Unsolved

This post is more than 5 years old

8 Posts

946

September 8th, 2008 08:00

Snap fails replicating, simulates fine

2008 09 08 11:03:17 DEV-SQL 000508 WARNING: Unable to find Navisphere CLI path. If CLARiiON arrays are being used, install Navisphere CLI into the default path: C:\Program Files\EMC\Navisphere CLI\. Navisphere CLI is not needed for other arrays.

2008 09 08 11:03:17 DEV-SQL Unable to find java path, which is needed if you are using a Release 16 array or this client does not support NaviSECCLI. Please put the java path in the environment variable "PATH" or "ERM_JAVA_PATH" on the client if necessary.

2008 09 08 11:03:17 DEV-SQL 000076 ERROR: The catalog header is invalid for version 5.

2008 09 08 11:03:17 DEV-SQL 004052 WARNING: Unable to find Invista CLI path. If Invista instances are being used, install InvCLI into the default path: C:\Program Files\EMC\INVCLI\.

2008 09 08 11:03:17 DEV-SQL 026084 ERROR: Storage Services operation setRestoreObject failed with an error as follows: Storage Services has terminated with an error.

2008 09 08 11:03:17 DEV-SQL 026152 ERROR: An internal error occurred on the Replication Manager host as follows: processMark - getTopOfTree failed.

2008 09 08 11:03:17 DEV-SQL 000071 ERROR: The mount activity has failed.

2008 09 08 11:03:24 packmanager 102023 ERROR: The mount of replica from application set SQL created at 2008 09 08 11:02:28 to host dev-sql failed.


Not sure what this all means. I have navisphere CLI installed on all hosts, does it need to be on the RM server? I do not believe it is there. The weird thing is it simulates fine with no warnings or errors. The host I am trying to mount it to is a VM. The SQL server I am trying to snap is the default instance, and the server I am trying to mount it to it is a named instance, named the original servername. I suppose this is the right way to do this? Any advice for an RM newb?

8 Posts

September 8th, 2008 10:00

I did not have the RMSP2 installed, so that was causing some problems I believe. Here are the new error messages, that seem to be progress. This seems like it could be related to the fact that it's on a VM, so I don't really have a LUN attached to this host, it is attached to the server the VM is running on. Can snaps be done to VMs? or just clones?

2008 09 08 13:45:49 DEV-SQL 000343 ERROR: Solutions Enabler function ClarPdevList failed with return code 3,595. Error message: No objects of the selected type were found.

2008 09 08 13:45:49 DEV-SQL 000343 ERROR: Solutions Enabler function ClarPdevList failed with return code 3,595. Error message: No objects of the selected type were found.

2008 09 08 13:45:49 DEV-SQL 000343 ERROR: Solutions Enabler function ClarPdevList failed with return code 3,595. Error message: No objects of the selected type were found.

2008 09 08 13:45:49 DEV-SQL 000382 ERROR: No storage group could be located for host DEV-SQL on CLARiiON APM00082000453. Add a storage group for DEV-SQL with at least one LUN from APM00082000453.

2008 09 08 13:45:49 DEV-SQL 000509 ERROR: Failed to enable one or more replication devices on the mount host. Replication devices that could be enabled will now be disabled.

2008 09 08 13:45:49 DEV-SQL 026223 ERROR: An unexpected internal error occurred: SS_base: executeMount_implementation: failed to import shadow copy.

2008 09 08 13:45:49 DEV-SQL Trying to disable all the replication devices that were enabled during the previous operation.

2008 09 08 13:45:49 DEV-SQL 026084 ERROR: Storage Services operation ExecuteMount failed with an error as follows: Storage Services has terminated with an error.

2008 09 08 13:45:49 DEV-SQL Shared resources are now available for another mount restore or unmount operation.

2008 09 08 13:45:49 DEV-SQL 026084 ERROR: Storage Services operation processExecute failed with an error as follows: Storage Services has terminated with an error.

