[Logging] Please advise people to use reset() for long running processes

[Seldaek]

See https://twitter.com/seldaek/status/1370115718499422208 for details.

The latest recipe has a buffer limit now so it's less of an issue I guess, but still reset is cleaner.

[kbond]

Is this an issue for messenger workers? Should reset be called between jobs (WorkerMessageHandledEvent event)?

[Seldaek]

I'd say yes that would probably be a good idea. The one trade-off is that you lose context in case you have an issue that is triggered by something that happened in a previous job, but those are rather rare occurrences.

[Seldaek]

In my worker implementations I typically do a reset() before starting processing the first job (so that any logging happening while spinning up a worker is cleared) and then after every job processed.

[kbond]

symfony/symfony#40761 solves this I believe.

[Seldaek]

Not really. I mean it solves it for messenger workers, but pretty much every project I work on has custom built long term processes, so I suspect this is a fairly common thing people do, and the logging docs should still mention reset() IMO.

[carsonbot]

Thank you for this issue.
There has not been a lot of activity here for a while. Has this been resolved?

[carsonbot]

Friendly ping? Should this still be open? I will close if I don't hear anything.

[Seldaek]

Can be closed as #16609 is enough to keep track of this.