Skip to content

Conversation

@wenningerk
Copy link

@wenningerk wenningerk commented Dec 10, 2019

Counterpart on pacemaker-side is ClusterLabs/pacemaker#1957

Send a ping to pacemakerd to make it start all the subdaemons.
Periodically send a ping and memorize which timestamp pacemakerd
sent marking that subdaemons were alright.
If this timestamp becomes older than watchdog_timeout stop
triggering the inquisitor.
When pacemakerd-pinging returns shutdown-completed assume all
resources are down and exit pacemaker-watcher gracefully.

admin_uuid);

crm_ipc_send(pacemakerd_channel, cmd, 0, 0, NULL);
free_xml(cmd);

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is the meat of the new API right here :) Then the scaffolding is 10 times as much. I would think we just need a pacemakerd client object with connect, disconnect, and ping methods, and new/free functions for that. Not sure what the naming should look like.

State query ping of pacemakerd prevents pacemakerd from
starting any sub-daemons (and thus services) if sbd can't
reach it via ipc. As a health-check get timestamp from
pacemakerd. On shudown fetch info about graceful
shutdown from pacemakerd.
@wenningerk
Copy link
Author

Now using pacemakerd-api.

Use new pacemakerd-api provided by pacemaker.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants