SMO/Snapdrive Data Protection Issue

dreimer · ‎2011-05-16

SnapManager for Oracle 3.1P1

Snapdrive 4.2P1

DFM 4.0.2 (on Redhat Enterprise Linux 5) w/Protection Manager and Provisioning Manager

OS: Oracle Unbreakable Linux 5 32bit

DB: Oracle 11g R2

Netapp: 8.0.1RC3X16 (Simulator)

I have successfully configured this environment to the point where I can backup, clone and restore databases. When I attempt to apply a Protection poliy to the database profile in SMO, the dataset is appears to be created within Protection Manager but then disappears (and the Protection Policy checkbox in the Policies tab of the SMO profile is cleared). The logs seem to indicate a snapdrive error as follows:

[ INFO] DS-10012: Creating dataset smo_drs-ora-d01_samplep
[ INFO] DS-10019: Setting protection policy of dataset smo_drs-ora-d01_samplep to ep_Mirror.
[ INFO] SMO-05057: Updating members in dataset "smo_drs-ora-d01_samplep".
[ INFO] SD-00016: Discovering storage resources for /samplep.
[ INFO] SD-00017: Finished storage discovery for /samplep.
[ INFO] SD-00016: Discovering storage resources for /archive1.
[ INFO] SD-00017: Finished storage discovery for /archive1.
[ INFO] SD-00016: Discovering storage resources for /control.
[ INFO] SD-00017: Finished storage discovery for /control.
[ INFO] DS-10014: Adding members to dataset smo_drs-ora-d01_samplep: [/archive1, /control, /samplep]
[ERROR] FLOW-11019: Failure in DatasetMemberSync: SD-10028: SnapDrive Error (id:2429 code:102) Could not get the list of all devices: .
[ERROR] FLOW-11008: Operation failed: SD-10028: SnapDrive Error (id:2429 code:102) Could not get the list of all devices: .
[ERROR] SMO-13032: Cannot perform operation: Profile Update. Root cause: SMO-05058: Error while updating members in dataset "smo_drs-ora-d01_samplep": FLOW-11019: Failure in DatasetMemberSync: SD-10028: SnapDrive Error (id:2429 code:102) Could not get the list of all devices: .
[ INFO] DS-10013: Deleting dataset smo_drs-ora-d01_samplep
[ INFO] SMO-07433: Returning the database to its initial state: samplep(OPEN).
[ INFO] SMO-13039: Successfully aborted operation: Profile Update
[ERROR] SMO-13048: Profile Update Operation Status: FAILED
[ INFO] SMO-13049: Elapsed Time: 0:01:33.284
REMOTE-00004: RemoteObjectFactory shutdown gracefully.

The snapdrive command: snapdrive storage show -all

returns the following error: 0001-625 Command error: Could not get the list of all devices

Other snapdrive commands seem to work fine (i.e. snapdrive snap list -v -filer sim-ntap-1 returns a list of all snapshots on the sim-ntap-1 storage controller)

jakub_wartak · ‎2011-05-18

Hi,

1. set some rather large number of of allowed snapdrive trace log files via snapdrive.conf.

2. Restart the snapdrive daemon.

3. Execute again "snapdrive storage show -all" and grep for "ASSISTANT EXECUTION" inside the /var/log/sd-trace* log files. If you find correct log file then you have to copy it somewhere (those are auto-rotated), read it deeply, probably some OS level command is failing...

-Jakub.

dreimer · ‎2011-05-18

With the help of Netapp Support I was able to get this resolved.

The snapdrive storage show -all command seemed to be failing while trying to get a list of devices. Apparently because the iSCSI daemon was running on the Oracle host (installed and running by default), but was not configured, it was failing while trying to enumerate the devices. After disabling the iSCSI daemon on the Oracle host, I was able to successfully run the snapdrive storage show -all command, and able to apply an SMO Protection policy.

Thanks.