MPU Meeting Tuesday 11th March 2014


Alastair has incorporated Chris's comments into TartarusWorkFlow. Stephen has no more comments.

Virtual DICE

Toby has made LDAP changes but not yet changed the ACLs. Once he's done that Chris will test Virtual DICE logins from EdLAN.

LCFG client refactoring

No activity.


No activity.

LCFG port to RHEL7 or similar

Chris has been blogging his Fedora 20 progress in the SL7 port diary. The next stage is to get updaterpms managing the RPMs on an F20 machine, with its RPM list coming from an LCFG profile. Once that's done we should all be able to install an F20 VM with a minimal install, add the local yum repositories trivially, then fire up the client and updaterpms and have the package and component configuration automatically managed from then on.

Miscellaneous development

  • Alastair has fixed the updaterpms LCFG status check script.
  • Chris has changed the background for the login screens in the student labs to this. The survey closes on 30 April and the screens will be set back to normal shortly afterwards. (RT66025.)


  • When Stephen shipped the LCFG client update a few related problems had to be fixed - for instance the install profiles and PXE install images had to be changed.
  • Chris has done his pandemic topics and Alastair is working through his.
  • Chris has concluded that "HP Passport" is not relevant to DL180 firmware updates. He'll ask the IT Forum what other people have done.
  • Alastair and Stephen's latest theory about RT65774 is that it works fine until the user logs into a single-screened machine at which point gconf messes up the configuration. The solution would therefore be to set things correctly conditional upon hostname in the startup scripts.
  • Friday's fibrechannel switch failure made it evident that check_multipath no longer flags up the failure of a path. We'd like to know why. Also it would be good to know why the card on budapest could no longer talk to one port. Stephen will try a different card.
  • Stephen reminded us of the need to check file anomalies mentioned in the rkhunter reports and periodically update the stored file properties, as described in his rkhunter component documentation.
  • There's a problem affecting OpenAFS with recent RHEL kernels (408 series onwards for RHEL 6). It's triggered when volumes are multiply mounted on different paths, which is alien to the Linux kernel way of doing things. We don't do that much so shouldn't be badly affected. One problem area might be the package directories so we will refrain from updating the kernel on brendel for now.

This week

  • Alastair
    • Order a spare 600GB disk for waterloo (hot spare)
    • Ask George - what does the TXretransmit value mean for switch connections? (Can't see this using rfe -xl )
    • Consider how to make metropolitan usable by users
      • Decide whether we do this now or not
      • ISOs
      • minimal docs (mostly manual)
      • they'll use virt-manager, but not create machines or change config
    • circulate table of LCFG bugs
    • Consider activities list
    • Look at gnome issue with NX - try inftest account
    • Double check latest web security reports
    • systemd project
      • start writing in blog
      • Modify lcfg components/rc scripts list as a result of COs talk.
      • Complete lcfg-systemd component
      • Update SystemdSchemaProposal
      • Consider how components will work with systemd
    • Pandemic - network
    • Ask Alison to write pandemic NX documentation. - discuss with Alison and Craig
    • Check multipath nagios check script (to look for failures and optionally a loss of redundancy)

  • Chris
    • SL7
      • Try getting a machine to be updaterpms managed from package lists via profile
    • Looking at systemd and F20
    • Announce new release of Virtual DICE to blog (and make clearer now more supported)
    • Ask others in Uni how they keep track of HP firmware updates

  • Stephen
    • LCFG client refactor stage 1
      • report
      • schedule debrief meeting
      • bring for sign off
    • LCFG client refactor stage 2
      • continue development and docs
    • Consider dhcpd component changes
    • Consider activities list
    • Check with SEE what they did to improve NX performance
      • make any easy changes
      • open up staff.nx and announce
    • Start writing up discussion document wrt Centos and SL
    • Look at systemd
    • Reboot hare to test firmware update
    • Pandemic
    • UKUUG talk
    • Swap budapest fibre card

-- AlastairScobie - 11 Mar 2014

Topic revision: r9 - 23 Sep 2019 - 13:33:37 - AlastairScobie
This site is powered by the TWiki collaboration platformCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback
This Wiki uses Cookies