T78319 ask to prevent scap error reporting on hosts that are under maintenance. To do so we would need an export of such hosts from Icinga which will be consumed by scap.
We are going to use etcd for the varnish caches, so seems it can feed scap as well.
| Status | Subtype | Assigned | Task | ||
|---|---|---|---|---|---|
| Open | None | T78319 [scap] Suppress/de-emphasize errors from hosts marked has being under maintenance | |||
| Open | None | T100777 expose hosts in maintenance state so we can prevent scap from running on them |
That doesn't sound like a very good idea. It should be the other way around: we set hosts under maintenance somewhere else (e.g. etcd ;)) and from there it propagates wherever needed (pybal, icinga, scap).
+1 for just getting etcd tracking the state of all the MW servers and then switching scap to get the list of hosts to update from there rather than the static files left over from the dsh days.