Then followed by this a little later:

2008 09 08 13:45:49 DEV-SQL 000149 ERROR: Mount failed. Performing undo of storage operations.

2008 09 08 13:45:51 DEV-SQL 000071 ERROR: The mount activity has failed.

2008 09 08 13:45:54 packmanager 102023 ERROR: The mount of replica from application set SQL created at 2008 09 08 13:41:15 to host dev-sql failed.

8 Posts

September 8th, 2008 19:00

Well, I have it down to this now, after getting the raw data mapped LUN visible on the VM. Someone feel free to step in anytime! I don't have a host called dev-sql, because it's a VM. Is there some way I need to create a host in navisphere? Do I need to do something in my fabric manager? Pretty lost and new to this, any help is appreciated.


2008 09 08 22:17:51 DEV-SQL 000382 ERROR: No storage group could be located for host DEV-SQL on CLARiiON APM00082000453. Add a storage group for DEV-SQL with at least one LUN from APM00082000453.

2008 09 08 22:18:24 DEV-SQL 000509 ERROR: Failed to enable one or more replication devices on the mount host. Replication devices that could be enabled will now be disabled.

2008 09 08 22:18:24 DEV-SQL 026223 ERROR: An unexpected internal error occurred: SS_base: executeMount_implementation: failed to import shadow copy.

2008 09 08 22:18:24 DEV-SQL Trying to disable all the replication devices that were enabled during the previous operation.

2008 09 08 22:19:04 DEV-SQL 026084 ERROR: Storage Services operation ExecuteMount failed with an error as follows: Storage Services has terminated with an error.

2008 09 08 22:19:04 DEV-SQL Shared resources are now available for another mount restore or unmount operation.

2008 09 08 22:19:04 DEV-SQL 026084 ERROR: Storage Services operation processExecute failed with an error as follows: Storage Services has terminated with an error.

Message was edited by:
texrob

257 Posts

September 9th, 2008 00:00

Hi Texrob,

Maybe if you could let us know what youare trying to achieve and in what type of environment, host OS, etc, it would better help us advise you.

I do recommend you look at the specifics in the user guide about mounting to an ESX host, as there are a number of steps which need to be performed prior to attempting to mount.

cheers
James.

8 Posts

September 10th, 2008 06:00

Thanks for the response.

The host is an HP BL460 blade in a C7000 chassis. It is a dual proc, quad core, 16gb of RAM blade. It is running two drives mirrored internally, with the ESX hypervisor installed onto those internal drives, so there is no host OS, or rather ESX is the host OS. It is ESX 3.5 update 2 with the time bomb patch applied, go VMWare!

I am trying to use RM to replicate a snap of our SQL databases to a VM host, for reporting purposes.

As far as the steps, I have done all the steps in the RM admin guide for ESX. I have created the snap devices in the storage group, and discovered the new LUNs in the infrastructure client and mapped them as RDM to the host. I am able to see the LUNs in the host OS in disk management in windows (server 2003 R2). RM writes to the tempundodir fine, it just seems to think it's not communicating properly.

Can I not do a SQL specific RM to a VM host? Do I need to try and just do a RAW lun snap? I hope not, because I really need it to mount automatically and be SQL aware.

257 Posts

September 11th, 2008 00:00

Hi Texrob,

OK, the error "000382 ERROR: No storage group could be located for host" means that for reasons unknown (as yet), solutions enabler cannot find a match for the VM host on the Clariion. This could be for many reasons.

In order to continue to debug this, we would need the pertinent section of the RM debug log files at the time of error.

I would recommend that you set the mount host in debug mode in the console, run the mount again and open a support call so they can diagnose the issue in far more detail that here.

But what I would ask you to do is come back and share your results with the rest of us.

Cheers
James

8 Posts

September 11th, 2008 06:00

I will have to do that. I was just trying to hit this thing when I have time, as it's not a top priority short term. I'll get a ticket opened at some point here soon.

0 events found

No Events found!

Top