davidpablocohn
commented
3 months ago Hrrmmmmm... As you've observed, that shouldn't be happening - thanks for catching/documenting that!
I'm neck deep in farm stuff tomorrow, but will put on my code snorkel on Friday and try diving into this.
On Tue, May 21, 2024 at 5:00 PM Michael D Labriola @.***> wrote:
I saw this happen the other day, but couldn’t figure out what I’d done to trigger it… If you modify just one logger in the yaml config, then push the yellow “Definition is stale. Reload” button, all the loggers are still running and the one you modified hasn’t seen it’s changes yet. If at that point you stop and restart JUST the logger you actually modified, you end up with ONLY that logger running. For some reason, when you stop the one logger, OpenRVDAS stops ALL the loggers but displays them all as still running in the web UI (happy little green buttons)… so that’s bad.
The quick fix is to push the “cruise mode” button (which has turned into an empty label) and select “all-on” to have it start them all back up.
For now, the safe way to refresh just a single logger seems to be:
- modify the config
- Reload the stale config
- open the cruise mode dialog
- select “all-logging” and “Change Mode”.
And that seems like it can tell that only the one logger really needs to restart, at least I think that's what I'm seeing.
This is happening to me while running v1.9.0 with about a dozen loggers reading UDP streams and writing them all to disk after applying timestamps. On Ubuntu 22.04.
— Reply to this email directly, view it on GitHub https://github.com/OceanDataTools/openrvdas/issues/392, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFO7V3XYNDXIZKNJX3T27O3ZDPNYJAVCNFSM6AAAAABICRXHS2VHI2DSMVQWIX3LMV43ASLTON2WKOZSGMYDSMZQG42TINY . You are receiving this because you are subscribed to this thread.Message ID: @.***>
veggiemike
I saw this happen the other day, but couldn’t figure out what I’d done to trigger it… If you modify just one logger in the yaml config, then push the yellow “Definition is stale. Reload” button, all the loggers are still running and the one you modified hasn’t seen it’s changes yet. If at that point you stop and restart JUST the logger you actually modified, you end up with ONLY that logger running. For some reason, when you stop the one logger, OpenRVDAS stops ALL the loggers but displays them all as still running in the web UI (happy little green buttons)… so that’s bad.
The quick fix is to push the “cruise mode” button (which has turned into an empty label) and select “all-on” to have it start them all back up.
For now, the safe way to refresh just a single logger seems to be: 1) modify the config 2) Reload the stale config 3) open the cruise mode dialog 4) select “all-logging” and “Change Mode”.
And that seems like it can tell that only the one logger really needs to restart, at least I think that's what I'm seeing.
This is happening to me while running v1.9.0 with about a dozen loggers reading UDP streams and writing them all to disk after applying timestamps. On Ubuntu 22.04.