Parca Operator
Parca provides continuous profiling for analysis of CPU and memory usage, down to the line number and throughout time. Saving infrastructure cost, improving performance, and increasing reliability.
This operator builds a simple deployment of the Parca server and provides a relation interface such that it can be integrated with other Juju charms in a model. A good example of this is the juju-introspect operator, which when related to this charm will enable continuous profiling of the Juju controller its attached to.
Usage
You can deploy the operator as such:
# Deploy the charm
$ juju deploy parca --channel edgeOnce the deployment is complete, grab the address of the Parca application:
$ juju show-unit parca/0 --format=json | jq -r '.["parca/0"]["public-address"]'Now visit: http://<parca-address>:7070/ to see the Parca dashboard.
Profiling a Juju Controller
You can deploy this charm alongside juju-introspect to profile a running controller like so:
# Bootstrap a new Juju controller on LXD
$ juju bootstrap localhost lxd
# Switch to the controller model
$ juju switch controller
# Deploy the juju-introspect charm to the controller machine
$ juju deploy --to=0 juju-introspect --channel edge
# Deploy the Parca charm
$ juju deploy parca --channel edge
# Relate the two charms to enable scraping
$ juju relate parca juju-introspectIf you visit the Parca dashboard, you should now see profiles for the related Juju controller.
Configuration
By default, Parca will store profiles in memory. This is the current default, as the persistence settings are very new and prone to breaking! The default limit for in-memory storage is 4096MB. When Parca reaches that limit, profiles are purged.
The in-memory storage limit is configurable like so:
# Increase limit to 8192MB
$ juju config parca memory-storage-limit=8192If you wish to enable the experimental storage persistence, you can do as such:
$ juju config parca enable-persistence=true