Re: Network (fwd)

[Frederick Gray <fegray@berkeley.edu>]

> ---------- Forwarded message ----------
> Date: Mon, 10 Oct 2005 19:37:36 +0200
> From: Peter Winter <winter@npl.uiuc.edu>
> To: David W Hertzog <hertzog@uiuc.edu>, Peter Kammel <kammel@npl.uiuc.edu>
> Cc: David Webber <dwebber@npl.uiuc.edu>
> Subject: Network
> 
> Hello to Illinois,
> > this morning Robert Scheuermann from muSR came to me and he asked if one of > you (David or Peter) were around. Since not, he asked if we were affected 
> by PSI network problems. The reason is that there was one yesterday night 
> and the muSR experiment is then really interrupted somehow. So his aim is 
> to ask for more support from PSI (since there is no 24h service) but for 
> that he actually wants other groups to support that. Therefore, the 
> question is, are we also disturbed in case of PSI network problems?
> I was asking around here and we are of the opinion that not (the FTP 
> archive might not work but this actually can wait til the next day). 
> Nevertheless, I herewith forward this question and I can then give him your 
> answer.

Hi, Peter,

I basically agree with your assessment.  Our local network, with a single 
multihomed computer between the DAQ network and the outside world, is designed 
to insulate us from PSI network problems.  We can still take data and
put it in local storage without the PSI network.  We have enough local
disk space to cover about 24 hours of production data.  There is a scenario
that has caused problems for us several times in the past, though, where the 
archive has crashed at 18:00 Friday with no one available to reset it 
until 7:30 Monday morning.  The weekend "Pikett" operator didn't know how to 
deal with the problem and was not able to call in anyone knowledgeable.  
The archive is a critical system, and if they can't fix the underlying problem 
that causes it to crash frequently, they at least need to have a trained
babysitter available every day when the accelerator is operating.  It would be
nice to have someone on call 24 hours/day, but at a minimum there should be
some coverage on every day the accelerator is running, including holidays.

Thanks very much,

-- Fred

-- Fred Gray / Visiting Postdoctoral Researcher                 --
-- Department of Physics / University of California, Berkeley   --
-- fegray@berkeley.edu / phone 510-642-2438 / fax 510-642-9811  --