Dan Everton

Testing this out again it looks like it is just the "application" search that is the problem for us. The other search types (projects, serverGroups, instances, loadbalancers, and securityGroups) all...

Some more digging it looks like this call https://github.com/spinnaker/clouddriver/blob/master/clouddriver-ecs/src/main/java/com/netflix/spinnaker/clouddriver/ecs/provider/view/EcsServerClusterProvider.java#L373 which calls to https://github.com/spinnaker/clouddriver/blob/master/clouddriver-ecs/src/main/java/com/netflix/spinnaker/clouddriver/ecs/cache/client/EcsCloudWatchAlarmCacheClient.java#L76 is really expensive for us as it iterates every alarm to figure out if it has an...

Digging in to this again I think the issue with search might just be as simple as implementing this https://github.com/spinnaker/clouddriver/blob/master/clouddriver-ecs/src/main/java/com/netflix/spinnaker/clouddriver/ecs/provider/view/EcsServerClusterProvider.java#L576 in the provider. The root cause of the issue seems...

It could be related. The lookup logic in the `clouddriver-ecs` module does seem to be written in a "get all the data and then filter" style which gets expensive as...

From what I can tell (though I haven't tested this) the issue is that `getClusterSummaries()` is returning the full data set when it should be just returning summary data. That...

I've run in to this as well with though it's moved on from sddm crashing to pretty much all of the KDE apps segfaulting inside the nvidia driver. Doing a...

Something horrible like this works. There is runuser/daemon in the /etc/init.d/functions file but they don't really help with forking. ``` sh touch $pidfile chown ${LS_USER} $pidfile nice -n ${LS_NICE} su...