Welcome to the new Gigaspaces XAP forum. To recover your account, please follow these instructions.

Ask Your Question
0

What is the correct way to shutdown a GSA, etc. started with gs-agent.sh?

What is the correct way to shutdown a GSA, etc. started with gs-agent.sh?

Thanks, Rowland

{quote}This thread was imported from the previous forum. For your reference, the original is [available here|http://forum.openspaces.org/thread.jspa?threadID=3336]{quote}

asked 2009-12-22 03:35:41 -0500

river2sea gravatar image

updated 2013-08-08 09:52:00 -0500

jaissefsfex gravatar image
edit retag flag offensive close merge delete

4 Answers

Sort by ยป oldest newest most voted
0

You will need to kill the GSA process (after you undeployed the application).
This in turn will terminate all the GSC/GSM/LUS managed by the GSA proecss on the machine.
We will be adding CLI/API to shutdown the GSA with future versions.
Shay

answered 2009-12-22 09:27:54 -0500

shay hassidim gravatar image
edit flag offensive delete link more

Comments

OK, thanks for the info Shay.

Rowland

river2sea gravatar imageriver2sea ( 2009-12-22 11:41:01 -0500 )edit
0

The admin API can help you to determine the status of a PU. Usually it takes few seconds per PU instance to be undeployed.

The org.openspaces.core.cluster.ProcessingUnitUndeployingListener can help you here. See:

org.openspaces.admin.pu.ProcessingUnit. addLifecycleListener(ProcessingUnitInstanceLifecycleEventListener eventListener) and getProcessingUnitInstanceProvisionStatusChanged()

answered 2017-03-09 13:49:18 -0500

shay hassidim gravatar image
edit flag offensive delete link more
0

Hi Shay,

What about the same operation through the Admin API? I'm invoking the GridServiceAgent shutdown method, but not all the processes are killed. Is it mandatory to first undeploy the application, also via Admin API, and only then invoke the shutdown method? Or is the grid service agent implementation supposed to handle this 'clean up' internally (I'd expect it to happen, in a client standpoint).

Thanks!

answered 2017-03-09 12:55:56 -0500

pedro_brigatto gravatar image
edit flag offensive delete link more
0

You should undeploy and later shutdown the agent. Once you shutdown the agent - all the processes it manages should be terminated as well. If this is not happening - report this issue to support.

answered 2017-03-09 13:02:09 -0500

shay hassidim gravatar image
edit flag offensive delete link more

Comments

Thanks Shay. I'm gonna follow your suggestions. Just one more question: after the undeployment, by calling the shutdown, for how long should I wait to determine something is wrong? I know the clean up is supposed to take some time, but is there a way to determine a reference so that we know something went wrong? Or, once all PUs were already undeployed, the clean up should be very quick ...? What parameter can I use to have any reasonable conclusion about the process?

pedro_brigatto gravatar imagepedro_brigatto ( 2017-03-09 13:10:53 -0500 )edit

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Question Tools

1 follower

Stats

Asked: 2009-12-22 03:35:41 -0500

Seen: 40 times

Last updated: Mar 09