Puredisk content router process (spoold) failing to startup.
| Article:TECH169920 | | | Created: 2011-09-20 | | | Updated: 2012-03-08 | | | Article URL http://www.symantec.com/docs/TECH169920 |
Problem
The spoold process is not starting. After running a PureDisk services startup "/etc/init.d/puredisk start" the spoold process shows as "failed".
Starting PureDisk ContentRouter: spoold failedError
Checking the /Storage/log/spoold/spoold.log on the content router we can see this error message
August 23 12:47:03 INFO [46992284444128]: Database Manager: initializing
August 23 12:47:03 INFO [46992284444128]: Database Manager: initialization complete
August 23 12:47:03 INFO [46992284444128]: Storage Cache Manager: Initialized in 0.44 seconds.
August 23 12:47:03 INFO [46992284444128]: Storage Cache Manager: initialization complete
August 23 12:47:03 INFO [46992284444128]: Spooler Cache Manager: initializing
August 23 12:47:03 INFO [46992284444128]: Spooler Cache Manager: initialization complete
August 23 12:47:03 INFO [46992284444128]: DataStore Manager: initializing
August 23 12:47:03 ERR [46992284444128]: -1: ds_checkIdentity: identity file /Storage/data/.identity does not exist
August 23 12:47:03 ERR [46992284444128]: 25004: _storeInit: the identity of data store is not valid, spoold cannot con tinue.
August 23 12:47:03 ERR [46992284444128]: 26016: Store Manager: Activate failure.Cause
Checked the /Storage/data directory to see if this .identity file existed. The .identity file should contain the storagepool ID and the contentrouter ID
# ls -al
total 40
drwxr-x--- 3 root root 8192 Aug 23 13:03 .
drwxr-xr-x 16 root root 8192 Jan 26 2011 ..
-rwxr-x--- 1 root root 8 Jul 1 2010 .cnf
-rwxr-x--- 1 root root 56 Jul 1 2010 .cnt
-rwxr-x--- 1 root root 56 Jul 1 2010 .path.cnt
drwxrwxrwx 2 root root 96 Aug 23 13:03 deref
As the file does not exist, it can be generated by running the command
/opt/pdcr/bin/spoold --genidentity
We then see that a .identity file exists
# ls -dl .id*
-rw-r--r-- 1 root root 45 Aug 23 14:16 .identity
And the file contains the required information
# cat .identity
[Main]
StoragePoolID = 1
ContentRouterID = 2
Before starting the contentrouter process it should be investigated why this file was missing and why there are not other files in /Storage/data.
Checked the /etc/fstab and /Storage/data is a separate file system to /Storage.
/dev/vx/dsk/PDDG/PDCRVOL /Storage/data vxfs suid,_netdev 0 0
This file system is not mounted so that is why the .identity file does not exist and why there is no other files in /Storage/data.
Trying to manually mount /Storage/data results in an error
# mount /Storage/data
UX:vxfs mount.vxfs: ERROR: V-3-21268: /dev/vx/dsk/PDDG/PDCRVOL is corrupted. needs checking
Checking the /var/log/messages on the content router we can see errors
Aug 22 19:00:55 server1 kernel: vxfs: msgcnt 6 mesg 039: V-2-39: vx_writesuper - /dev/vx/dsk/PDDG/PDCRVOL file system super-block write error
Aug 22 19:00:56 server1 kernel: vxfs: msgcnt 7 mesg 037: V-2-37: vx_metaioerr - vx_logbuf_write - /dev/vx/dsk/PDDG/PDCRVOL file system meta data write error in dev/block 0/752965
Aug 22 19:00:56 server1 kernel: vxfs: msgcnt 8 mesg 031: V-2-31: vx_disable - /dev/vx/dsk/PDDG/PDCRVOL file system disabled
Solution
After making sure the backend storage has no issues an FSCK of the file system should resolve the problem with the file system
/opt/VRTS/bin/fsck -t vxfs /dev/vx/dsk/PDDG/PDCRVOL
Once complete you should be able to mount the file system
#mount /Storage/data
Then start the Puredisk services
#/etc/init.d/puredisk start
NOTE:- Once the system is back up and running, it may be worth doing a consitenct check in case any data has been lost due to the file system issues.
|
|
Article URL http://www.symantec.com/docs/TECH169920
Terms of use for this information are found in Legal Notices
Thank you.