AWS Thinkbox Discussion Forums

mark slave bad after N errors

Im not sure if this works, i turned this feature on, and set it to 5. But the slave just keeps picking up & erroring on the same job, ~15 times now.

cheers,
laszlo

Hey Laszlo,

Thanks! We can reproduce this here. We’re looking into it.

Cheers,

  • Ryan

This will be fixed in beta 11.

I think this bug resurfaced?

The repo is set to mark the slave bad after 5 fails
error_no_blacklist.jpg

Where there any other jobs in the queue at the time that this slave could be working on? If a slave is marked as bad for a job, it can retry it later if there are no other “good” jobs in the queue for it to render. The frequency at which this occurs can be configured in the Repository Options under Job Settings -> Failure Detection -> Slave Faiure Detection (yes, we see the typo in the settings window, which will be fixed in the next beta). If you set the “Frequency at which a slave will attempt…” setting to 0%, that will prevent slaves from every retrying jobs that they have been marked bad for.

Cheers,

  • Ryan

Hm, maybe there werent any,… i ll keep an eye on this, and set that value to 0, thanks!

Privacy | Site terms | Cookie preferences