Project planning meeting - 18 September 2006
Present: Stuart Yeates (SY), Graham Klyne (GK), Mike Fraser (MF, part of meeting)
Last report: SakaiVre/PlanningProgress/20060904
This report: SakaiVre/PlanningProgress/20060918
Next meeting: Mon, 2 Oct 2006, 09:00. SakaiVre/PlanningProgress/20061002
- Review agenda
- Confirm date/time of next meeting
- Discussion of specific issues
- Future plans, with Mike Fraser
- Search requirements, with Mike Fraser
- Conversion to Debian Linux
- Review actions outstanding at last meeting
- New activities and issues arising from current work
2. Activity since last report
Since the last meeting, we have run into some serious problems with the new system hosting environment (see below), and have focused on documentation and other future tasks for the time being. In particular, GK is focusing on documentation and handover activities, while SY is examining portal customization and federated search activities for the final phase of the project.
- We have experienced continuing problems with transfer of our installation to Debian. Now, it is the Shibboleth SP software that seems to be incompatible with the Debian setup. This is rather unexpected and very disappointing, since so far the Shibboleth software has been very dependable.
- Disaster! The system hosting the Xen virtual machine for our new Sakai and Shibboleth SP installation, as well as the SPIE IdP, has become caught in a boot/crash/reboot cycle. And the sysadmin is away for the month. It appears to be a problem induced by a Debian update incompatible with the Xen/Debian kernel of the VM host system.
2.1. Actions closed or completed
[SY] Ensure that gotchas with Sakai installation (esp need for trailing '/' on paths) are recorded at SakaiNotes. Done.
[GK] Check Christian's availability with Mike Fraser, in case of problems using the SPIE IdP (which we don't anticipate) Closed - this is rather overtaken by events.
[SY+GK] Modify and document Sakai configuration to persist user database. (Currently, new users added to Sakai are lost on each restart of the system.) The configuration change seems reallly easy: see http://maillist.ox.ac.uk/ezmlm-cgi?3855:mss:135:200607:pfcnekgcelolllgcmnmp. This change needs to be applied to Sakai when the new installation is complete (see 20060803.2). Done. See http://wiki.oss-watch.ac.uk/SakaiNotes, configuration section.
2.2. Actions progressed
[SY] Analysis of search requirements. SY has reactivated this activity, looking at options for discovery of people and research interests based in part on directory data and personal FOAF information. See discussion section, SakaiVre/LDAPToFOAFIdea, and SakaiVre/AvailableToolReview. (Stuart: I think a page or some words about the origin of the search requirements leading to these ideas would be helpful - #g.)
- [GK/SY] Write Shibboleth configuration document, in particular showing how the various configuration elements and functions are inter-related. GK has started work on a Shibboleth configuration chart, which is now substantially complete.
2.3. New activities and notes
- [GK] Notify Nicky Tomlin of my planned leaving date.
- [SY] Provide MF with a couple of paragraphs of information about FOAF based federated search work, for presentation to Rob Crouchley, to confirm this is in line with the overall project plan.
[SY] Draw up plans for customization of the Oxford portal demonstrator. (This activity was actually started in the last period, but wasn't formally raised as an action in the previous report. See SakaiVre/VreCustomisations.)
2.4. Summary of ongoing actions carried forward
- [SY] Analysis of search requirements.
[GK] Initiate process of SDSS federation credentials. Have obtained an SDSS CA certificate, but have not yet joined the SDSS federation. See http://maillist.ox.ac.uk/ezmlm-cgi?3855:mss:84:200606:hbmpafckdeclpnjmmgmb. The current plan is to enroll the new machine as an SDSS SP (using the new certificate obtained), and use the SPIE IdP, which is already enrolled with SDSS. Dependent on action 20060807.2.
- [GK/SY] Write Shibboleth configuration document, in particular showing how the various configuration elements and functions are inter-related. Also, note relationship between Shibboleth and Eduperson attribute schema, noting in particular Shibboleth's treatment of attributes as a flat namespace, coupled with their more structured interpretation by Shibboleth-using applications.
- [GK] Write up things learned about Sakai authentication and authorization structure and the consequent options for integration with Shibboleth.
