not backported to rhel6. So I am thinking is it possible to also run the jenkins test on a rhel6 host for each patch? Then we can find the break in time. and the developer don't need to run the test cases on two systems for one patch.
I'd like to join Mark's plea. I believe that the only problem is of finding the resources for another Jenkins slave.
Missing an el6 slave is a pain point for me - I've accepted Saggi's http://gerrit.ovirt.org/9986 and now Vdsm's master branch fails to build on EL6 on the following grounds:
OK, we're working on getting actual RHEL 6 entitlements for the new hosts at Rackspace, which has one server to be dedicated to multiple Jenkins slaves.
It will be a week or more for those hosts to be available. Once I have the details, I'll forward them to the Jenkins admins so they can get started ASAP. (We don't need to wait for Puppet, we can puppetize the server later.)
- Karsten
_______________________________________________ Infra mailing list Infra@ovirt.org http://lists.ovirt.org/mailman/listinfo/infra
we have two rhel 6 jenkins slaves on ec2 today - is there an issue with them?
they are running 6.2, is that enough? can we upgrade them to 6.3? i can run the same job on them if it will help for now.
_______________________________________________ Infra mailing list Infra@ovirt.org http://lists.ovirt.org/mailman/listinfo/infra
----- Original Message -----
From: "Eyal Edri" <eedri@redhat.com> To: "Itamar Heim" <iheim@redhat.com> Cc: infra@ovirt.org Sent: Monday, January 21, 2013 9:46:36 AM Subject: Re: Run jenkins test on a rhel6 host
----- Original Message -----
From: "Itamar Heim" <iheim@redhat.com> To: "Karsten 'quaid' Wade" <kwade@redhat.com> Cc: infra@ovirt.org Sent: Monday, January 21, 2013 6:54:09 AM Subject: Re: Run jenkins test on a rhel6 host
On 01/20/2013 01:09 PM, Karsten 'quaid' Wade wrote:
On 01/20/2013 06:34 AM, Dan Kenigsberg wrote:
On Thu, Jan 17, 2013 at 03:58:55PM +0800, Mark Wu wrote:
Hi guys,
I am a vdsm developer. Recently I submit a vdsm patch fix a break on fedora 18. I just verify it on fedora, but not on RHEL6 before merged. Unfortunately, my patch called a function which has a bug on rhel6. That bug was fixed in fedora, but not backported to rhel6. So I am thinking is it possible to also run the jenkins test on a rhel6 host for each patch? Then we can find the break in time. and the developer don't need to run the test cases on two systems for one patch.
I'd like to join Mark's plea. I believe that the only problem is of finding the resources for another Jenkins slave.
Missing an el6 slave is a pain point for me - I've accepted Saggi's http://gerrit.ovirt.org/9986 and now Vdsm's master branch fails to build on EL6 on the following grounds:
what job runs currently per patch for vdsm that will verify this patch, and you want it running on EL?
OK, we're working on getting actual RHEL 6 entitlements for the new hosts at Rackspace, which has one server to be dedicated to multiple Jenkins slaves.
It will be a week or more for those hosts to be available. Once I have the details, I'll forward them to the Jenkins admins so they can get started ASAP. (We don't need to wait for Puppet, we can puppetize the server later.)
- Karsten
_______________________________________________ Infra mailing list Infra@ovirt.org http://lists.ovirt.org/mailman/listinfo/infra
we have two rhel 6 jenkins slaves on ec2 today - is there an issue with them?
they are running 6.2, is that enough? can we upgrade them to 6.3?
i can run the same job on them if it will help for now.
