Re: apparent crashes persist.
On Sun, Jun 25, 2006 at 02:53:17PM +0000, Dimitris Lampridis wrote:
> On Sunday 25 June 2006 12:15, firstname.lastname@example.org wrote:
> > Left it running overnight. 9 full passes through memory, *no* faults
> > detected.
> If you search recent posts on the list, you'll see that I also had problems
> with my memory and they were not discovered by memtest. So, as others have
> suggested too, maybe its better to open your box, remove memories and try
> them one by one for some time. If the problem persists with every single
> memory module, then either your memories are OK, or all of them are bad (but
> the latter, being such an extreme case, would have been probably detected by
Took a while to find just how to reduce the number of memories from 2 to
1. It turns out to be possible, but against all intuition the single
memory had to be installed in the second slot, which was named A1.
Got the machine wo work for awhile with only one G memory, then it
crashed in the usual way. Tried it with the other memory in that slot,
and it worked for longer, but still got around to crashing today.
It crashes quickly when running Pan, when using mozilla firefox with
complex web pages, and slowly when using xjig. Other things probably
crash it too.
It took a while to get to the tests involving Pan and the complex
web0sites, because just to complicate things the DSL line I use for
network died and the phone company had to get around to fixing it.
But they crashed quite quickly when I finally got around to it.
I have an idea that it is network activity (the mouse in X also talks on
the net internally, I believe) during graphics activity that does it in.
No definitive evidence for that, but I do seem to be doing both when it
Once again, when it crashes, I can sometimes still manage to use a ssh
connection to get in from elsewhere. What information should I collect,
and how should I analyse it?
> What is ur memory configuration by the way? I remember you said 2G, but is it
> 2x1G, 4x512M, something else? and what about the manufacturer?
2x1G. But now I'm running 1x1G. Kingston memory.
memtest86+ never finds a thing wrong.
Could be, of course that it's other hardware.
> To UNSUBSCRIBE, email to debian-amd64-REQUEST@lists.debian.org
> with a subject of "unsubscribe". Trouble? Contact email@example.com