Whats the best way to monitor deployment activity?

[Lowell]

Anyone know the best way to monitor deployment activity of a splunk server? I've found DeploymentMetrics coming from the deployment serer, and I see DeploymentClient and DeployedApplication messages being forwarded from my deployment clients.

There seem to be cases where I see deployments occuring on the clients (reported by DeployedApplication), but I don't see them in DeploymentMetrics. Has anyone else observed anything like that?

Anyone have any helpful deployment tracking searches they would like to share?

---

Here is the basis of my existing searches:

Search 1:

index="_internal" sourcetype="splunkd" source=*metrics.log Component="DeploymentMetrics"

Search 2:

index=_internal sourcetype=splunkd Component="DeployedApplication"

Update: Note. If your using Splunk 4.1, then you have to replace Component with component. (These kind of upgrade bugs really can take a lot of time to straighten out.)

[oreoshake]

Similar searches

If you want to see apps that have been recently installed/updated

source="*splunkd.log" index=_internal "DeployedApplication - Checksum mismatch" OR "Installing" | rex "app: (?<app_name>\\w*)"

If you want to see which clients have "phoned home" for updates:

 source="/opt/splunk/var/log/splunk/splunkd_access.log" index=_internal /services/broker/phonehome | rex "/services/broker/phonehome/connection_[\\d|\\.]*_\\d*_(?<src>.*)_(?<client>.*)_" | dedup client

If you want to see your forwarders restarting (not directly related to the deployment, but still helpful)

source="*splunkd.log" index="_internal" ShutdownHandler punct="--_::.____-____._"

And this was helpful when my server classes overlapped and hosts were sending data to two indexers

source="*splunkd.log" index="_internal" "Connected to" | rex "Connected to (?<indexer>.*)" | stats count by host,indexer | stats count by host | search count>1