These were accepted.
Report from Computing Executive Group
Nothing to report.
Reports from units.
Ian asked that people consider the 3 questions, in particular the third regarding how many new fibre connections will be required in the Forum.
Toby summarised the planned deployment of Prometheus w/c 19th July. Further information can be found in the US Unit Meeting Summary.
Should you need to keep a machine at SL5.4, you will need to define the DICE_STICK_WITH_SL54 macro at the top of the machine profile.
F12 machines should now be upgraded to F13 as we are no longer supporting F12. We will shortly have a script (along the lines of exam-lockdown) to assist with lab upgrades by removing the labourious task of editing each machine profile manually.
Iain is still working on RAT specific rpms so most are not as yet available. If you do require any of them, just e-mail Iain. Toby mentioned an issue with fonts in EMACS. Iain suggested that Toby try moving his .emacs out the way. It could be down to missing fonts. Toby will get back to Iain.
This has affected KB machines only (but not pyrolisk). It was suggested that there could be 4 possible causes
Neil has written a script that tries to identify which volumes are being hammered. Monitoring network traffic as a means to identify the problem has proved difficult due to a lot of 'noise'. Stephen mentioned that he could merge the patch with the next AFS update if required.
Topics for discussion
Although not officially on the list this was briefly discussed. It was felt that this needs to be tracked so will be added to the Action list. As it is not obvious where the problem lies, although a problem in kernel space seems likely, all Units should give some thought to the issue and it will be brought back as a discussion point at the next meeting. US-Unit will try to link all such tickets together and gather as much information as possible in the meantime.
The next meeting should be on Wednesday 28th July in room 4.31, Alison to convene.
Please contact us with any comments or corrections.
Unless explicitly stated otherwise, all material is copyright The University of Edinburgh