Re: ext2 filesystem corruption?!?!?? (fwd)

Doug Ledford (
Mon, 07 Apr 1997 19:04:55 -0500


> A return of 0 proves its neither hardware or ext2, but a failure does not
> indicate anything.
> Sadly, a return of 0 does not *prove* anything.

Technically correct :)

> It merely gives you more
> confidence that it's neither hardware nor ext2.

The test bypasses ext2, so a return of 0 would indicate passing hardware, but
would say nothing about ext2.

> It's entirely possible for a
> test not to trigger a case where the hardware misbehaves, for example, even
> though the hardware might still have problems in some scenarios.

This is also correct, although the testing I've performed has indicated that
the particular test in question will show a lot of errors that might otherwise
pass undetected. Where I first started using this test was to stress tagged
queueing on the aic7xxx driver (especially in relation to the PAGE_ENABLED
flag). Using this test, I've turned up errors under extremely heavy load
whereas other people have left PAGE_ENABLED for months now without a hiccup.
But, then again, they never placed the load on the driver that I did. It's a
false simulation, but it will show up errors on the order of 10 times faster
than doing something like plugging your machine into a live news feed :)

* Doug Ledford                      *   Unix, Novell, Dos, Windows 3.x,     *
*    873-DIAL  *     WfW, Windows 95 & NT Technician   *
*   PPP access $14.95/month         *****************************************
*   Springfield, MO and surrounding * Usenet news, e-mail and shell account.*
*   communities.  Sign-up online at * Web page creation and hosting, other  *
*   873-9000 V.34                   * services available, call for info.    *