Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

redhat 6.2 on a classic...

0 views
Skip to first unread message

Angel

unread,
Oct 29, 2002, 3:41:58 PM10/29/02
to
Hello folks,

I have redhat 6.2 running on two sparc classics (stop looking at me like that,
they were cheap) but I'm having some issues with them...

- First of all, when I try to install a new kernel rpm on them, it fails with
a lot of dependency errors, all referring .o files. Any idea what's wrong here?

- Second, when I leave them running without monitor & keyboard. the message
file quickly fills up as mingetty keeps crashing and being respawned by init.
I currently have mingetty disabled in /etc/inittab, but I was wondering if
there is a better solution.

- Last, both take quite a while to respond to an incoming ssh connection,
although once the connection is made, everything runs fine. Are they just too
old to run sshd smoothly, or is there something I can do about that? I'm
running OpenSSH_3.4p1 on them.

Thanks in advance.

--
I could be SPEWS, but even if I was I wouldn't admit it.

"Nothing is scarier than cluelessness in action."
-Couvier's Law

Hugo van der Kooij

unread,
Dec 21, 2002, 6:57:11 PM12/21/02
to
Angel wrote:
> Hello folks,
>
> I have redhat 6.2 running on two sparc classics (stop looking at me like that,
> they were cheap) but I'm having some issues with them...
>
> - First of all, when I try to install a new kernel rpm on them, it fails with
> a lot of dependency errors, all referring .o files. Any idea what's wrong here?

Sounds like something is broken here. But hard to tell without a sample
of the exact errors.

Untill now my two Sparc machines are cruising around with a higly
updated version of Red Hat Linux 6.2 (see also the package list on
http://hvdkooij.xs4all.nl/software/)

> - Second, when I leave them running without monitor & keyboard. the message
> file quickly fills up as mingetty keeps crashing and being respawned by init.
> I currently have mingetty disabled in /etc/inittab, but I was wondering if
> there is a better solution.

This seems to prove my previous point. Thy are somehow hosed and you end
up with all these pieces.

> - Last, both take quite a while to respond to an incoming ssh connection,
> although once the connection is made, everything runs fine. Are they just too
> old to run sshd smoothly, or is there something I can do about that? I'm
> running OpenSSH_3.4p1 on them.

Sparc CPU's are bad at this crypto thing.

Hugo.


Tamas Gyursanszky

unread,
Dec 22, 2002, 8:19:43 PM12/22/02
to
On Sun, 22 Dec 2002 00:57:11 +0100, Hugo van der Kooij <hvdk...@vanderkooij.org> wrote:
> Angel wrote:

>> - Last, both take quite a while to respond to an incoming ssh connection,
>> although once the connection is made, everything runs fine. Are they just too
>> old to run sshd smoothly, or is there something I can do about that? I'm
>> running OpenSSH_3.4p1 on them.
>
> Sparc CPU's are bad at this crypto thing.

If you run sshd out of x|inetd, with the server arg -i, then
it uses smaller key sizes with take less time to generate.

From man sshd..

-i Specifies that sshd is being run from inetd. sshd is normally
not run from inetd because it needs to generate the server key
before it can respond to the client, and this may take tens of
seconds. Clients would have to wait too long if the key was
regenerated every time. However, with small key sizes (e.g.,
512) using sshd from inetd may be feasible.

I've not tried, but I suspect you could use the -i arg when running
sshd as a standalone daemon.

0 new messages