You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
TTR is a mechanic to release crashed / stalled jobs back into the queue.
Sometimes jobs are slow but not stalled.
Consider the DownloadJob from the docs; I could easily make this job exceed its TTR by throttling the connection at the server.
However since the process is alive and actually doing what it is supposed to do (assuming it is still actually receiving more than 0 b/s) we might want to tell the queue that we're still busy and it shouldn't release our job to someone else just yet.
Beanstalk protocol uses touch for this, and this shouldn't be hard to implement for other drivers.
The database driver currently doesn't do TTR at all (#249). When it does, all we need to do to extend the time is update the reserved_at field.
For other drivers I'm not sure how to implement this exactly, but I don't think it'll be hard.
The text was updated successfully, but these errors were encountered:
TTR is a mechanic to release crashed / stalled jobs back into the queue.
Sometimes jobs are slow but not stalled.
Consider the
DownloadJob
from the docs; I could easily make this job exceed its TTR by throttling the connection at the server.However since the process is alive and actually doing what it is supposed to do (assuming it is still actually receiving more than 0 b/s) we might want to tell the queue that we're still busy and it shouldn't release our job to someone else just yet.
Beanstalk protocol uses
touch
for this, and this shouldn't be hard to implement for other drivers.The database driver currently doesn't do TTR at all (#249). When it does, all we need to do to extend the time is update the
reserved_at
field.For other drivers I'm not sure how to implement this exactly, but I don't think it'll be hard.
The text was updated successfully, but these errors were encountered: