An,
I answered below inline.
BTW the patch has been reviewed and given a +2, so its ready to be merged.
Thanks,
Brady
On 28/10/16 19:30, An Ho wrote:
Hi Brady,
To better assess the impact of this patch 47739, could someone from
your team please help us identify the following:
Regression: Is this bug a regression of
functionality/performance/feature compared to Boron?
No it is not, its been a failure for some time, and we just found it in
OPNFV.
Severity: Could you elaborate on the severity of this bug? Is this a
BLOCKER such that we cannot release Boron-SR1 without it? Is there a
workaround such that we can write a release note?
This is quite severe. With just a simple modification to a Service
Function Forwarder (SFF) or Service Function (SF) all related Rendered
Service Paths (RSPs, the actual service chains) are deleted. The RSP
should only be deleted if either the SFF is deleted, the SF is deleted,
or either is modified in such a way that the RSP is affected.
Testing: Could you also elaborate on the testing of this patch? How
extensively has this patch been tested? Is it covered by any unit
tests or system tests?
Unit Tests were modified and added to ODL to test this patch. I also
created a distro to test this patch in OPNFV, which is quite extensive.
Impact: Does this fix impact any dependent projects?
No
Best Regards,
An Ho
[1] https://git.opendaylight.org/gerrit/#/c/47739/
[2] https://bugs.opendaylight.org/show_bug.cgi?id=7039
*From:*Brady Allen Johnson [mailto:[email protected]]
*Sent:* Friday, October 28, 2016 8:40 AM
*To:* Brady Johnson; An Ho
*Cc:* [email protected]; sfc-dev opendaylight;
[email protected]
*Subject:* Re: [release] [sfc-dev] Netvirt Blocker bug: Problems with
the netvirt classifier in SFC
Here is the patch in stable/boron:
https://git.opendaylight.org/gerrit/#/c/47739/
And here is the associated bug:
https://bugs.opendaylight.org/show_bug.cgi?id=7039
This is being tested in OPNFV now. Once successful, we'll merge the
patch and close the bug.
Regards,
Brady
On 27/10/16 18:17, Brady Johnson wrote:
We did some more testing with more debug logs and found the problem.
The rsp (rendered service path, the actual service chain) is being
created correctly, but we see it gets deleted 1 second later,
before the netvirt classifier is created, thus the log message by
netvirt that the rsp doesn't exist.
We investigated further to see why/who deletes the rsp. We figured
out that tacker creates everything correctly for the first rsp and
then starts creating a second rsp. When tacker is creating the
second rsp, it modifies the sff (service function forwarder),
which currently causes SFC to delete the rsp.
If the sff is deleted or if certain fields are modified, the rsp
should be deleted. But in this case, the sff mods are minimal and
the rsp should not be deleted. So the fix to sfc is when an sff is
modified, only delete the rsp when absolutely necessary.
I can have this fix ready on Friday mid-day GMT+1. Will this still
be in time for boron SR1?
Regards,
Brady
On Oct 26, 2016 23:05, "An Ho" <[email protected]
<mailto:[email protected]>> wrote:
Hi Brady and SFC Team,
Would your team be able to provide the full logs with trace
enabled to help address Sam’s concerns with missing RSP, etc.
Best Regards,
An Ho
*From:*Sam Hague [mailto:[email protected]
<mailto:[email protected]>]
*Sent:* Wednesday, October 26, 2016 1:44 PM
*To:* An Ho
*Cc:* Brady Allen Johnson; Robert Varga; Manuel Buil;
[email protected]
<mailto:[email protected]>;
[email protected]
<mailto:[email protected]>; sfc-dev opendaylight
*Subject:* Re: [release] Netvirt Blocker bug: Problems with the
netvirt classifier in SFC
I reassigned this bug back to sfc. netvirt can't do anything
without an RSP and netvirt does not create RSP's either. That
needs to be debugged why the RSP isn't there - either wrong name
given or simply doesn't exist at the time the acl is requested.
Full logs with trace enabled for sfc and netvirt should provide
the info to find the problem.=
On Wed, Oct 26, 2016 at 1:49 PM, An Ho <[email protected]
<mailto:[email protected]>> wrote:
Hi Sam and NETVIRT Team,
Could you please provide us some insight for bug 7039 [1]. Does
your team have plans to provide a fix for Boron-SR1? What is the
ETA for merging the fix or should we retarget the bug for
Boron-SR2 instead? Please note that a dependent end user (OPNFV
Colorado 2.0) will be impacted by this blocker issue.
Best Regards,
An Ho
[1] https://bugs.opendaylight.org/show_bug.cgi?id=7039
-----Original Message-----
From: [email protected]
<mailto:[email protected]>
[mailto:[email protected]
<mailto:[email protected]>] On Behalf Of
Brady Allen Johnson
Sent: Wednesday, October 26, 2016 10:17 AM
To: Robert Varga; Manuel Buil; [email protected]
<mailto:[email protected]>;
[email protected] <mailto:[email protected]>
Subject: Re: [release] Netvirt Blocker bug: Problems with the
netvirt classifier in SFC
Robert,
We wanted to get this fix into OPNFV Colorado 2.0 which releases
October 27, but I doubt that will be possible. OPNFV Colorado 3.0
is scheduled for December 1, but ODL Boron SR2 will release
December 15, so if the fix isnt in ODL Boron SR2, we'll completely
miss OPNFV Colorado if we want to use an "official" build.
If it wont be possible to get this fix in ODL Boron SR1, then
we'll have to use a distro taken from ODL Boron post SR1.
Personally I dont mind doing this, but this makes others in OPNFV
nervous, since these sorts of builds are usually "permanent".
Manuel did a great analysis of this problem. He even gave stack
traces of where the problem is. Ive been at the OpenStack summit
and havent had time to try to fix it. If we could get an idea of
how complicated it would be to fix the problem, that would
probably help us decide how to proceed. If its a simple fix, then
maybe we could try to get it into SR1. Otherwise we'll just have
to go with an ODL post SR1 build.
Thanks,
Brady
On 26/10/16 17:27, Robert Varga wrote:
> On 10/26/2016 04:30 PM, Brady Allen Johnson wrote:
>> Hello,
>>
>> We reported this problem to the Netvirt email list on October
21 and
>> got no response.
>>
>> I consider this to be a blocker for OPNFV SFC as it is causing
CI failures.
>>
>> Here is the bugzilla report:
>>
>> https://bugs.opendaylight.org/show_bug.cgi?id=7039
>>
> Hello Brady,
>
> would this be an SR1 blocker or can it be postponed to SR2?
>
> Thanks,
> Robert
>
_______________________________________________
release mailing list
[email protected] <mailto:[email protected]>
https://lists.opendaylight.org/mailman/listinfo/release
_______________________________________________
sfc-dev mailing list
[email protected] <mailto:[email protected]>
https://lists.opendaylight.org/mailman/listinfo/sfc-dev
_______________________________________________
release mailing list
[email protected] <mailto:[email protected]>
https://lists.opendaylight.org/mailman/listinfo/release
_______________________________________________
sfc-dev mailing list
[email protected]
https://lists.opendaylight.org/mailman/listinfo/sfc-dev
