I am currently trying to coordinate the snapshot and SnapMirror replication sequence between Oracle on VMs, application on VM, and Oracle on external physical servers that are all part of an application that needs to be quiesced as close to possible at the same time to ensure a valid recovery point.
I have been able to get everything to work as planned with the exception of triggering the SnapMirror update for the external DBs. I can quiesce Oracle on the VMs, take a SnapShot of the VMs, quiesce Oracle on the external servers, but have not been able to figure out where I can place a config just to simply initiate the SnapMirror update for the Oracle DB that is on the physical servers.
This is the current sequence in the main config file (configs being called from a main config file):
I created a config that will trigger the SnapMirror update for the volumes that are related to the Oracle instance (running on the system identified as 'camel' above). Since I already have a config that will quiesce the DB I placed this in the section on the GUI for 'SnapShot Create Command' - it will always error. I tried placing it in other areas, same error.
[Thu Dec 1 19:28:48 2011] DEBUG: Command ["C:\Program Files\Netapp\NetApp_Snap_Creator_Framework\scServer3.4p2\snapcreator" --profile SCRATCH_ALL --config SCRATCH_scr_SM --action snap --policy hourly] finished with
exit code: 
[Thu Dec 1 19:28:48 2011] ERROR: Command ["C:\Program Files\Netapp\NetApp_Snap_Creator_Framework\scServer3.4p2\snapcreator" --profile SCRATCH_ALL --config SCRATCH_scr_SM --action snap --policy hourly] failed with return code  and message 
[Thu Dec 1 19:28:48 2011] ERROR: [scf-00103] Running snapshot create command ["C:\Program Files\Netapp\NetApp_Snap_Creator_Framework\scServer3.4p2\snapcreator" --profile SCRATCH_ALL --config SCRATCH_scr_camel_SM --action snap --policy hourly] failed with exit code  and message 
Basically I need to ensure I can create the SnapShot and trigger the SM update while all the applications are quiesced. I am assuming I am just specifying the config to trigger the SnapMirror update in the incorrect area.
After completing additional testing with the integration with vSphere; creating the configurations to quiesce, create NetApp SnapShot, and replicate using the ‘VMware vSphere’ plugin works great for all applications that are part of the vSphere infrastructure.
Since I have additional Oracle DB’s that are not part of the vSphere configuration (on external host), is it possible to add the operations to quiesce and un-quiesce this configuration within a single profile/config (I was hoping I could use the configuration created with the ‘VMware vSphere’ plugin adding the profile/config of the external Oracle DB in the sequence somewhere within the main config file)? I have specific startup/shutdown sequences that are application dependent that I would like to coordinate between the vSphere DBs and the external Oracle DBs (that way I can take advantage of the PRE-EXIT commands, etc... in the event of failures)/
I tried adding the profile/config (i.e. APP_QUIESCE_CMD04) of the external Oracle DB to the MAIN configuration that was created as part of the ‘VMware vSphere’ plugin and, of course, as show above, did not work. I have tried adding the external Oracle DB profile/config in other PRE/POST areas, and of course, did not work (as most are intended to perform operations on the Oracle servers themselves, not initiate additional Snap Creator operations on the Snap Creator server).
Example of what I am trying to accomplish in single config file:
Just in case anybody was curious, the easiest way I was able to figure out how to coordinate the shutdown of Oracle databases on VMware and physical machines and replicate a consistent environment to a DR system was to create a windows bat file that would coordinate the operations between the systems. Basically with the following sequence:
1. For VM1 - quiesce Oracle DB
2. For VM1 - use VIBE plugin to quiesce VM
3. For VM2 - quiesce Oracle DB
4. For VM2 - use VIBE plugin to quiesce VM
5. For physical server1 - quiesce Oracle DB
6. For physical server2 - quiesce Oracle DB
7 . On NetApp - create snapshots on DB volumes and ESX Datastore and replicate to DR
8. For physical server2 - unquiesce Oracle DB
9. For physical server1 - unquiesce Oracle DB
10. Unquiesce VM2 (VIBE)
11. Unquiesce VM1 (VIBE)
12. Unquiesce VM2 DB
13. Unquiesce VM1 DB
All applications that needed to be replicated at the same time (in conjunction with the physical systems) within VMware are on a dedicated datastore.
Here is an example of what the bat file (called via Windows Task Scheduler):
:: ## Variables
set /a sdm=1%startDate:~4,2% - 100
set /a sdd=1%startDate:~7,2% - 100
set /a sth=%startTime:~0,2%
set /a stm=1%startTime:~3,2% - 100
set /a sts=1%startTime:~6,2% - 100