Hi James,<div>Agreed the 1st seems attractive; the only issue is perhaps whether it is worthwhile to have someone at</div><div>the site (e.g., an operator) a bit earlier in the process in case there's a snag that requires local hands.</div>
<div>Gene/Peggy - would this be hard/prohibitive?</div><div>Alternatively, we can just say that we're not expecting to get started with science observations until 1800 </div><div>at the earliest (=2 hours after we have operator support on site; assumes 1 hour of potential interaction</div>
<div>with James' group and then at least another hour for testing/setup for science observing).</div><div>Joe</div><div><br><div class="gmail_quote">On Mon, Nov 29, 2010 at 1:59 PM, James Robnett <span dir="ltr"><<a href="mailto:jrobnett@aoc.nrao.edu">jrobnett@aoc.nrao.edu</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;"><br>
After a bit of debate we're now leaning toward doing the<br>
filehost replacement Saturday January 1st assuming all the<br>
various parties are amenable.<br>
<br>
During a filehost replacement every Linux box needs to be<br>
down or at least rebooted at very specific times (plural) to<br>
get around stale NFS filehandle issues and general lack of<br>
infrastructure servers. That includes all linux systems at<br>
the EVLA and VLBA ops machines (but not VLBA site machines).<br>
<br>
The scheduling has been heavily influenced by a desire to<br>
avoid affecting normal work hours and EVLA observing. Since<br>
we're not observing part of the 1st anyway the 1st seems to fit<br>
the bill.<br>
<br>
What we'd propose is scheduling it for something like noon<br>
to 4pm. The late start is designed to give folks a chance<br>
to recover from a potentially late night and more importantly<br>
coincide with the operator availability at 4pm.<br>
<br>
I suspect we'll be done sometime before 3pm and spend the<br>
next hour or so checking systems at which point we can hand<br>
things over to Ops for checking at 4pm.<br>
<br>
Right now everything's tentative but I wanted to bounce this<br>
off folks that might need to do be involved to ensure proper<br>
restart of Widar and M&C systems as well as VLBA ops.<br>
<br>
The alternative suggestions have been the evening of the<br>
29th (front end of double maintenance day) or during the day<br>
some other time.<br>
<br>
Personally the 1st seems like the best option.<br><font color="#888888">
<br>
james<br>
<br>
</font></blockquote></div><br><br clear="all"><br>-- <br>Dr. Joseph P. McMullin <br>EVLA <br>Ph: +1 575 835 7315<br>E-mail: <a href="mailto:jmcmulli@nrao.edu" target="_blank">jmcmulli@nrao.edu</a><br>
<a href="http://science.nrao.edu/evla/" target="_blank">http://science.nrao.edu/evla/</a><br><br>
</div>