I'd like to point out that this sort of thing could happen, for example, when the host itself has to reboot, at which point all the running Linodes are shut down, and then restarted after the host comes back up. At this point, *every* Linode that was up before the reboot would be taking up CPU, probably leading to quite a low bogomips value.

[edit: spelling]

IIRC, it was originally used for the ftape driver which required a tuned timing loop to work properly. I have no idea if that driver is still supported, or whether anything else uses this value.

_________________
Rgds
Stephen
(Linux user since kernel version 0.11)

Just as an FYI... after host20's problem and reboot this week, I noticed the bogomips are now rated to 3853.51. While it is a pretty bogus number as previously stated, just thought I would make mention of it.

So, is there any possibility of getting some sort of MRTG or CACTI graphs of host CPU, Load Average and IO rates? I'd be nice to use that to determine when to run certain cron jobs or determine if a slowness is just inside my VPS or due to the host being busy.