Also look for failures at the bus level. I have had cases where some component was mucking up the bus, and got long pauses because the OS kept resetting the bus. These caused huge pauses, like 30 seconds or more, and due to how various OS components mostly do synchronous disk access, that kind of thing can have pretty wide-ranging effects.
[For Linux, usually the errors will show up in the dmesg output. No idea where to look for other operating systems.] -scott On Tue, Nov 10, 2015 at 1:52 AM, Simon Slavin <slavins at bigfraud.org> wrote: > > On 10 Nov 2015, at 8:18am, OBones <obones at free.fr> wrote: > > > However, using tools to read the SMART values for this harddrive, I > noticed that it had been eating up spare sectors for quite a while. > > That's actually a good suggestion for you: look at the SMART values, run > any hard disk diagnostics you have, etc.. Once you've backed-up, of course. > > > So to answer, yes, it is definitely possible that the hardware is > crashing and windows does not notice it. > > You will see errors that components report to Windows. But a dodgy sector > on a drive may just cause the driver to retry reading repeatedly (mystery > delays) at the driver level, and be reported as an error only if reading > fails completely. > > Simon. > > _______________________________________________ > sqlite-users mailing list > sqlite-users at mailinglists.sqlite.org > http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users >
