Hi Nagu,

The *reboot* seems more than enough, because amfd actually can finalize 
handle after a few retries and re-init again.
As your view, we can accept clm to return TRYAGAIN/TIMEOUT from stopping 
track during switchover but it's supposed to succeed as amfd has become 
standby.
Otherwise, we can just add a log error (not call exit()) and investigate 
why clm returns TRYAGAIN/TIMEOUT when switchover is done.
How's about that?

Thanks,
Minh

On 01/12/16 20:19, Nagendra Kumar wrote:
> Hi Gary,
>> I’m not sure if reboot the standby node is acceptable. It might be standby 
>> for AMF, but active for other SGs.
> It is very rare to come upto the point that Standby Amfd has to reboot the 
> node.
> The CLM stop API should succeed in normal situations(When it has become 
> Standby) if it fails during transition(Act to Quiesced).
> I am also saying to retry if it doesn't stop for couple of times.
> I don't find any issue with the approach.
>
> Thanks
> -Nagu
>> -----Original Message-----
>> From: Gary Lee [mailto:gary....@dektech.com.au]
>> Sent: 01 December 2016 14:26
>> To: Nagendra Kumar
>> Cc: minh chau; Praveen Malviya; hans.nordeb...@ericsson.com;
>> long.hb.ngu...@dektech.com.au; opensaf-devel@lists.sourceforge.net
>> Subject: Re: [PATCH 1 of 1] AMFD: Do not let non-active AMFD receives CLM
>> track callback [#2141]
>>
>> Hi guys
>>
>>
>>> On 1 Dec. 2016, at 4:24 pm, Nagendra Kumar <nagendr...@oracle.com>
>> wrote:
>>> Hi Minh,
>>>     But fixing small thing is complicating the code. You can just try to
>> stop CLM track again in Quisced->Standby transition and even if it fails you
>> may try few times as it is Standby and it is not doing much job accept that 
>> it is
>> receiving some checkpoints. If that fails, it is ok to reboot the Standby 
>> node.
>> I am sure this is very rare case and the added code may not be getting
>> chance to get executed.
>>
>> I’m not sure if reboot the standby node is acceptable. It might be standby 
>> for
>> AMF, but active for other SGs.
>>
>>> Also, it is important to investigate why CLM is giving try-again.
>>>
>>> Thanks
>>> -Nagu
>> Thanks
>> Gary
>>


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most 
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
_______________________________________________
Opensaf-devel mailing list
Opensaf-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/opensaf-devel

Reply via email to