Infrastructure meetings take place every other Thurs. at 9:00 Pacific on the BlueJeans infrastructure-meeting channel: https://bluejeans.com/383721668
Date
Goals
- Alignment of NCSA-provided services with program needs
- Ensure effective use of the current NCSA infrastructure
- Refinement and continuous improvement of services, resources and processes
- Plan for near- and medium-term activities
Discussions
| Item | Who | Notes |
|---|---|---|
| Review of last meeting notes |
| |
1TB/1M quotas for /home | Unknown User (mbutler) | Is being moved forward. Starting with documentation – (much more is needed in documentation. resources and policies) Only 1 user is effected by this data, and John Swinbank didn't want to hold up this process by only waiting for this 1 user to report back on what to do with his data. Data for the 1 user should be moved to /projects or /datasets. |
| DRP processing is beginning | Hsin-Fang Chiang | A queue for the processing has been set up and running this week. 1/2PB of storage has been added to /datasets for this data. |
| Qserv nodes harddrive reorg | Unknown User (aloftus) and Fritz Mueller | Reorg of the master and dax1 harddrives are being done this week for the Kubernetes roll out for the PDAC. |
| Developers guide update | Unknown User (mbutler) | Tickets and other mentions above are being added to the developers guide. |
| DTN nodes and new VMsphere | Unknown User (mbutler) | To add the DTN nodes and new vsphere VM nodes to the LSST environment requires changes to the firewall. The impact of the firewall environment could be felt across the PDAC/Slurm cluster/Kubernetes environments and could take out GPFS for the entire LSST machine enviornment for a short time. The impact was too great just to add these 2 different systems to the infrastructure, so it was decided to not undergo this firewall change until other groups could take advantage of this outage such as a new GPFS version.. or something like that. |
| Kubernetes roll out | The NFS solution for /home is being worked on. Straight NFS works fine for /projects /scratch and /datasets, but NCSA is trying an authenicated NFS for the /jhome environment. If this doesn't work by Friday, Monday a straight NFS solution will be rolled out, and an authenicated NFS solution will be still worked on for the future as a research test implementation. | |
| ATS archiver system | Unknown User (mbutler) | The ATS archiver system is being configured with the pathfinder and James Parsons here at NCSA. The system is to be shipped to Tucson in the first week of May after it's been tested at NCSA. |
| Nebula (for Igor) | Igor Gaponenko | Igor and Michelle had a discussion about the current enviornment of Nebula, and all the trouble that Igor has had with it since the changes in January. Michelle will take an action to go and find the administrator of Nebula, and see if the snapshot or the external storage are drivers to the current problems that Igor is having. All these images were working just fine before January just fine before the Nebula update. |
| PDAC Status | Gregory Dubois-Felsmann | Wasn't in attendance. |
| Topics for next meeting |
Action items
Please enter action items in the form
Responsible Person, Due Date, Description
- Unknown User (mbutler), next meeting, nebula current problems list and find a working solution?