[ProgSoc] Black box question

Anand Kumria wildfire at progsoc.uts.edu.au
Thu Feb 3 13:11:43 EST 2011


Having done this a few times, I've found the simplest way is to have
them boot into single user mode and then bring up the network and then
ssh separately.

That means you can identify fsck (or other) issues, and if it does
come up - you have complete remote access to diagnose further.

I've heard of ssh not responding remotely when the machine was
swapping itself to death as well.

Cheers,
Anand

On Thu, Feb 3, 2011 at 12:45 PM, jedd <jedd at progsoc.org> wrote:
> On Wednesday 02 February 2011 22:11:08 you wrote:
>> Might still be worth a check, since the only other thing that comes to
>> mind is if someone accidentally unplugged the server's network
>> cable... The only indication I had of a corrupt file system last time
>> it happened was that nothing could write to their log files even
>> though everything appeared to be mounted correctly.
>
>  You might be onto something - though my gut feel is that the
>  reboot would have identified this, as the fsck on boot would
>  have stalled.  I'll get my sister to do a 'mount' and see if any
>  volumes are RO.
>
>  The network cable is okay, as ping works fine.  I did think
>  that changing the cable might be a cautious move, and I've
>  got a stash of spare cat5's near the server.
>
>  j.
>
> _______________________________________________
> Progsoc mailing list
> Progsoc at progsoc.org
> http://progsoc.org/cgi-bin/mailman/listinfo/progsoc
>
>



More information about the Progsoc mailing list