Hi,
sccs01, which exports /nfs/slac/g/hps2,
needs to be re-ip'd and the outage has been
scheduled from 2pm to 3pm TODAY.
Access to that area should be avoided during
that period and jobs that might access that
area should be stopped well before. unix-admin
contacted me earlier today because they were concerned
about the amount of nfs traffic they were seeing.
After looking in ganglia
(http://ganglia.slac.stanford.edu:8080/ganglia/fileservers/graph.php?g=packet_report&z=large&c=nfs-auto&h=sccs01.slac.stanford.edu&m=load_one&r=2_weeks&s=descending&hc=4&mc=2&st=1395937168)
I see that there is some process triggering
high network activity daily starting at about
9am and going to about noon but sometimes
an hour or more later. It would be appreciated
if this could be avoided/stopped today or at least delayed
until after the outage.
Cheers,
Homer
NB:
Also, I noticed that that area is still
close to 100% full. When an area becomes that
full then the performance can be severly
degraded until space is freed. Could some data
please be moved to another hps area presumably
on the large 30TB area. The occupancy should
be less than 95%. Of course, please wait until
after the outage to move any data.
########################################################################
Use REPLY-ALL to reply to list
To unsubscribe from the HPS-SOFTWARE list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1
