[k8s] how to get the job sutatus change event

[shengkaixuan]

skypilot is a great project,run it in k8s, how can i be alerted when the job status changed?, like the status changed to FAILED.
`
Fetching and parsing job queue...

Job queue of cluster llm
ID NAME SUBMITTED STARTED DURATION RESOURCES STATUS LOG
3 - 1 hr ago 1 hr ago 1h 25m 37s 1x[QUADRO-RTX8000:1] RUNNING ~/sky_logs/sky-2024-07-23-16-08-46-454657
2 - 1 hr ago 1 hr ago 10m 21s 1x[QUADRO-RTX8000:1] SUCCEEDED ~/sky_logs/sky-2024-07-23-15-44-46-107347
1 - 1 hr ago 1 hr ago 4m 48s 1x[QUADRO-RTX8000:1] FAILED ~/sky_logs/sky-2024-07-23-15-34-45-318133
`

[concretevitamin]

@shengkaixuan Just to understand the use case better, are you trying to programmatically alert the user via email, slack, a custom dashboard, or some other system?

[shengkaixuan]

just want an api to send the event to a custom message system.
By the way, will skypilot offer a web ui in the future to show all the tasks?

[concretevitamin]

Check out #2106 from @MaoZiming

[shengkaixuan]

thank you, it would be better to put it in the doc.
where can i find all the predefined environment variables?
i do not think the doc cover all the predefined environment variables. https://skypilot.readthedocs.io/en/latest/running-jobs/environment-variables.html#skypilot-environment-variables

[cblmemo]

thank you, it would be better to put it in the doc. where can i find all the predefined environment variables? i do not think the doc cover all the predefined environment variables. https://skypilot.readthedocs.io/en/latest/running-jobs/environment-variables.html#skypilot-environment-variables

Hi @shengkaixuan ! Could you elaborate on what environment variables you find missing?