On 2008-07-14T12:27:58, HIDEO YAMAUCHI <[EMAIL PROTECTED]> wrote:
> Hi All,
>
> I confirmed it about watchdog of Heartbeat.
>
> The environment that I confirmed has two resources in two
> nodes(active/standby).
>
> The resource of one eyes checks starting VIP.
> The resource of two eyes is VIP(IPaddr).
>
> In the next sequence, a phenomenon happened.
>
> 1)I start Heartbeat.(version 2.1.3)
> 2)Two resources start in the first node.
> 3)The second node is a DC node.
> 4)In the first node, I do kill of a master process.
> 5)The second node detects the death of the first node.
> 6)And, in the second node, a resource of VIP of a check starts.
> but watchdog does not yet exercise it in the first node.
> 7)Fail in the start of the check resource of VIP in the second node by a
> delicate timing.
>
> Because the interval of watchdog becomes deadtime+10ms.
>
> I turn on a delay in start time(start_delay) of the first resource, and I
> solve this problem.
> This is delay time to wait for complete of watchdog.
>
> Is not the interval of right watchdog shorter than deadtime?
I'm sorry, I'm not sure I understand what you're saying.
The watchdog timer is not a replacement for stonith; step 6+7 then would
be prevented.
But yes, it appears that the watchdog timeout should be less than
deadtime.
Regards,
Lars
--
Teamlead Kernel, SuSE Labs, Research and Development
SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nürnberg)
"Experience is the name everyone gives to their mistakes." -- Oscar Wilde
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
