Scheduled Maintenance on 2013-05-15
The next first working Wednesday of the month we will go into Scheduled Downtime. It will last from 9:00 to 20:00, but we will return to operation as soon as we finish.
As usual, CMS and Atlas queues will be closed 24 hours before the maintenance, and LHCb queue will close 48 hours before the maintenance.
Summary of interventions
We will perform the following operations on the cluster:
Network intervention
- Description: Update firmware of 2 Nexus 5548 switches
- Affected nodes:
- Notes: To be done between 9:00 and 10:00
Upgrade dCache head nodes to 2.2 from 1.9.12
Upgrade dCache pools to 2.2 from 1.9.12
Firmware upgrade on DS3500 controller
- Description:One of the controllers still requires it's firmware to be updated
- Affected nodes:
- Notes: This is a carry over from the previous maintaince as it failed
Install CVMFS on new hardware
- Description: Install CVMFS on new IBM hardware
- Affected nodes: cvmfs1.lcg.cscs.ch
- Notes:
Other minor updates
- Disk parameters of DM devices on
se09.lcg.cscs.ch
are not correct. Need to reboot the system.
- Modify entry of CVMFS on WNs to point to the new cvmfs server
/etc/cvmfs/default.local
should have: CVMFS_HTTP_PROXY="http://cvmfs1.lcg.cscs.ch:3128|http://cvmfs.lcg.cscs.ch:3128"
- Increase RAM available to
cream03.lcg.cscs.ch
and modify the swap usage reporter script /etc/glite-ce-cream-utils/glite_cream_load_monitor.conf
Also this system has been migrated to SL 6.4 and updated with latest packages (including UMD-2 CREAM-CE)
- Move the following VMs to
kvm02
and kvm03
and update them to SL59:
-
arc01
-
arc02
NOTE: this is a Xen VM that requires extra work.
-
lrms01
-
ganglia
-
nagios
-
pub