Linux Working Group

AGENDA LOCKED

Meeting Date

• TEAMS: 2022-09-07

Invited

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

Attendees

• Anthony, Dave, Devon, Guoxiang, Todd, Clayton, Nathan, O, Fraser

Review and accept previous meeting minutes.

• CsLWGMeeting20220824

Review last meeting's Action Items

• ubuntu1804-006
  • Machine has been shipped back. Will re-rack and test ourselves.
  • work ongoing
• Lori - will book a meeting re: suexec-flex
  • Need info from Lori on this
  • This is a package that is maintained in-house. It needs patching and re-building.
  • O: I recall that this is an odyssey framework tool.
• Nextcloud upgrade provisioning: https://rt.uwaterloo.ca/Ticket/Display.html?id=1208589
  • Nathan needs clarity on SALT master governance that is blocking him from configuring it. Anthony is going to talk to Dave and attempt to get this clarity in RT above.
    • Feel free to make any configuration changes as necessary
  • DONE
• OpenVPN upgrade: containers and 2FA: https://rt.uwaterloo.ca/Ticket/Display.html?id=1231092
  • Work is ongoing
• DC directory system load of Odyssey data: https://rt.uwaterloo.ca/Ticket/Display.html?id=910469
  • Lori: This is a request for Clayton.
  • Clayton will follow up with Daniel on details and put relevant info in RT above.
  • DONE
• Mike Patterson [UW-RT #1231841] wants us to patch/upgrade cups1.cs.uwaterloo.ca.
  • Fraser will take point on this work moving forward. Anthony offered assistance.
  • It would be handy to have some on campus testers while Fraser works on this. O assigns Costin and Ed for this help.
  • Anthony is communicating with Mike as well.
  • Test server has been setup with latest INF-Ubuntu stable and is being worked on.
  • Work is ongoing

Proposed Agenda Items

• www.cs.uwaterloo.ca outage postmortem [ Anthony, 20min ]
  • Problem was solved and communication occurred in Teams and RT
  • What was the root problem and how can we prevent it ?
  • Lori, Nathan ?
    • The DC proxy was restarted and it was resolved. Logs didn't indicate anything specific. Best guess: IST broke something in the early morning.
    • Has anyone followed up with IST about this ?
  • Fraser and Anthony re-iterate the high-profile nature of this URL and website that has far-reaching "political" impact oncampus and externally for CS.
  • Are there other detectable failure states that occur earlier than the website being down ?
  • Perhaps some monitored tests/checks at relevant points in the "hosting stack". One such point is HA-Proxy.
  • Next step: IST's networking group ought to be contacted to determine whether they made any changes on the day of the incident.
• Review status of printing in CS [ Dave, 20min ]
  • Most clients today appear to talk directly to printers, unlike when networked printing services were originally contrived.
  • Nonetheless, existing networked printing servers appear to provide some value as well.
  • any howtos or twiki pages on using these servers ?
    • https://cs.uwaterloo.ca/twiki/view/CF/Printing
  • can these servers have logging turned on so that we can have more concrete usage-data, at least over time, to drive future decisions on maintaining these services ?
    • there is financial data available with respect to spending on provisioned printers that can be useful (paper, toner, maintenance, etc.), indirectly.
    • there is logging configuration that can be tuned on these servers. Fraser will take a look at this and touch base with Anthony on criteria.

Comments