[ClusterLabs] cofigured trace for Virtual Domains - automatic restart ?

[Lentes, Bernd]

----- On Sep 18, 2021, at 1:19 AM, kgaillot kgaillot at redhat.com wrote:

> 
> If the agent meta-data advertises support for the 1.1 standard and
> indicates that the trace_ra parameter is reloadable, then Pacemaker
> will automatically do a reload instead of restart for the resource if
> the parameter changes.

>From where do i know that my RA supports 1.1 ?
I'm running on SLES 12 SP5 with resource-agents-4.3.018.a7fb5035-3.51.1.x86_64.
My crm does not support "crm resource reload":

ha-idg-1:~ # crm resource help
Resource management

At this level resources may be managed.

All (or almost all) commands are implemented with the CRM tools
such as crm_resource(8).

Commands:
       ...
        refresh          Recheck current resource status and drop failure history
        restart          Restart resources
      ...

> 
> trace_ra is unusual in that resource agents don't define the parameter
> themselves, the ocf-shellfuncs shell include looks for it instead. It
> would be nice to come up with a general solution that all agents can
> use rather than modify each agent's meta-data individually, but either
> approach would work.

Bernd
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 2217 bytes
Desc: S/MIME Cryptographic Signature
URL: <https://lists.clusterlabs.org/pipermail/users/attachments/20210918/76c7ff0e/attachment-0001.p7s>