cancel
Showing results for 
Search instead for 
Did you mean: 

New fileshare ressource fail on the second node

Hello guys,

I need your help about a new fileshare that I create on a cluster ressource.

I created the new fileshare as the old ones already excist on the same ressource.

When I try to do a failover to the second node, the old fileshares bring online.

But the new one no, and that's make the ressource in faulted state.

I transfer the ressource to the first node, and all fileshares bring online including the new created.

Conclusion : the fileshare can brings online on the first node but fails on the second node.

I checked the configuration between the olds fileshare and this new one, and I didn't find any difference.

Can you please help me about this issue.

Thank you in advance.

 

 

 

 

1 Solution

Accepted Solutions
Accepted Solution!

Hi Jimb2k, Error 3 is path

Hi Jimb2k,

Error 3 is path not found and error 2310 is share does not exist.

I would ignore the error 2310 as it is part of the cleanup process.  If the folder did not exist or could not be found during the online entry point, then you would expect the clean entry point to not find the share on the server.

You can bring up the VMDg and MountV resource alone on the second node. Then check to see if the D:\Testshare folder is visable on that node. 

If it is there, then check NTFS permissions on the folder.  The folder will need to have permissions set so that the "Local System" account has at least read access to the folder.  If the folder does not have this permission listed then add it to the ACL.

FYI - "Local System" is the account that HAD runs as.  If you have changed the startup account that HAD is running with then you should change it back to "Local System".

Let us know if this helps.

Thank you,

Wally

View solution in original post

5 Replies

Hi Jimb2k, On the second

Hi Jimb2k,

On the second node, please check the FileShare_A.txt log that is located in the %vcs_home%\log folder.  It should have so details on why the resource faulted when you attempted to online it on that node.  If you don't see anything that stands out to you, please share the entries that are made during the failed online attempt and we might be able to point you in the right direction.

Thank you,

Wally

Please provide details of

Please provide details of change:

Extract from main.cf as it is now and details of what you have changed (perhaps post extract of previous main.cf)

Extract from engine_A.log when resource fails to online on second node and also include where it succeeds on first node.

Mike

Hello Guys, This is what I

Hello Guys,

This is what I found on the logs as asked :

2015/08/09 18:38:38 VCS ERROR V-16-10051-10505 FileShare:TEST-FileShare-1:online:Failed to open folder D:\Testshare [14:3]
2015/08/09 18:40:40 VCS ERROR V-16-2-13066 Thread(10644) Agent is calling clean for resource(TEST-FileShare-1) because the resource is not up even after online completed.
2015/08/09 18:40:40 VCS ERROR V-16-10051-10519 FileShare:TEST-FileShare-1:clean:Failed to unshare folder D:\Testshare as (\\FileShare\Testshare) [2:2310]
2015/08/09 18:40:40 VCS ERROR V-16-2-13071 Thread(10644) Resource(TEST-FileShare-1): reached OnlineRetryLimit(0).

 

The configuration of the cluster is based on SAN so the failover take the same data on the same disk.

Thank you for your help.

1. Is the dir name correctly

1. Is the dir name correctly mentioned?

2. Does the user have sufficient permission to access the directory on the node where the issue occurs? Are you able to access the directory manually with the particualr user?

3. Are you able to manually share the folder outside VCS control?

4. Can you provide main.cf extracts and directory permissions from both the nodes?

 

Regards,

Sudhir
 

Accepted Solution!

Hi Jimb2k, Error 3 is path

Hi Jimb2k,

Error 3 is path not found and error 2310 is share does not exist.

I would ignore the error 2310 as it is part of the cleanup process.  If the folder did not exist or could not be found during the online entry point, then you would expect the clean entry point to not find the share on the server.

You can bring up the VMDg and MountV resource alone on the second node. Then check to see if the D:\Testshare folder is visable on that node. 

If it is there, then check NTFS permissions on the folder.  The folder will need to have permissions set so that the "Local System" account has at least read access to the folder.  If the folder does not have this permission listed then add it to the ACL.

FYI - "Local System" is the account that HAD runs as.  If you have changed the startup account that HAD is running with then you should change it back to "Local System".

Let us know if this helps.

Thank you,

Wally

View solution in original post