_______________________________________________ Infra mailing list Infra@ovirt.org http://lists.ovirt.org/mailman/listinfo/infra
_______________________________________________ Infra mailing list Infra@ovirt.org http://lists.ovirt.org/mailman/listinfo/infra
On Mon, Jan 21, 2013 at 02:59:55AM -0500, Eyal Edri wrote:
----- Original Message -----
From: "Eyal Edri" <eedri@redhat.com> To: "Itamar Heim" <iheim@redhat.com> Cc: infra@ovirt.org Sent: Monday, January 21, 2013 9:46:36 AM Subject: Re: Run jenkins test on a rhel6 host
----- Original Message -----
From: "Itamar Heim" <iheim@redhat.com> To: "Karsten 'quaid' Wade" <kwade@redhat.com> Cc: infra@ovirt.org Sent: Monday, January 21, 2013 6:54:09 AM Subject: Re: Run jenkins test on a rhel6 host
On 01/20/2013 01:09 PM, Karsten 'quaid' Wade wrote:
On 01/20/2013 06:34 AM, Dan Kenigsberg wrote:
On Thu, Jan 17, 2013 at 03:58:55PM +0800, Mark Wu wrote:
Hi guys,
I am a vdsm developer. Recently I submit a vdsm patch fix a break on fedora 18. I just verify it on fedora, but not on RHEL6 before merged. Unfortunately, my patch called a function which has a bug on rhel6. That bug was fixed in fedora, but not backported to rhel6. So I am thinking is it possible to also run the jenkins test on a rhel6 host for each patch? Then we can find the break in time. and the developer don't need to run the test cases on two systems for one patch.
I'd like to join Mark's plea. I believe that the only problem is of finding the resources for another Jenkins slave.
Missing an el6 slave is a pain point for me - I've accepted Saggi's http://gerrit.ovirt.org/9986 and now Vdsm's master branch fails to build on EL6 on the following grounds:
what job runs currently per patch for vdsm that will verify this patch, and you want it running on EL?
http://jenkins.ovirt.org/job/vdsm_unit_tests_manual_gerrit/ The name of the job is quite bad for something that runs automatically.. It currently runs per patch on Fedora - I'd like to see it run on EL6, too.
----- Original Message -----
From: "Dan Kenigsberg" <danken@redhat.com> To: "Eyal Edri" <eedri@redhat.com> Cc: infra@ovirt.org, "Itamar Heim" <iheim@redhat.com> Sent: Monday, January 21, 2013 11:57:47 AM Subject: Re: Run jenkins test on a rhel6 host
On Mon, Jan 21, 2013 at 02:59:55AM -0500, Eyal Edri wrote:
----- Original Message -----
From: "Eyal Edri" <eedri@redhat.com> To: "Itamar Heim" <iheim@redhat.com> Cc: infra@ovirt.org Sent: Monday, January 21, 2013 9:46:36 AM Subject: Re: Run jenkins test on a rhel6 host
----- Original Message -----
From: "Itamar Heim" <iheim@redhat.com> To: "Karsten 'quaid' Wade" <kwade@redhat.com> Cc: infra@ovirt.org Sent: Monday, January 21, 2013 6:54:09 AM Subject: Re: Run jenkins test on a rhel6 host
On 01/20/2013 01:09 PM, Karsten 'quaid' Wade wrote:
On 01/20/2013 06:34 AM, Dan Kenigsberg wrote:
On Thu, Jan 17, 2013 at 03:58:55PM +0800, Mark Wu wrote: > Hi guys, > > I am a vdsm developer. Recently I submit a vdsm patch fix > a > break > on fedora 18. I just verify it on fedora, but not on RHEL6 > before > merged. Unfortunately, my patch called a function which > has a bug on rhel6. That bug was fixed in fedora, but not > backported > to rhel6. So I am thinking is it possible to also run the > jenkins > test on a rhel6 host for each patch? Then we can find the > break > in > time. > and the developer don't need to run the test cases on two > systems > for one patch.
I'd like to join Mark's plea. I believe that the only problem is of finding the resources for another Jenkins slave.
Missing an el6 slave is a pain point for me - I've accepted Saggi's http://gerrit.ovirt.org/9986 and now Vdsm's master branch fails to build on EL6 on the following grounds:
what job runs currently per patch for vdsm that will verify this patch, and you want it running on EL?
http://jenkins.ovirt.org/job/vdsm_unit_tests_manual_gerrit/
The name of the job is quite bad for something that runs automatically..
it started as a job to run manual gerrit patches (like a 'developer job'), i guess at some point it was change to run automaticly. changed the job name now. i'll have it running on rhel as well.
It currently runs per patch on Fedora - I'd like to see it run on EL6, too.
On 20/01/2013 23:46, Eyal Edri wrote:
they are running 6.2, is that enough? can we upgrade them to 6.3?
just yum update them? a job should do this nightly and run to detect issues changing in the distro (assuming the core vdsm job is stable)
i can run the same job on them if it will help for now.
