FQ Alert: ASTRO

[Deb]

Andrew is on his way to the NOC to bring ASTRO back online.

Everything should be working properly within an hour....

Deb

[Terra]

ASTRO had a problem with it's virtual swapping memory/drive which caused a kernel OOPS to register...[nbsp][nbsp]This is usually attributed to a bad block being remapped on the SCSI drives out from underneath of it...[nbsp][nbsp]Upon swapping the task back in, the block containg a portion of executable code was remapped and zero'd out...[nbsp][nbsp]On non-swap partitions this is handled quite gracefully, but due to the aggresive nature of Linux on swap caches this is not always foolproof...[nbsp][nbsp]:(

It's one of those things that at least 3 planets must be in perfect alignment...[nbsp][nbsp];)

I have already taken the partition offline for further badblock testing and will remap it manually before bringing the other swap back online...

Our apologies for any inconvenience this has caused...

--
Terra
sysAdmin
FutureQuest