Re: [Rule-list] Introduction

[Geoff Burling]

On Thu, 18 Apr 2002, Martin Stricker wrote:

> Geoff Burling wrote:
> >
> > On Tue, 16 Apr 2002, Martin Stricker wrote:
>
> > I suspect the CDROM drive because occasionally when slinky (both 0.1.5
> > & 0.1.9a) unpacked the rpms, occasionally I would see on the console
> > screen a report of a MD5 sum mismatch. I also saw reports of a few
> > other kinds of errors (e.g., one was a cpio read error.) Later, I
> > could unpack the rpm package without any problems -- & the packages
> > slinky had varied between the two versions.
>
> It might be the CD-ROM drive. Did you try a lens cleaner?
> But then it might be something else: Did you check the connectors of the
> cables? Sometimes it helps to pull them out, brush them and put them
> back in. Also, if the cables have been curled up or bent they might be
> damaged. I once thought I had a bad harddisk, but a new cable fixed
> it...
>
I had a number of other clues that the CD drive was about to pass on to
the Great InterNetwork in the Sky: trouble getting the drive door to
open, getting the drive to acknowledge that the CD was in the drive.
What should have told me clearly was that at one point I got the
equivalent of a hex dump reported to me when the drive tried to mount a
CD early on; I only know this was the drive's death song when I saw a
post in my local Linux Users' Group email list when practically the
same error message & the puzzled question, ``What does this mean?"

I bit the bullet & bought a brand new CD drive. Slinky did her bit
without a hitch after that.
>
> > Bringing the system back up as I write this failed to recreate the
> > 2.50+ load average. Now I know I didn't imagine this . . . (but it's
> > not the first time I've experienced unreproduceable results.)
>
And Michael Fratoni <address@hidden> wrote:

> This is still quite alpha quality. I expect unreproduceable results. ;)
>
Restarting the test machine a few times after powering it down explained
what happens. anacron apparently starts slocate within the first 30 minutes
after power up. The first time slocate builds the database, the cpu gets
the work out I described; after that, the load may peak a little above
1.00, but performance is not otherwise affected.

It turns out to be nothing major, but will surprise someone who is not
expecting it (like yours truly). Maybe something to add to the user
documentation?

Geoff