Autosys Vmotion issue
Article ID: 263014
Updated On:
Products
Issue/Introduction
Hi Team,
We are facing issues in Autosys servers after the Windows team performed the Vmotion activity.
We did follow the below-mentioned KB articles, but still we ended up doing the sequence reboot of all the servers.
We are seeing agents and jobs going to starting state automatically.
Kindly look into the issue and provide us the further solution to resolve the issue.
Regards,
Shravani K
Environment
Autosys R12.x
Cause
As of R12.1SP1 we do not support VMoption as a High availability option
Resolution
Currently, Autosys does not support Vmotion as a high-availability option
The document link provided states the following:
Do not use vMotion to move the virtual machines to other ESXi hosts.
At the bottom it also states the following:
Before allowing a move, shut down the AutoSys Workload Automation processes as you typically would when performing any maintenance.
This way processing ceases until the processes are brought back up, and then new connections are re-established.
Additional Information
Notes:
1) currently there is no support for VMoption operations in a High availability situation where the application will not experience an outage.
2) VMotion can be used for Disaster recovery with the understanding the application will need to be properly shut down and started before normal operations will resume.
This is documented in the below link:
https://techdocs.broadcom.com/us/en/ca-enterprise-software/intelligent-automation/autosys-workload-automation/12-0/Before-You-Begin/installation-considerations/ca-workload-automation-ae-and-vmware-vmotion.html
If your client would like to set up a Disaster Recovery /High availability solution I will be glad to set up a call/Webex to go over the support options.
Currently, Supported DR/HR methods.
1) For the database side we support Oracle cluster /RAC or for MS SQL server we support clustering- You would point the Scheduler and WCC to the VIP for the HA instance of the database
2) For Autosys Scheduler you would setup a Primary and Shadow scheduler
3) for EEM you would need to setup EEM in a multi-write setup
4) We do not have HA for WCC, however, you can set up multiple WCCs behind a Load balancer, and should one go down users can continue to use the other WCC with the understanding that they would need to logout and log back in to establish a new session.
Feedback