----- Original Message -----
From: "Itamar Heim" <iheim@redhat.com> To: "Eyal Edri" <eedri@redhat.com> Cc: infra@ovirt.org, "Karsten 'quaid' Wade" <kwade@redhat.com> Sent: Monday, January 21, 2013 11:25:49 AM Subject: Re: Run jenkins test on a rhel6 host
On 20/01/2013 23:46, Eyal Edri wrote:
they are running 6.2, is that enough? can we upgrade them to 6.3?
just yum update them? a job should do this nightly and run to detect issues changing in the distro (assuming the core vdsm job is stable)
don't think we can just yum update, its public so we can't use repos. we need an rhn key/channel for 6.3 i suppose. once we'll have foreman+puppet managing the slaves it will be easier to handle various oss
i can run the same job on them if it will help for now.
On 21/01/2013 01:44, Eyal Edri wrote:
----- Original Message -----
From: "Itamar Heim" <iheim@redhat.com> To: "Eyal Edri" <eedri@redhat.com> Cc: infra@ovirt.org, "Karsten 'quaid' Wade" <kwade@redhat.com> Sent: Monday, January 21, 2013 11:25:49 AM Subject: Re: Run jenkins test on a rhel6 host
On 20/01/2013 23:46, Eyal Edri wrote:
they are running 6.2, is that enough? can we upgrade them to 6.3?
just yum update them? a job should do this nightly and run to detect issues changing in the distro (assuming the core vdsm job is stable)
don't think we can just yum update, its public so we can't use repos. we need an rhn key/channel for 6.3 i suppose.
yes you can. its an AMI with a configured rhn entitlement/repo.
once we'll have foreman+puppet managing the slaves it will be easier to handle various oss
i can run the same job on them if it will help for now.
updated both vms to 6.3 also added new vdsm jobs to run on them.. http://jenkins.ovirt.org/job/vdsm_unit_tests_gerrit_el/ eyal. ----- Original Message -----
From: "Itamar Heim" <iheim@redhat.com> To: "Eyal Edri" <eedri@redhat.com> Cc: infra@ovirt.org, "Karsten 'quaid' Wade" <kwade@redhat.com> Sent: Monday, January 21, 2013 11:46:03 AM Subject: Re: Run jenkins test on a rhel6 host
On 21/01/2013 01:44, Eyal Edri wrote:
----- Original Message -----
From: "Itamar Heim" <iheim@redhat.com> To: "Eyal Edri" <eedri@redhat.com> Cc: infra@ovirt.org, "Karsten 'quaid' Wade" <kwade@redhat.com> Sent: Monday, January 21, 2013 11:25:49 AM Subject: Re: Run jenkins test on a rhel6 host
On 20/01/2013 23:46, Eyal Edri wrote:
they are running 6.2, is that enough? can we upgrade them to 6.3?
just yum update them? a job should do this nightly and run to detect issues changing in the distro (assuming the core vdsm job is stable)
don't think we can just yum update, its public so we can't use repos. we need an rhn key/channel for 6.3 i suppose.
yes you can. its an AMI with a configured rhn entitlement/repo.
once we'll have foreman+puppet managing the slaves it will be easier to handle various oss
i can run the same job on them if it will help for now.
----- Original Message -----
From: "Ewoud Kohl van Wijngaarden" <ewoud+ovirt@kohlvanwijngaarden.nl> To: "Eyal Edri" <eedri@redhat.com> Cc: "Itamar Heim" <iheim@redhat.com>, infra@ovirt.org Sent: Wednesday, January 23, 2013 3:39:21 PM Subject: Re: Run jenkins test on a rhel6 host
On Tue, Jan 22, 2013 at 03:13:28PM -0500, Eyal Edri wrote:
updated both vms to 6.3
It looks like the nodes with rhel_62 are now offline. Maybe that needs to be fixed.
its not that, it's just the rhel_62 label was replaced with rhel_63. i changed the jobs that used it and it's running now.
