+ netvirt-dev So far we weren’t able to reproduce the issue when statistics are disabled (still undergoes testing), but I’m not sure what effects it might have on the netvirt/genius code.
I’ve seen some Flow & Group listeners in genius/netvirt (If I understand correctly StatisticsGatheringUtils.java is creating/updating those entities – not sure whether there are others). I think the genius ones are dead code, but the one in netvirt seems like something that needs to be looked into (l2gw’s ElanGroupListener.java). I’m not really familiar with the l2gw implementation, maybe others can provide an input.. Thanks, Koby From: Abhijit Kumbhare [mailto:[email protected]] Sent: Wednesday, December 14, 2016 7:11 PM To: Shuva Jyoti Kar <[email protected]> Cc: Kochba, Alon <[email protected]>; An Ho <[email protected]>; Aizer, Koby <[email protected]>; Vishal Thapar <[email protected]>; openflowplugin-dev <[email protected]>; Anil Vishnoi <[email protected]>; Luis Gomez <[email protected]>; Colin Dixon <[email protected]>; [email protected]; Jozef Bacigal <[email protected]>; Sam Hague <[email protected]> Subject: Re: openflowplugin blocker bug So we will continue investigating this in the coming days but the fix will most likely be in the SR3 timeframe. Till then the workaround of turning off stats as mentioned by Shuva might be the option. On Tue, Dec 13, 2016 at 5:38 AM, Shuva Jyoti Kar <[email protected]<mailto:[email protected]>> wrote: Hi Alon, As discussed with Koby over IRC today, with statistics turned off we mightnot see this issue. As I understand netvirt doesnot(as of now depend much on OfPlugin statistics) hence with stats off if the issue is not reproduced, I think we can consider that to be a valid workaround for SR2. Since the primary way of reproducing the issue involves bringing interfaces up/down iteratively the rca might take some involvement from both the teams – netvirt and ofplugin. Thanks, -shuva From: Kochba, Alon [mailto:[email protected]<mailto:[email protected]>] Sent: Tuesday, December 13, 2016 5:36 PM To: Abhijit Kumbhare; An Ho Cc: openflowplugin-dev; Anil Vishnoi; Luis Gomez; Colin Dixon; [email protected]<mailto:[email protected]>; Jozef Bacigal; Shuva Jyoti Kar; Sam Hague Subject: RE: openflowplugin blocker bug Hi Abhijit, I commented on the bug [1]. This bug is very problematic in that it leaves the DPN unusable, though it appears as available so VMs are allocated to it. It has been encountered several times and reproduction details/info was provided, would appreciate your feedback on the bug itself. Due to the nature of this bug I think it's important we at least analyze the root cause and provide a workaround for SR2. [1] https://bugs.opendaylight.org/show_bug.cgi?id=6908 Thanks, --alon From: Abhijit Kumbhare [mailto:[email protected]] Sent: Tuesday, 13 December 2016 04:04 To: An Ho <[email protected]<mailto:[email protected]>>; Kochba, Alon <[email protected]<mailto:[email protected]>> Cc: openflowplugin-dev <[email protected]<mailto:[email protected]>>; Anil Vishnoi <[email protected]<mailto:[email protected]>>; Luis Gomez <[email protected]<mailto:[email protected]>>; Colin Dixon <[email protected]<mailto:[email protected]>>; [email protected]<mailto:[email protected]>; Jozef Bacigal <[email protected]<mailto:[email protected]>>; Shuva Jyoti Kar <[email protected]<mailto:[email protected]>> Subject: Re: openflowplugin blocker bug If I am not mistaken - I think Anil had backed out the following change: https://bugs.opendaylight.org/show_bug.cgi?id=6917. Not sure if it got merged Anil? About https://bugs.opendaylight.org/show_bug.cgi?id=6908 - added Alon (as he has raised the priority to blocker yesterday). Alon - would like to check with you if you think it is OK to defer this bug (6908) to the next SR since we need to get this SR out and the investigation/fix by Shuva/Jozef may need a few days? On Mon, Dec 12, 2016 at 5:20 PM, An Ho <[email protected]<mailto:[email protected]>> wrote: Hi OPENFLOWPLUGIN Team, We would like to kindly request an update on these OPENFLOWPLUGIN Boron Blocker Bugs? Please let us know the ETA for a fix or if we should retarget the blocker bug for Boron-SR3 instead. Bug 6908 - failed to update port status - all new OVS ports fail from this point. Is this only for Carbon or does it apply to Boron-SR2 as well? Otherwise, we should retarget for Carbon. Bug 6917 - Flow matching function (operational flow reconciliation) is not stable. Luis reopened this bug on 12/08. Are there plans to fix it now, or can we retarget for Boron-SR3 for more time? Best Regards, An Ho [1] https://bugs.opendaylight.org/show_bug.cgi?id=6908 [2] https://bugs.opendaylight.org/show_bug.cgi?id=7343
_______________________________________________ openflowplugin-dev mailing list [email protected] https://lists.opendaylight.org/mailman/listinfo/openflowplugin-dev
