-
I can't figure out how to properly update plugins, preferably with zero downtime.
But this fill junky and this way I lose all task execution logs. How to properly update plugins without losing your schedule and logs? |
Beta Was this translation helpful? Give feedback.
Replies: 2 comments 2 replies
-
Moving to discussion. This is not an issue. |
Beta Was this translation helpful? Give feedback.
-
Well, at my day job we just stop Cronicle entirely, update what we need, then start it again. But I guess since you're using Docker that doesn't work, because if you stop Cronicle, the container exits? I'm guessing that's what the problem is. So, you'll have to script a sequence of actions using the REST API:
To enable/disable the scheduler, use the update_master_state API. To wait for all jobs, poll the get_active_jobs API, and wait for it to return no results. To abort all jobs, first use the get_active_jobs API to get the IDs of all active jobs, then call abort_job on each one to abort it. Good luck! |
Beta Was this translation helpful? Give feedback.
Well, at my day job we just stop Cronicle entirely, update what we need, then start it again. But I guess since you're using Docker that doesn't work, because if you stop Cronicle, the container exits? I'm guessing that's what the problem is.
So, you'll have to script a sequence of actions using the REST API:
To enable/disable the scheduler, use the update_master_state API.
To wait for all jobs, poll the get_active_jobs API, and wait for it to return no results.
To abort all jobs, first use the get_active_jobs API to get the IDs of all active jobs, then call abort_job on each …