Scripted shutdown of Cluster and restart problem

[phil76]

hello,


we have an active / passive cluster in use and want in case of a power failure or maintenance shutdown the cluster via script controls and later restart again.

but now we have the following problems:

there is no API command to stop the cluster and if we shutdown the second and then shutdown the primary node and powerup the primary first, the cluster is stopped and all replication targets are set from source to destination.

Then you must manually set all replication targets from target to source and start all replication task manually and then start the cluster again.

in case of maintenance could be the cluster manually stop and then start again later easily, but during power failure unfortunately this does not work because there is no script command to stop cluster.

can anyone help me?

[To-M]

We are looking into this to have the API functions for the Failover, it might be in the next release for the DSS V7 but I am not 100% sure due to testing and the many other feature requests that are currently in the que but we will have it for sure. The one way to mitigate the issue is to make sure to drop down the Secondary 1st then after the Primary then when bring the systems up bring up the Secondary 1st then the Primary.

[phil76]

We are looking into this to have the API functions for the Failover, it might be in the next release for the DSS V7 but I am not 100% sure due to testing and the many other feature requests that are currently in the que but we will have it for sure. The one way to mitigate the issue is to make sure to drop down the Secondary 1st then after the Primary then when bring the systems up bring up the Secondary 1st then the Primary.

had previously tested before but i tested it again.
Unfortunately this does not work so well.
If the primary node starts after completely started secondary node, all replications are stopped then all replications set from source to target and the cluster consequently stopped.

[To-M]

I forget to add a step. So when you down the Destination server then the Source server then bring up the Destination server and the Source then go to the Source and start the Volume Replication tasks then once they are consistent then start the Failover.

[phil76]

The one way to mitigate the issue is to make sure to drop down the Secondary 1st then after the Primary then when bring the systems up bring up the Secondary 1st then the Primary.

Hello,

if I start the secondary first and then if that is completely booted up the primary, the cluster is stopped and the cluster replication targets in primary are set to target.
I have now tried already all combinations but the cluster is always stopped at startup.


i think if you shutdown one of the two nodes with active cluster it stops the replication and sets itself to replication target


P.S. I have 5 days ago already answered here but somehow my answer has disappeared

[To-M]

Keep in mind that if both are down then restarted then Source of Primary on the Volume Replication for the volumes will be in Destination mode and you need to switch to Source mode then you can start the Failover service.

[phil76]

in case of a clean shutdown due to power failure it must be possible the system automatically restart without contact an admin.

The only thing missing is the API command to stop and restart the cluster.
Thats what I expect from a so expensive software.
A quick solution before this was implementet can be to give me the HTML link to automatically login and the link to Start/Stop Cluster button.
Is this possible ?

[To-M]

Our GUI language is different then a HTML link, many use an application that does a Mouse Controller Recording and Automates Mouse Actions to perform such actions.

[To-M]

Update to the Stop and Start cluster - We have made a small update for the latest V7 using API. If interested send a support ticket asking for this.

[phil76]

Thanks for your work.

got it from your ftp and will test this tomorrow.