Please note that this report pertains to activity during November 2011.
For Open Doors Day we set our digital signs to show information relevant to the event. Please let us know if you have events coming up and would like custom signage in your area to support the event.
Also, there is a fault with one of our rarely used ISDN lines. ICS/BT are aware of the problem and will advise when it is fixed.
There is no news to report relating to the compute cluster.
Please remember that dMail will support connections from Macs with Office 2011 as well as IMAP clients such as Apple Mail.
Our Windows 7 migration project is progressing division by division. We have completed migration for CAHId, SLSL&T, MMB, CSI, MRC, MM, SCILLS, and PS as well as many support areas. We are progressing migrations for BCDD, GRE and CDB. Our parallel activity introducing Windows terminal servers for divisions has completed one for BCDD and GRE divisions, MAJF group and also support staff, most recently for CLS Finance. We are progressing servers for MMB division and MRC.
We have successfully completed our evaluation of Remote Desktop Gateway services for LIFESCI-AD Terminal Servers and planning deployment by the end of Q4 2011. This will mean that SSL-capable RDP clients can connect without first establishing a VPN connection, and so users will be able to use these servers more easily from home and other institutions while travelling. Note that only Windows clients are able to use this service without additional cost, although cost-effective clients are available via iTap for Mac OS X, iOS and Android devices.
We have virtualised the GJB Oracle server, which involved migrating the disks into SVC and the machine into our virtual infrastructure.
ICS have advised that the ongoing issues with the eDirectory portal have been resolved and the division secretaries are now working on getting the email lists up to date as quickly as possible. We are reconfiguring some of the lists to be automatically populated based on HR and Registry information, along with manual additions/exclusions to ensure the correct membership for these larger lists.
During November there have been several incidents resulting in significant performance degradation with the University's email, IMAP and webmail service. Some of these issues seem to relate to problems with the underlying storage subsystems within ICS.
The email migration project, dMail, is planning to migrate staff and student mail between 6-9 Jan 2012 and "to have the dMail service fully operational by the start of semester 2". More information is available from the project's blog at: http://blog.dundee.ac.uk/dundeemail/.
We intend to ensure that all mail is copied over but it is vital that all Groupwise archived mail is transferred to our central gwarchive server. If you are not sure of the state of your Groupwise archived mail please log a call with Life Sciences Computing. The University also intends to provide a preview of Office 365 to all staff in December. Proxy rights as well as sorting and filtering rules are not transferred so this will give staff time to recreate these before the actual switch over in January.
All Windows machines and Macs running OS X 10.5 and above will be able to access Office 365 Exchange Online via the web application using a web browser. The best way on Windows is to use Outlook 2010, which will only be available on Windows 7 machines. If you have a Windows XP machine at work and wish to use Outlook 2010 please log a call to upgrade to Windows 7 as soon as possible. On Macs with OS X 10.5.8 or higher there is a choice between Outlook 2011 and Apple's Mail/iCal applications. Most modern Linux machines will be able to use the web application and in addition IMAP access is also supported.
On 7 Nov the MSI network went offline when the Uninterruptible Power Supply (UPS) devices failed during our standard battery replacement procedure. This caused a 5 minute outage while we isolated the UPS in bypass mode. We reconfigured the power supply connections during the next day's maintenance period so that the switches are fed directly from the mains and also from the UPS. We also completed the battery replacement procedure during the same 7-9am period.
On 18 Nov we modified the configuration of the UPS on the WTB Mezz floor to support the autodialler for Estates & Buildings.
On 22 Nov we applied client software updates to the experimental Cisco AnyConnect SSL VPN service (we currently use Cisco IPSec VPN).
On 1 Nov we made changes to the CLS UoD_WiFi and eduroam wireless networks to use a new ICS login server which fixed recent login problems. On 22 Nov the VPN and wireless authentication services were offline - service was restored by 08:45 on 23 Nov.
As part of our Windows 7 project, we have also developed a new password management tool to support the requirement for a new directory, LIFESCI-AD, as well as to help synchronise password changes between UoD and Life Sciences passwords. The best way to change your password is via the Life Sciences Directory website at: https://directory.lifesci.dundee.ac.uk/password/change
There are no issues to report.
There are no issues to report.
On 4 Nov ICS carried out urgent work between 6-8pm to update their storage array and filestore cluster.
On Saturday 5 Nov we realigned the cabinet containing the SVC engines, and this work was completed without any service interruption.
On 8 Nov we moved the DAC lab to Alloy storage platform.
On 10 Nov we reported problems for users of the Alloy storage platform, who were affected by the policy engine (used to move data between the tiers of fast and slow storage) continually running. We paused the scanning process until the evening to regain the usual performance levels.
We had planned MRC fileserver migration for Saturday 12 Nov but there was a problem that prevented this work from proceeding. The same problem also affected Mascot, Proteomics and DDU storage. We restarted the node and that restored functionality.
However on 14 Nov we were alerted to problems for MRC PPU staff unable to access their shared drives following the work that was started on 12 Nov. The service was restored by 09:35 on 14 Nov.
Following move of the home and MRC shares to a new storage array we have increased the size of the homes/H: drive quota from 5GB to 25GB. This only applies to researchers who had the standard 5GB quota. Staff who have co-located their homes and group shares together will not see a change to their quota.
On 22 Nov we restriped the /GPFS filesystem across all the disks to increase write performance. This adversely affected performance until the operation completed but users had been notified before initiating this change.
On Sunday 27 Nov our storage virtualisation controller developed an issue regarding its management interface. Following IBM advice we upgraded the software which restarts the nodes, without affecting the service during this process. On 29 Nov we upgraded the software on our IBM V7000 Solid State Storage Array, which is based around the same technology as the SVC unit. Again, all went smoothly with no loss of service.
On 29 Nov, during the weekly maintenance period, we completed migration of dgn-smb, gh-smb, cpd-smb and gac-smb servers and shares.
We are progressing project work to migrate current static sites to the College's Drupal instance and standalone sites to Aegir hosted Drupal sites - approx 20 sites have been migrated so far. We are also investigating the feasibility of providing a self service web deployment tool based on Aegir to standardise functionality wherever possible, and we are evaluating Apache Solr and custom solutions that allow for sharing of content and enhanced search facilities across all sites.
We have obtained development access to the REST APIs and conducted initial trials to integrate the Pure CRIS system with Drupal based sites.
We have improved functionality of LSC/Mantis integration to improve user experience when interacting with the LSC team. We have also received interest from our CLS Finance and Workshop teams in using this improved design for managing their ticketing activity. If you manage a service area and would like to consider using this, or our online resource booking system (Bumblebee), please let us know.
