Status:
04.02.24, 18:30
hyades-hyades6 are available for use.
04.03.24, 07:20
All machines up, except:
- hyades-hyades6. Some work remain (BIOS and firmware updates).
- electra3 and beehive25: Hardware and/or software issues.
We expect the remaining machines to come up before lunch.
owl18-owl24 are now retired.
03.03.24, 21:05
- We enable login now.
- Workstations and older compute nodes are up.
- Some work remain on the newer compute nodes with Infiniband. They are up, but login is disabled. Will be available tomorrow.
03.03.24, 08:30
- Integration of new Hitachi storage complete. Migration in progress.
- One name server replaced (main), one reinstalled with RHEL9 (backup).
- /mn/stornext/d23 and /mn/stornext/astro created.
- Move of /astro/local from alruba2 to stornext in progress.
- Infiniband switches upgraded.
- RHEL9 upgrades ongoing.
01.03.24, 10:50 StorNext work proceeding according to plan
29.02.24, 19:15 Shutdown complete
29.02.24, 18:00 Starting shutdown
Downtime:
We are planning downtime on the computer system from Thursday February 29 at 18.00 to Sunday March 3.
During the downtime we will:
- Integrate new Hitachi storage hardware in the system.
- Start migration from the old Hitachi system to the new system.
- Replace two old StorNext gateway servers with new hardware.
- Replace hardware for one of our old name-servers
- Move /astro/local from NFS-server to StorNext volume
- Install a new 1/2 PB file system (/mn/stornext/d23)
- Upgrade the Infiniband swithces in the clusters with new firmware and software.
- Upgrade the operating system from RHEL 7 to RHEL 9 on the remaining workstations (betelgeuse, papsukal, sabik, sothi, bellatrix, zedaron, castor) and as many of the remaining servers as time allows.
Early warning:
We had planned to upgrade StorNext to version 7.2 during the downtime, but the software is delayed. There will be another downtime on May 3. - 4. where we do this upgrade.