|
From: Bruce A. <ba...@gr...> - 2008-04-13 15:04:49
|
Hi Simon, I would suggest that you take the second approach and schedule self-tests at the most probable time that you computer is on. This means that you will miss tests from time to time. That's OK, as long as they are run on a regular basis. My advice however would be to do a LONG not a short self test once per week. You want to know if some sectors of your disk have become unreadable; this is the best way to find out. Cheers, Bruce On Sun, 13 Apr 2008, Simon Bachmann wrote: > Hi, > > I'm setting up smartmontools to monitor my laptop's HD. Thanks to the > very good manpages, the basic setup was pretty easy, but now I've got a > question about scheduling self-tests with the '-s' option. My computer > does not run 24/24 hours a day and there isn't really a time in day > where it's pretty sure it is on. However, there are days in the week > where I (almost) always will use it. So my first thought was to use > something like this: > -s S/../../7/.. > to run a short self-test every Sunday, no matter at what time. > The problem is, if I understand things correctly, that with such a > regex, a short self-test will be run after every device poll that day. > Running a self-test every 30 minutes doesn't seem to make much sense, so > is there a way to avoid this? > Or should I just schedule self-tests at a time where it is most probable > that the computer is on and live with the fact that sometimes longer > periods of time will pass between two self-tests? After all I don't > really have any "vital" data on the disc, so the paranoia-level must not > be that high... > > Thanks in advance > > Simon > > |