Subject: RE: VERY slow ssh logins to uVAX
To: 'NetBSD/vax Discussion List' <port-vax@NetBSD.ORG>
From: Gregg C Levine <>
List: port-vax
Date: 05/07/2005 16:56:15
Hello from Gregg C Levine
I'm probably landing here in this subject too late to make a
difference, but here goes:

I seem to recall this happening, when someone, who I can't recall,
gave me temporary access privileges to his system, via SSH. It indeed
moved very slowly. I also included latency from the Internet, plus the
delays imposed by my ISP within its networking.

And this was during my dialup days.

By contrast, when doing an SSH link between my Linux box, who's on
downtime at the moment, and my client's offices in Manhattan, I saw
much the same thing. I also remembered to take into account the
latencies caused by my ISP, who's the same group, despite moving to
DSL some years ago,  and the poorly managed network belonging to my
client. (Poorly managed, because I'm not the fellow who's doing the
managing!) Now I was using a freely available client via the office
systems, at my client in Manhattan, and my setup here.

Now I'm not thinking this subject is starting to get old and gray, but
some of these posts look like restatements. Not that yours is, Greg,
or the ones from the fellow in Canada, but others are. Including the
fellow who started this discussion by throwing old news at us.
Gregg C Levine
"The Force will be with you...Always." Obi-Wan Kenobi
"Use the Force, Luke."=A0 Obi-Wan Kenobi

> -----Original Message-----
> From: []
> Behalf Of Greg A. Woods
> Sent: Saturday, May 07, 2005 3:24 PM
> To: NetBSD/vax Discussion List
> Subject: Re: VERY slow ssh logins to uVAX
> [ On Friday, May 6, 2005 at 23:10:39 (+0200), Klaus Klein wrote: ]
> > Subject: Re: VERY slow ssh logins to uVAX
> >
> > On Friday, 6. May 2005 02:04, Greg A. Woods wrote:
> > > [ On Thursday, May 5, 2005 at 15:01:40 (-0700), Aaron J. Grier
wrote: ]
> > > > which crypto algorithms use floating point?
> > >
> > > I don't really know for sure (esp. w.r.t. OpenSSL and OpenSSH),
> > > than what "egrep 'float|double' */*" shows in
> > > (which seems to be most of them :-)
> >
> > What you're seeing there are the timing calculations of the
> > algorithms' _benchmark_ modules.
> Yes, that could well be.
> In any case the problem is endemic to OpenSSH and it is easily
solved by
> using SSH-3.x from pkgsrc/security/ssh2 instead.  :-)
> This thread more or less mirrors one a couple years ago:
> There may indeed also be some way to improve the performance of
> by using better compiler optimization flags, especially on sparcs
> there seems to be a wider diversity of performance-related CPU
> over the years.  However I don't think anyone has yet had the
> to benchmark the differences, if any, between an ideally optimized
> of OpenSSH against an identically compiled SSH-3.x.  All I know is
> I was surprised by the complaints in the above referenced thread
until I
> realized that those folks were using OpenSSH and I wasn't seeing
> any similar problems because I was using SSH-3.x  ;-)
> --
> 						Greg A. Woods
> H:+1 416 218-0098  W:+1 416 489-5852 x122  VE3TCP  RoboHack
> <>
> Planix, Inc. <>          Secrets of the Weird