Re: After rebooting the both nodes the dependent service is down on on of them
- From: yulkayulka@xxxxxxxxx
- Date: 21 Sep 2006 02:25:36 -0700
I just want to note that in the case above windows doesn't run the
service for the first time because the cluster service failed to run.
On the the second try cluster service succeeds to start, but for an
unknown reason windows doesn't start the dependent service.
yulkayu...@xxxxxxxxx wrote:
The dependent service I'm talking about it the application we are
developing. If it's installed to work in cluster, it fails if cluster
is down. It should be up after the MSCS service is up.
It's not running just on node B, node where the cluster failed to run
for the first time.
Edwin vMierlo wrote:
so, on both nodes the cluster service is running, but your "dependent
service" is not running
Some Questions for you:
1) what is the "dependent service" ? e.g what application is it ?
2) is it not running on node A ? not running on node B (10 min delay) ? or
not running on both nodes ?
3) why does this "dependent service" to be dependent on the cluster service
?
<yulkayulka@xxxxxxxxx> wrote in message
news:1158827891.228641.184460@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Hi,to
The solution you've suggested doesn't work. I increased the "Time to
display list of operating systems" to 10 minutes and rebooted both of
the cluster machines, after they are up the same error in the event
manager appeared on the node with the reboot delay and the dependent
service is down.
Is there any other solution you can think of?
Thanks, Yulia.
yulkayulka@xxxxxxxxx wrote:
I have an understanding why cluster doesn't start at one of the nodes,
but it seems to me a bug that when the cluster restarted after the
first failure, windows doesn't restart the dependent service.
Thanks, Yulia
John Toner [MVP] wrote:
FYI, the reason that this happens is because both nodes are attempting
time,start at the same time as the others have hinted. When cluster service
starts, it attempts to arbitrate for the quorum device using SCSI
reserve/release/reset commands. If both nodes are starting at the same
sometimesthey will interrupt each other's arbitration attempt and one or
process inboth nodes will fail to start cluster service.
If you ever have trouble sleeping, you can read more about this
the following KB article: http://support.microsoft.com/kb/309186.
Regards,
John
<yulkayulka@xxxxxxxxx> wrote in message
news:1158758953.897029.311600@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Right, they are rebooted at the same time.
This is an important case that could happen for example during
electricity failure
.
- References:
- Re: After rebooting the both nodes the dependent service is down on on of them
- From: John Toner [MVP]
- Re: After rebooting the both nodes the dependent service is down on on of them
- From: yulkayulka
- Re: After rebooting the both nodes the dependent service is down on on of them
- From: yulkayulka
- Re: After rebooting the both nodes the dependent service is down on on of them
- From: Edwin vMierlo
- Re: After rebooting the both nodes the dependent service is down on on of them
- From: yulkayulka
- Re: After rebooting the both nodes the dependent service is down on on of them
- Prev by Date: Re: After rebooting the both nodes the dependent service is down on on of them
- Next by Date: Re: After rebooting the both nodes the dependent service is down on on of them
- Previous by thread: Re: After rebooting the both nodes the dependent service is down on on of them
- Next by thread: Re: After rebooting the both nodes the dependent service is down on on of them
- Index(es):
Relevant Pages
|