"Green w/ Veritas" wrote:
>
>Has anybody run into an issue running VCXS 3.5 on Sol 9 and getting "monitor
>procedure did not complete within the expected time"? It seems to hang the
>node b/c I can't even run format after that...The second node on the cluster
>does not have this issue..
>
>Thanks


I mean VCS 3.5

ok, let's explain what veritas cluster server does first:

1. it can online your resources
2. it can offline resources
3. it can monitor (tell if it is online or offline) the resources.


How does it online ? offline and monitor ?

Mostly with scripts.

If you go and check out /opt/VRTSvcs/bin

You will see a directory for each resource type. So say you want to go
to the IP resource, just change to the IP directory.

In the directory, you will find Perl scripts to online, offline, monitor
and clean.

OK, so now that we've determined that VCS will run scripts, why will a
script timeout ?

For a monitor script, VCS will allow 60 seconds (by default) to complete.

So, if your monitor script did not complete within 60 seconds, could be
because your machine was too busy ?

As you say that you could not even run format, perhaps something was
wrong on the disks on the system, slowing the whole system (not only
VCS) right down !!!

So, the problem was not VCS, but rather something else on the system.


Go and have a look in the syslog file (depends on your OS where this is)
to find out if something else was happening on your system at that time,
which would have kept everything (not only VCS) so busy that even
running s little script could not complete in 60 seconds

Green w/ veritas wrote:
> "Green w/ Veritas" wrote:
>
>>Has anybody run into an issue running VCXS 3.5 on Sol 9 and getting "monitor
>>procedure did not complete within the expected time"? It seems to hang the
>>node b/c I can't even run format after that...The second node on the cluster
>>does not have this issue..
>>
>>Thanks
>
>
>
> I mean VCS 3.5

Thanks for your input.

I came across the same problem on my cluster running Solaris too.

VCS ERROR V-16-2-13027 (hostname) Resource (diskgroup name) - monitor procedure
did not complete within the expected time.

However, no other relevant error (eg. SAN disks connection error) appeared
on /var/adm/messages. VCS just brought offline the resource and deport the
dgs.

Anyone can advise ?
Thanks a lot




Me wrote:
>ok, let's explain what veritas cluster server does first:
>
>1. it can online your resources
>2. it can offline resources
>3. it can monitor (tell if it is online or offline) the resources.
>
>
>How does it online ? offline and monitor ?
>
>Mostly with scripts.
>
>If you go and check out /opt/VRTSvcs/bin
>
>You will see a directory for each resource type. So say you want to go
>to the IP resource, just change to the IP directory.
>
>In the directory, you will find Perl scripts to online, offline, monitor

>and clean.
>
>OK, so now that we've determined that VCS will run scripts, why will a
>script timeout ?
>
>For a monitor script, VCS will allow 60 seconds (by default) to complete.
>
>So, if your monitor script did not complete within 60 seconds, could be

>because your machine was too busy ?
>
>As you say that you could not even run format, perhaps something was
>wrong on the disks on the system, slowing the whole system (not only
>VCS) right down !!!
>
>So, the problem was not VCS, but rather something else on the system.
>
>
>Go and have a look in the syslog file (depends on your OS where this is)

>to find out if something else was happening on your system at that time,

>which would have kept everything (not only VCS) so busy that even
>running s little script could not complete in 60 seconds
>
>Green w/ veritas wrote:
>> "Green w/ Veritas" wrote:
>>
>>>Has anybody run into an issue running VCXS 3.5 on Sol 9 and getting "monitor
>>>procedure did not complete within the expected time"? It seems to hang
the
>>>node b/c I can't even run format after that...The second node on the cluster
>>>does not have this issue..
>>>
>>>Thanks
>>
>>
>>
>> I mean VCS 3.5