Story so far -- interactive nodes, NFS (nobackup), and EOS are done.  Power coming back in GCC, so work continues otherwise

> On Sep 14, 2023, at 7:52 AM, David A. Mason <[log in to unmask]> wrote:
> 
> This will commence shortly -- will update as things progress!
> 
>> On Sep 13, 2023, at 1:25 PM, David A. Mason <[log in to unmask]> wrote:
>> 
>> Just a reminder this is happening TOMORROW!
>> 
>>> On Sep 7, 2023, at 2:47 PM, David A. Mason <[log in to unmask]> wrote:
>>> 
>>> Greetings,
>>> 
>>> There is a planned power outage in one of the computer rooms housing half our worker nodes next week, and we are going to take that opportunity to catch up on some routine maintenance on the Tier 1 and LPC.  One of those tasks will be to reboot the LPC NFS disks (nobackup) which will be maximally disruptive to LPC jobs and interactive work.  EOS and Tier 1 dCache nodes will also be rebooted.  Expect jobs during this time to fail.  We will try to keep the work to the morning hours, depending on the progress of the power infrastructure work, but have scheduled the downtime for the full day.  We'll make noise as we progress.  We expect to begin holding condor on the nodes to lose power the day before, and the downtime itself should commence at 8 AM FNAL time.
>>> 
>>> Also coincident with this is upgrading cmslpcgpu3 to EL8.  There is already a message on the node you will see when you log in.  This machine will be completely wiped, including any local data that is stashed there, so if you have something on the local data disk you care about (you shouldn't) please retrieve it ASAP or say your final goodbyes.  We will leave cmslpcgpu1 and 2 at SL7 for the time being, but plan to eventually also upgrade them this year.
>>> 
>>> Thanks for your patience!
>>> 
>>> --Dave
>> 
>> 
>