On Wed, Jan 23, 2013 at 09:45:21AM -0500, Eyal Edri wrote:
----- Original Message -----
From: "Ewoud Kohl van Wijngaarden" <ewoud+ovirt@kohlvanwijngaarden.nl> To: "Eyal Edri" <eedri@redhat.com> Cc: "Itamar Heim" <iheim@redhat.com>, infra@ovirt.org Sent: Wednesday, January 23, 2013 3:39:21 PM Subject: Re: Run jenkins test on a rhel6 host
On Tue, Jan 22, 2013 at 03:13:28PM -0500, Eyal Edri wrote:
updated both vms to 6.3
It looks like the nodes with rhel_62 are now offline. Maybe that needs to be fixed.
its not that, it's just the rhel_62 label was replaced with rhel_63. i changed the jobs that used it and it's running now.
Thanks.
I donate boten\tiggy for the cause. Have eedri configure them as Jenkins slaves. Problem solved ----- Original Message -----
From: "Dan Kenigsberg" <danken@redhat.com> To: "Mark Wu" <wudxw@linux.vnet.ibm.com>, "Saggi Mizrahi" <smizrahi@redhat.com> Cc: infra@ovirt.org Sent: Sunday, January 20, 2013 9:34:52 AM Subject: Re: Run jenkins test on a rhel6 host
On Thu, Jan 17, 2013 at 03:58:55PM +0800, Mark Wu wrote:
Hi guys,
I am a vdsm developer. Recently I submit a vdsm patch fix a break on fedora 18. I just verify it on fedora, but not on RHEL6 before merged. Unfortunately, my patch called a function which has a bug on rhel6. That bug was fixed in fedora, but not backported to rhel6. So I am thinking is it possible to also run the jenkins test on a rhel6 host for each patch? Then we can find the break in time. and the developer don't need to run the test cases on two systems for one patch.
I'd like to join Mark's plea. I believe that the only problem is of finding the resources for another Jenkins slave.
Missing an el6 slave is a pain point for me - I've accepted Saggi's http://gerrit.ovirt.org/9986 and now Vdsm's master branch fails to build on EL6 on the following grounds:
ERROR: testMethodBadParameters(reactorType='tcp') (jsonRpcTests.JsonRpcServerTests) ---------------------------------------------------------------------- Traceback (most recent call last): File "/home/mine/.rpm/BUILD/vdsm-4.10.3/tests/testrunner.py", line 65, in wrapper return f(self, *args) File "/home/mine/.rpm/BUILD/vdsm-4.10.3/tests/jsonRpcTests.py", line 273, in testMethodBadParameters JsonRpcInternalError().code) File "/usr/lib64/python2.6/contextlib.py", line 34, in __exit__ self.gen.throw(type, value, traceback) File "/home/mine/.rpm/BUILD/vdsm-4.10.3/tests/jsonRpcTests.py", line 97, in constructServer yield server, jsonClientFactory File "/home/mine/.rpm/BUILD/vdsm-4.10.3/tests/jsonRpcTests.py", line 269, in testMethodBadParameters with self.assertRaises(JsonRpcError) as cm: TypeError: failUnlessRaises() takes at least 3 arguments (2 given)
====================================================================== ERROR: testMethodMissingMethod(reactorType='tcp') (jsonRpcTests.JsonRpcServerTests) ---------------------------------------------------------------------- Traceback (most recent call last): File "/home/mine/.rpm/BUILD/vdsm-4.10.3/tests/testrunner.py", line 65, in wrapper return f(self, *args) File "/home/mine/.rpm/BUILD/vdsm-4.10.3/tests/jsonRpcTests.py", line 258, in testMethodMissingMethod JsonRpcMethodNotFoundError().code) File "/usr/lib64/python2.6/contextlib.py", line 34, in __exit__ self.gen.throw(type, value, traceback) File "/home/mine/.rpm/BUILD/vdsm-4.10.3/tests/jsonRpcTests.py", line 97, in constructServer yield server, jsonClientFactory File "/home/mine/.rpm/BUILD/vdsm-4.10.3/tests/jsonRpcTests.py", line 254, in testMethodMissingMethod with self.assertRaises(JsonRpcError) as cm: TypeError: failUnlessRaises() takes at least 3 arguments (2 given)
participants (6)
-
Dan Kenigsberg -
Ewoud Kohl van Wijngaarden -
Eyal Edri -
Itamar Heim -
Karsten 'quaid' Wade -
Saggi Mizrahi