add example diagrams

[remylouisew]

Adding diagrams and tables depicting how metrics are calculated when a job completes vs when a job fails.

[nstogner]

Some of these metric names dont match what is being produced. You probably need to query the metrics in the project to see exactly what is there (search for megamon_alpha_ and remove the "alpha" part from the final doc. PS: I think you checked in a duplicate diagram.

[remylouisew]

Ok the metric names should be correct now (hopefully). I see two different diagrams, let me know if you still don't

[nstogner]

Hey @remylouisew! I am still seeing some differences. Here are the metrics that are produced today:

(where * might be jobset or nodepool or jobset_nodes)

megamon_*_up
megamon_*_up_time_seconds
megamon_*_down_time_seconds
megamon_*_interruption_count
megamon_*_recovery_count
megamon_*_up_time_between_interruption_seconds
megamon_*_up_time_between_interruption_mean_seconds
megamon_*_up_time_between_interruption_latest_seconds
megamon_*_down_time_initial_seconds
megamon_*_down_time_between_recovery_seconds
megamon_*_down_time_between_recovery_mean_seconds
megamon_*_down_time_between_recovery_latest_seconds

PS: Technically the megamon_ part is actually configurable.

[remylouisew]

I'll wait to submit the next fix until the names are more stable, since they keep changing. Thanks for reviewing!

…

On Mon, Dec 30, 2024 at 5:50 PM Nick Stogner ***@***.***> wrote: Hey @remylouisew <https://github.com/remylouisew>! I am still seeing some differences. Here are the metrics that are produced today: (where * might be jobset or nodepool or jobset_nodes) megamon_*_up megamon_*_up_time_seconds megamon_*_down_time_seconds megamon_*_interruption_count megamon_*_recovery_count megamon_*_up_time_between_interruption_seconds megamon_*_up_time_between_interruption_mean_seconds megamon_*_up_time_between_interruption_latest_seconds megamon_*_down_time_initial_seconds megamon_*_down_time_between_recovery_seconds megamon_*_down_time_between_recovery_mean_seconds megamon_*_down_time_between_recovery_latest_seconds PS: Technically the megamon_ part is actually configurable. — Reply to this email directly, view it on GitHub <#2 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACEBWG7ZYUIXAJIQ5HX5YD32IHZ4TAVCNFSM6AAAAABRXSHO5SVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKNRWGA3DIOJZG4> . You are receiving this because you were mentioned.Message ID: ***@***.***>

-- Remy Welch | +1 240 515 4204 Customer Engineer | Google Cloud <http://cloud.google.com/> | Google Inc. remywelch.com