MPU Meeting Tuesday 26th January 2016


Nothing happened this week.

LCFG Client Refactoring

Nothing happened this week.

SL7 Server Base

This week we continued to work on the support for modern network naming. This will all be documented, but in the meantime, Stephen blogged about the new network device names on 13 January: Network device naming.

Alastair has finished the new LCFG_NETWORK_SET_DEVICE macro. This allows us to continue to use old-style network names (e.g. eth0) in addition to new-style names (e.g. em1). See lcfg/metropolitan and lcfg/sauce for examples.

The interfaces in an ethernet bond can now be specified more precisely using new macros LCFG_OPTIONS_ETHERBOND_FIRST and LCFG_OPTIONS_ETHERBOND_SECOND. When not explicitly defined, these macros default to eth0 and eth1 respectively.

The sysconfig files generated by LCFG will contain the modern network names, but the old style network names will be mentioned in them too, in a comment, to make the relationship clearer.

The Nagios script which checks bonding will now complain if fewer than two of the bond's interfaces are active.


Nothing happened this week.

MPU Services SL7 Upgrade

We met to kick off this project and to agree a plan.

Miscellaneous Development

Disk encryption
The conversion script has been enabled for CO desktops on the develop release.
NetworkManager configure-and-quit option
By default this option is now unset.
"previous" release
this was fine last week (its first week), so the changes to the release scripts were a success, and the previous release is safe to use.
LCFG file registry
Stephen has branched lcfg-ngeneric to enable Kenny to work on his LCFG file registry scheme. He'll probably subclass LCFG::Component.
Slow LCFG slave servers
Stephen deleted a lot of old exam releases and zapped the caches on the slaves. We had a theory that this would get rid of enough cruft to speed up profile compilation somewhat, but sadly it's still mysteriously slower than it used to be. Even so, we think that it will be a good idea to tidy up old exam releases after each exam diet has finished.


Kernel security, and how to build a kernel
There was a kernel security vulnerability for which the fix was not immediately available, so Stephen built our own kernel as a stopgap. The official fix came out shortly afterwards, but the experience was not wasted because Stephen has now thoroughly refreshed the Rebuilding the DICE kernel documentation page. The kernel problem was number CVE-2016-0728 and was described in Redhat bug 1297475 and at
OpenAFS kernel module build problem
While rebuilding the kernel (see previous item) Stephen found that the OpenAFS kernel module would not build; one of its build scripts had an overly specific notion of what the kernel release version ought to be, rejecting anything with .inf in it. The script had to be hacked to get the module to build.
HP Elite 800 G2
This model is the latest "Select PC". It requires the 327 kernel series at a minimum, so this will be deployed to the stable release soon. Stephen has an 800 G2 at the moment so contact him if you want to see it.
AMD Catalyst Pro
This video driver can now be specified using dice/options/video_amd_catalyst_pro.h. This is provided in addition to the familiar Catalyst video driver whose header is still dice/options/video_ati_catalyst.h. honeypot
Alastair is building one.
Innovating Learning Week LCFG event
Alastair has the VM running; it still needs LCFG defaults and headers installed. Kenny is going to produce the course material, and he has suggested storing the course material in subversion for next time. Potential attendees can book a place here.
New Virtual DICE release
Chris has made the second Virtual DICE release of the academic year and will release it soon.

This Week

  • Alastair
    • Inventory project
      • continue working through TartarusWorkFlow
      • consider what next can be integrated into existing system, if anything
      • Check for systemic errors from clientreport
        • Look now that servers don't check monitors
      • Document clientreport
      • Document order sync code
      • Continue work on hpreport processing script
    • Remove default pool if ops meeting agrees
    • Experiment with different window managers under VNC (making the assumption that performance under NX will be similar)
    • Think of a use for 'atom'
    • Deploy encrypted /tmp and swap conversion script
      • Await result from running on CO desktopsNeed to fix a wee problem and test on another set of machines.
      • Need to warn users that Gnome3 may pop up a window about /tmp being full (when script is run)
    • SL7 base server
      • Localhome functionality - use mkhome_dir instead?
      • check metropolitan USB and CD
      • Continue work with FC and LVM
        • investigate interaction between multipath and UDEV
        • check nagios notices if FC cable removed
      • network and bonding
        • currently, whilst we still use eth0 and eth1 as lcfg tags, if eth0 is actually em1 physically, ifcfg-em1 will be created and not ifcfg-eth0
        • rename macro to be SET_DEVICE
        • more experimenting required (and documenting)
        • would be good if network component commented which tag was used to generate each ifcfg-{n} file
        • blog article required (give example of sauce and metropolitan)
    • Schedule MPU meeting to discuss systemd ordering
    • Continue building honeypot
    • Look through latest systemd changes Nothing too exciting.
    • Take MPU SL7 server project start to dev meeting - once Chris has published project plan
    • Fix VBox 5 config under SL7 327 series kernel
    • Finish work on ILW virtualbox image and look for docs branch in LCFG svn

  • Chris
    • Inventory project
      • continue working through TartarusWorkFlow
      • Look at clientreport modules for replacing firmwarereport
    • pkgsearch for SL7
      • reimplement as a yum web front end (yum search for keyword produce an html file of links to cgi to do yum info)
      • Need support multiple platforms
    • Liaise with George over iDRAC documentation (look through ops reports to remind)
    • SL7 -
      • test out rsync / rmirror (both client and server ends) - liaise with Neil
    • SL7 MPU servers
      • update project plan
    • RT tickets close
    • Continue investigating SL6 sleep problem
    • Schedule MPU stargazing meeting

  • Stephen
    • LCFG client refactor stage 1
      • schedule debrief meeting
    • LCFG client refactor stage 2
      • document API
      • blog article (once documentation complete)
    • Think about PD - Interested in ZeroMQ
    • Investigate kernel component pipe moan by using shell commands instead of RPM module => waiting on 7.2 => activities list
    • continue thinking about apacheconf
    • SL7 server
    • rkhunter config needs fixing
    • LCFG server symlink to exam branches - produce reporting script and discuss with Graham
    • Circulate dmesg proposal
    • Continue work on HP 800 G2

-- AlastairScobie - 26 Jan 2016

Topic revision: r7 - 23 Sep 2019 - 13:33:38 - 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