Suja,
   We have reverted the changes to use PortReason flag in master as well, as 
the current implementation is buggy and we have tests failing because of that.
Thanks,
Faseela

From: Faseela K
Sent: Saturday, March 10, 2018 1:31 AM
To: Suja T <suj...@ericsson.com>; D Arunprakash <d.arunprak...@ericsson.com>; 
'openflowplugin-dev' <openflowplugin-dev@lists.opendaylight.org>
Cc: 'genius-...@lists.opendaylight.org' <genius-...@lists.opendaylight.org>; 
'odl netvirt dev' <netvirt-...@lists.opendaylight.org>
Subject: RE: PortReason Flag not set properly during port deletes

Just an FYI.
Suja was later on able to reproduce the issue, with some openflowplugin logs 
enabled.
@Suja : Please update if you have some conclusion on the root cause.

Thanks,
Faseela

From: Faseela K
Sent: Friday, March 09, 2018 12:31 PM
To: Suja T <suj...@ericsson.com<mailto:suj...@ericsson.com>>; D Arunprakash 
<d.arunprak...@ericsson.com<mailto:d.arunprak...@ericsson.com>>; 
openflowplugin-dev 
<openflowplugin-dev@lists.opendaylight.org<mailto:openflowplugin-dev@lists.opendaylight.org>>
Cc: 
genius-...@lists.opendaylight.org<mailto:genius-...@lists.opendaylight.org>; 
odl netvirt dev 
<netvirt-...@lists.opendaylight.org<mailto:netvirt-...@lists.opendaylight.org>>
Subject: RE: PortReason Flag not set properly during port deletes

Suja,

   The logs which you have referred somehow gets these test cases passing, and 
that is when Reason flag is coming properly. If you check the run below where 
the test-case fails, you can see the PortReason flag coming as "Update".


https://logs.opendaylight.org/releng/vex-yul-odl-jenkins-1/netvirt-csit-1node-openstack-queens-gate-stateful-oxygen/50/odl_1/odl1_karaf.log.gz
  ==> tapcad3a2b8-50

I have triggered some more runs, let us see whether we will hit the error or 
not.


Thanks,
Faseela

From: Suja T
Sent: Friday, March 09, 2018 8:06 AM
To: Faseela K <faseel...@ericsson.com<mailto:faseel...@ericsson.com>>; D 
Arunprakash <d.arunprak...@ericsson.com<mailto:d.arunprak...@ericsson.com>>; 
openflowplugin-dev 
<openflowplugin-dev@lists.opendaylight.org<mailto:openflowplugin-dev@lists.opendaylight.org>>
Cc: 
genius-...@lists.opendaylight.org<mailto:genius-...@lists.opendaylight.org>; 
odl netvirt dev 
<netvirt-...@lists.opendaylight.org<mailto:netvirt-...@lists.opendaylight.org>>
Subject: RE: PortReason Flag not set properly during port deletes

Hi Faseela,
We have added logs in openflowplugin and we started CSIT to identify whether we 
are receiving the proper reason and updating the reason flag with exact reason.
When there is actual port delete we are submitting the transaction with the 
reason as delete for both update and remove event.
I have verified this scenario for the problematic port which you have mentioned 
for the failing test case.  we are providing the flag with proper reason.

Herewith, the link where we have run CSIT in stable/oxygen and master, and 
given the port name for which I verified this scenarios:

https://logs.opendaylight.org/releng/vex-yul-odl-jenkins-1/netvirt-csit-1node-openstack-queens-gate-stateful-oxygen/57/odl_1/odl1_karaf.log.gz
  port name: tap7de20c9a-42

https://logs.opendaylight.org/releng/vex-yul-odl-jenkins-1/netvirt-csit-1node-openstack-queens-gate-stateful-fluorine/34/odl_1/odl1_karaf.log.gz
 port name: tapa504adbf-93

Regards,
Suja
From: Faseela K
Sent: Thursday, March 08, 2018 2:02 PM
To: Suja T <suj...@ericsson.com<mailto:suj...@ericsson.com>>; D Arunprakash 
<d.arunprak...@ericsson.com<mailto:d.arunprak...@ericsson.com>>; 
openflowplugin-dev 
<openflowplugin-dev@lists.opendaylight.org<mailto:openflowplugin-dev@lists.opendaylight.org>>
Cc: 
genius-...@lists.opendaylight.org<mailto:genius-...@lists.opendaylight.org>; 
odl netvirt dev 
<netvirt-...@lists.opendaylight.org<mailto:netvirt-...@lists.opendaylight.org>>
Subject: PortReason Flag not set properly during port deletes

Hello openflowplugin-devs,
  We had started using the PortReason Flag given by openflowplugin to 
efficiently handle DPN disconnect scenarios in Genius.
  However in certain cases this flag is not coming with the right reason for 
genuine port delete scenarios, and there were failures in netvirt CSIT due to 
this.
  We had to revert the patch in Genius,  and we would like to have a solution 
for the issue, so that the patch can go back in Oxygen, as this is an important 
robustness fix.
  Let me know whether we should raise a JIRA for this.
  We have run netvirt CSIT only for the failing CSIT with respective TRACES 
enabled, the problematic port in the below log is "tapcad3a2b8-50".
  
https://logs.opendaylight.org/releng/vex-yul-odl-jenkins-1/netvirt-csit-1node-openstack-queens-gate-stateful-oxygen/50/odl_1/odl1_karaf.log.gz

Thanks,
Faseela
_______________________________________________
openflowplugin-dev mailing list
openflowplugin-dev@lists.opendaylight.org
https://lists.opendaylight.org/mailman/listinfo/openflowplugin-dev

Reply via email to