When running on multiple servers, worker-stop removes all worker entries from Redis, not just local ones

[brianV]

In our usage, we have multiple machines running resque workers.

When the 'resque:worker-stop --all' command is issued, the local workers are stopped and all the entries for workers on all machines are removed.

Workers which do not exist on the local machine (and therefore cannot be terminated) should not be removed from the active workers list.

[PhilETaylor]

yeah sounds about right :-(

Im just about to go live in a multi server environment - I would have hit this soon, so thanks for reporting, I'll investigate a distributed way of terminating all workers on all machines - maybe by setting a kill switch in redis that workers look for every poll...

[brianV]

Oh, that would be terrific if you could add that! Maybe add a --local switch at the same time to compliment --all!

…

On Tue, Oct 31, 2017 at 2:35 PM, Phil Taylor ***@***.***> wrote: yeah sounds about right :-( Im just about to go live in a multi server environment - I would have hit this soon, so thanks for reporting, I'll investigate a distributed way of terminating all workers on all machines - maybe by setting a kill switch in redis that workers look for every poll... — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#27 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAQoKBpxxVvTF32f-FObvxPD666P3iuaks5sx2iCgaJpZM4QM_su> .

[PhilETaylor]

This is still an issue (Nov 2019)