[GK+SY] Transfer Sakai installation to a Xen virtual machine on the SPIE system hardware. The new virtual machine is running Debian Linux (a supported kernel is necessary for reliable operation under Xen). The change to Debian is throwing up a few issues of system configuration that we need to work through. We'll keep the old machine running for a while as a Shibboleth test machine. See //wiki.oss-watch.ac.uk/DebianNotes and 20060721.2. We have started this migration. Tasks still be be completed include:
Install IPTables configuration. Done. See DebianNotes
Install SSSHBlack. Done. (Description of installation on debian still needed.)
Install and configure HFS backups. Done. Manual backups working, need to check whether weekly backups work.
Install Postfix Done
Install Logwatch Done
Configure Logwatch messages Done
Configure auto-updater Done
Install and configure Apache Done
Install and configure Tomcat Done
Install and configure JK connector Done
Install and configure basic Sakai portal. Done.
Install and configure Shibboleth SP Installed but not yet configured. (Note action 20060721.3.)
3.1. Future plans
(Mike Fraser was present for this part of the meeting.)
GK's project involvement will continue until the end of October, two weeks of that period (3-17 october) being annual leave.
We are awaiting Christian Fernau's return to resolve problems with the hosting environment (see below). Meanwhile GK is focusing on documentation, and SY is focusing on federated search and customization/branding of the portal demonstrator.
3.2. Search requirements
(Mike Fraser was present for this part of the meeting.)
SY is picking up the search aspects of this project.
MF noted that we should re-activate Matthew Dovey's JAFER work and include that in our portal. Also, there is mention in the project plan of Z39.50 and SRU via JSR-168 and WSRP protocols. SY noted that he had tracked the work to deploy JAFER in Sakai to Chris Awre at Hull.
MF noted that the Humanities VRE project aimed to create a demonstrator scholarly portal relating to the works of Jane Austen, whose requirements should be taken into account. I.e. what kind of information do Jane Austen scholars want to be able to find?
SY is looking at options and techniques for searching for people, driven by user requirements to find research collaborators and related work. (See SakaiVre/LDAPToFOAFIdea and SakaiVre/AvailableToolReview.) Stuart's idea is to allow researchers to augment their institutional directory information (from LDAP, etc.) with information about personal interests, provide mechanisms to present that as a FOAF profile, and hook into existing FOAF tools to find links between researchers. As we proceed in this area, we need to be very aware of privacy concerns, at the level of determining what particular information is disclosed, and in partocular that personal information is only disclosed when the subject has given informed consent for the specific information concerned.
(Stuart: I've come across some related material in my data web investigatioons, so we should discuss and maintain continuing contact about these ideas. We are also in discussion Dan Brickley (inventor of FOAF) about participation in the Data Webs project. I think there may be real synergies here. #g)
MF asked SY to provide a couple of paragraphs of information about his FOAF based work for presentation to Rob Crouchley, to confirm this is in line with overall project goals.
Also noted was Anne Bowtell's Research Discovery System, which is due to go live next month. This has a particular focus on the needs of the Medical Sciences Division.;
3.3. Conversion to Debian Linux
We have experienced continuing problems with transfer of our installation to Debian. Now, it is the Shibboleth SP software that seems to be incompatible with the Debian setup. This is rather unexpected and very disappointing, since so far the Shibboleth software has been very dependable.
The system hosting the Xen virtual machine for our new Sakai and Shibboleth SP installation, as well as the SPIE IdP, has become caught in a boot/crash/reboot cycle. And the sysadmin is away for the month. It appears to be a problem induced by a Debian update incompatible with the Xen/Debian kernel of the VM host system. Further progress on the conversion to debian is currently stalled due to these problems.
GK is now focusing his efforts on documentation of work conducted so far. A Shiobboleth configuration chart is almost complete, and has been reviewed by SY.
Other topics to be documented include:
- Shibboleth and "n-tier authentication"
- Shibboleth and Sakai authentication mismatches
- Specific issues with portal systems and WSRP for research environments
Most of the raw material for this exists in the project wiki, and needs pulling together into some kind of overview and synthesis.
4. Notes for next meeting
(Matters arising following the planning meeting.)
[GK] Notify Nicky Tomlin of my planned leaving date. Done.
-- GrahamKlyne 2006-09-18 11:02:45