Dear all,
I am trying to understand the problem that we had with our 2 nods mscs cluster last week.
It seems that we had a temporary public network failure on one node and then all the ip address goes offline for a moment.
The problem affect only one node (srvxxxxA)
The resources hosted on node srvxxxxxA was:
Cluster Group
SQL Server Group
File Share Group
We had a problem only with File Share Group.
Inside this group we have 4 share, 3 of them are configured with the default parameter "affect the group".
The last one do not have this advanced parameter set.
Attached you will find an extract of the cluster log:
This is a very strange situation because SQL Server was online and running perfect on node SRVxxxxA.
With the "shares" we had some problems and we end up in shutting down node SRVxxxxA in order to failover the group to the other node because all the shares was unreachable.
To me the problem may be this one:
1) the cluster takes offline the 3 shares that have the advanced properties "affect the group" set.
2) the cluster takes offline the last share that do not have the advanced properties "affect the group" set.
3) the cluster terminates his activity of failover because the last share inform to not affect the group , so the failover process stops.
Is this possible?
Please advise.
I am trying to understand the problem that we had with our 2 nods mscs cluster last week.
It seems that we had a temporary public network failure on one node and then all the ip address goes offline for a moment.
The problem affect only one node (srvxxxxA)
The resources hosted on node srvxxxxxA was:
Cluster Group
SQL Server Group
File Share Group
We had a problem only with File Share Group.
Inside this group we have 4 share, 3 of them are configured with the default parameter "affect the group".
The last one do not have this advanced parameter set.
Attached you will find an extract of the cluster log:
This is a very strange situation because SQL Server was online and running perfect on node SRVxxxxA.
With the "shares" we had some problems and we end up in shutting down node SRVxxxxA in order to failover the group to the other node because all the shares was unreachable.
To me the problem may be this one:
1) the cluster takes offline the 3 shares that have the advanced properties "affect the group" set.
2) the cluster takes offline the last share that do not have the advanced properties "affect the group" set.
3) the cluster terminates his activity of failover because the last share inform to not affect the group , so the failover process stops.
Is this possible?
Please advise.
