Linux Working Group



Meeting Date

  • TEAMS: 2021-11-02

Invitees - Attendees

  • Adrian, Anthony (group leader), Clayton, Guoxiang, Lori, Fraser, Devon, Nathan, Nick, Todd, Dave, Lawrence, Omar

Review and accept previous meeting minutes.

Proposed Agenda Items

Moving remaining data from NetApp to DFSc

Model is data moves to /srv/DFSc//{home,mail,region}

  • In General
    • Dan Berry mounts /opt/csw however RSG will look into moving the data directly to his disk.
    • Moving mail mx service to IST by end-of-month so delay moving both regions inbox volumes to first week of December (Hopefully by the first).
      • How would making fuller use of rsyslog services reduce email.
      • talk with Isaac about his use of web logging.
  • TEACHING
    • mail - can move in 30 minutes or so.
    • umount old home directory snapshots -
    • xhier regional trees
  • CS-GENERAL
    • mail
    • xhier regional trees

CS Teaching - slowness of systems - how to address? (All)

  • Ceph: New 420 systems online. Should improve things. Upgrade of gateway systems (NFS/Samba): block old client. Patch and re-upgrade client kernels/ceph drivers
    • all of CS-GENERAL MDS's now on new 420s since Nov 3.
    • Teaching u{0-2,7-9} are also on the new 420s.
    • as of today, we've started devolving have the MDS's on the 422's, didn't have MDS's on 421's.
    • CS-GENERAL home [https://rt.uwaterloo.ca/Ticket/Display.html?id=1186373#txn-28433886][benchmarks]] out 3 times faster than 2 years ago. Planning more performance test soon.
    • strays still a concern. Asked Devon to reduce ICINGA warning threshold. Upcoming Pacific version will eliminate stays issue.
    • client kernel updates (that includes CEPH patch) ASAP. Same with new NFS-Ganesha service.
  • cgroup limiting of memory usage
    • Proof-of-Concept system on ubuntu2004-000.student.cs since Tuesday (RT #1164800) limits to 64GB of memory per system (real hardware or KVM).
    • will deploy to rest of TEACHING general-use systems ASAP.
      • This doesn't need to be applied to CS-GENERAL region at this time.

Hosts for production DB (postgres) containers? (Lori and Nathan)

Interested in hosts with ~1TB of NVMe for DB storage, access to cephfs for backups
  • review where we have suitable disk space that can be repurposed as xfs volume for this service.

Upcoming Teaching exam/assignment schedule

Need to replace a drive (dc-422:sdk). Want to minimize impact. Asking Nic for a 48 hour window to do this.

Changes to Regions:

  • OpenEdx documented need for more processor cores so raised allocation to 64.

BUMPED to next meeting:

Review new pam stack setup and files. (Dave and Clayton)

  • deployment of /etc/ssh/sshd_config, /etc/pam.d/sshd, /usr/share/doc/pam-kerberos-yubikey-duo-message.txt, /usr/local/bin/pam_exec-internal_ip_test.sh script
  • after the deployment of above file, need to run "systemctl reload ssh"
Edit | Attach | Watch | Print version | History: r7 < r6 < r5 < r4 < r3 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r7 - 2021-11-04 - DaveGawley
 
This site is powered by the TWiki collaboration platform Powered by PerlCopyright © 2008-2024 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback