From aldavis at wi.mit.edu  Fri Sep 15 11:12:29 2006
From: aldavis at wi.mit.edu (Al Davis)
Date: Fri, 15 Sep 2006 11:12:29 -0400
Subject: [CSBi-HPC] NE47 power outage - Sat Sept 23rd
Message-ID: <450AC2DD.7010300@wi.mit.edu>


As part of the ongoing NE47 building renovation for Novartis, there will be a 
building wide power outage on Saturday Sept 23rd.  The power will be out the 
whole day which means there won't be any power in the computer server room. 
I'll be turning off the CSBi IBM cluster here in NE47 Friday night and turn it 
back on Sunday - plan accordingly.

MIT has agreed to pay for the installation of an electrical circuit to connect 
the computer room UPS to the building emergency generator, which means we'll be 
able to keep the lights on (and the computers) even when the building power goes 
out.  This is a fairly large project, so the ETA for this is still several 
months away and thus it won't be installed in time for this power outage.


Respond back if you have questions/concerns.

thanks, al
-- 
Al Davis                           aldavis at wi.mit.edu | aldavis at mit.edu
Systems Manager                    617.324.0519
CSBi & WI/MIT BioImaging Center    NE47 Rm 311 (500 Technology Sq)


From aldavis at wi.mit.edu  Sun Sep 24 18:29:03 2006
From: aldavis at wi.mit.edu (Al Davis)
Date: Sun, 24 Sep 2006 18:29:03 -0400
Subject: [CSBi-HPC] BIM cluster is not working
Message-ID: <451706AF.8090408@wi.mit.edu>

Having a problem getting the IBM cluster back up after the power outage.  There 
is a hardware problem with the system which needs to get fixed.  Working on 
several solutions, will keep you up to date.

al
-- 
Al Davis                           aldavis at wi.mit.edu | aldavis at mit.edu
Systems Manager                    617.324.0519
CSBi & WI/MIT BioImaging Center    NE47 Rm 311 (500 Technology Sq)


From aldavis at wi.mit.edu  Wed Sep 27 12:44:42 2006
From: aldavis at wi.mit.edu (Al Davis)
Date: Wed, 27 Sep 2006 12:44:42 -0400
Subject: [CSBi-HPC] IBM cluster is back up
Message-ID: <451AAA7A.9020905@wi.mit.edu>

The IBM cluster is back up after repairing the cluster management console system 
(replaced a dead system drive), reloading and then reconfiguring all the 
software until it could talk to the cluster nodes and tell them to turn on.

BIG thanks to IBMs Ed Geraghty, who helped throughout the whole process and knew 
the last bit of magic that enabled the management console to access the cluster.

Let me know if you find something that doesn't work as I haven't tested 
everything yet.

al
-- 
Al Davis                           aldavis at wi.mit.edu | aldavis at mit.edu
Systems Manager                    617.324.0519
CSBi & WI/MIT BioImaging Center    NE47 Rm 311 (500 Technology Sq)


From aldavis at wi.mit.edu  Fri Sep 29 08:04:42 2006
From: aldavis at wi.mit.edu (Al Davis)
Date: Fri, 29 Sep 2006 08:04:42 -0400
Subject: [CSBi-HPC] IBM cluster home dir is full
Message-ID: <451D0BDA.10209@wi.mit.edu>

The filesystem used for home directories on the IBM cluster mitwilogin is full 
(~650GBs).  This is preventing any user jobs from being processed.

Please remove any unneeded files in your home directory, so applications can run 
again.

thanks, al
-- 
Al Davis                           aldavis at wi.mit.edu | aldavis at mit.edu
Systems Manager                    617.324.0519
CSBi & WI/MIT BioImaging Center    NE47 Rm 311 (500 Technology Sq)