[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: Command to start/top a cluster gracefully



Thanks Nick for this!


That's what I suspected. I'll make some tests here to see if a simple big shutdown of the hosts do not break anything. ;)


Kind regards,

Marc


On 12/10/18 00:18, Nick Pilch wrote:

My team has found that, after we configure our nodes with ansible, things connect correctly themselves. So to shut down a cluster, we just stop the nodes, and to start a cluster, we just start the nodes. Occasionally we have to kick some services because they don't start in the right order. This is with 1.3. We recently made the big upgrade to 3.9, but have not rolled that out in all our environments yet, so we don't have much experience with it yet.


Now the app is a different story. Our app pods unfortunately have some startup order dependencies, so we use a script with oc commands to start up our app pods which makes sure the order is correct. I would think ideally, your app pods could just start up in any order and be able to handle it.


If your app needs to do certain things when you shut it down, then you'll have to have some custom automation for that I would imagine.



Nick Pilch
Cloud Operations
O: 650.567.4560
M: 510.381.6777
E: nick pilch bluescape com
999 Skyway Rd, Suite 145, San Carlos, CA 94070
Notice of Confidentiality: This message and any attachments are confidential. If you are not the intended recipient, please do not read or distribute. Alert the sender by reply email and delete this message immediately.


From: users-bounces lists openshift redhat com <users-bounces lists openshift redhat com> on behalf of Aleksandar Lazic <aleks me2digital eu>
Sent: Wednesday, October 10, 2018 1:36:16 PM
To: Marc Ledent; users lists openshift redhat com
Subject: Re: Command to start/top a cluster gracefully
 
Am 10.10.2018 um 11:22 schrieb Marc Ledent:
> Hi all,
>
> Is there a command to stop/start an openshift cluster gracefully. "oc cluster"
> commands are acting only for a local all-in-one cluster...

Do you mean something like this?

* Scale all dc/rc/ds to 0

* stop all node processes

* stop all master process

* stop all etc processes

* stop all docker processes

* shutdown all machines

I don't know a easier way maybe there is a playbook in the ansible repo

Regards

Aleks

> Thanks in advance,
> Marc
>
>
>
> _______________________________________________
> users mailing list
> users lists openshift redhat com
> http://lists.openshift.redhat.com/openshiftmm/listinfo/users


_______________________________________________
users mailing list
users lists openshift redhat com
http://lists.openshift.redhat.com/openshiftmm/listinfo/users

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]