The mysterious running out of memory and then running out of swap problem came back yesterday, resulting in downtime of around 6 hours (or if you're @hoom, roughly 24 hours). I was notified swiftly by BRiT not long after it happened, although because of timezones and a little bit of work interruption, I didn't fix it until 10am or so.
That's the first time it's happened since January, and the fix is still to forcefully reboot the VM because the console is too overloaded with swap errors from the kernel to respond.
Anyway, sorry about that and while I can't promise it won't happen again, at least the notification system to prod me to fix it is working properly now.
That's the first time it's happened since January, and the fix is still to forcefully reboot the VM because the console is too overloaded with swap errors from the kernel to respond.
Anyway, sorry about that and while I can't promise it won't happen again, at least the notification system to prod me to fix it is working properly now.
