Re:Unikernels: No Longer an Academic Exercise?
by toast0 on Tuesday October 23, 2018 @11:26PM
(#18289469)
The disk server could conceivably fail because a disk took an absurdly long time to respond to a command --- by all means, update the firmware, but in the mean time, restarting the disk server (and replaying any pending writes, somehow) will get you back on track. At least restarting the disk server in read only mode allows reads to continue to function -- maybe even allows for an orderly evacuation of the data.
[ Parent
| on HN ]
Re:Unikernels: No Longer an Academic Exercise?
by Fnoord on Wednesday October 24, 2018 @05:12AM
(#18290828)
Good thinking. Read-only mode is an interesting suggestion. Filesystems allow this as well, and it can allow for the live extraction of data having to resort to professional recovery means (ie. physical and expensive).
I've had times with bad sectors (Deathstar at the very least) where reads would result in complete OS lock up or very laggy situation. That can be solved by only trying so many times to read (though that was on Windows 9x with PATA).