I disabled a slave last week due to hardware issues. Over the weekend, the slave process either died, or the machine was rebooted, and now the slave is no longer marked as disabled in the repository.
Because of the hardware issues, I can’t SSH to the machine to check the logs or the status of the processes, but I thought I’d check and see if this was already a known issue.
This isn’t a known issue (it’s the first we’ve heard of it).
The disabled flag is stored in the Slave Settings object, and only the Monitor and Command applications can actually modify it (when saving out the settings for the slave). Did you notice if any other settings for the slave changed (like pools, groups, description, etc)? If so, maybe the Slave Settings object was corrupted and/or reverted back to the default settings at some point…
No, as far as I can tell, everything else is the same (including the comment I left on the slave). Its status hasn’t been updated in a week, so I don’t think it’s come back online or anything; the disabled status just seems to have been lost.
Strange. Anything in the slave history to show if it was re-enabled?
Ah right, there’s a slave history… Yeah, it looks like someone inadvertently re-enabled it. Mystery solved!
Sorry for the noise…