| 1 |
Minutes of meeting 19/03/2001 @ 2pm |
| 2 |
Location: UKC Computer Science Meeting Room |
| 3 |
|
| 4 |
Present: ab11, ajm4, pjm2, tdb1 |
| 5 |
|
| 6 |
The first part of the meeting was a usual "catch up" |
| 7 |
session. Each of the group members reported back on what |
| 8 |
they'd been doing, and where things stood. |
| 9 |
|
| 10 |
AJ was proud to announce that Conient was now complete, or |
| 11 |
at least as complete as it was going to be for the project. |
| 12 |
There were of course many more features that could have been |
| 13 |
added, but time was restrictive. This has been the case with |
| 14 |
most things. The GUI had been reworking to provide a sorted |
| 15 |
list of hosts, which was much easier to navigate. A few bugs |
| 16 |
had been ironed out with the layout, and a general tidy up |
| 17 |
had been done. |
| 18 |
|
| 19 |
Paul said the web reports were well in progress, maybe even |
| 20 |
finished. The pages for the "latest data" and "alerts" had |
| 21 |
been done, and linked together with the historical side of |
| 22 |
things. A summary alerts page had also been generated. |
| 23 |
The other group members were impressed by the webpages |
| 24 |
generally, but felt that a few more smaller things were |
| 25 |
required. Firstly a "one screen" summary of selected hosts, |
| 26 |
maybe for display on a "public monitor" near reception |
| 27 |
desks. Concern was also raised that the whole lot should be |
| 28 |
even more tightly integrated... so it appeared as one |
| 29 |
system, rather than a collection of three subpages. |
| 30 |
|
| 31 |
Paul also added at this stage that winhost had been |
| 32 |
finished. It now reported uptime correctly, and the icon had |
| 33 |
been made blue (instead of pink). The few minor bugs that |
| 34 |
existed had been ironed out, and an installer had been made. |
| 35 |
AJ reported that it now worked on his system, and didn't |
| 36 |
give an OCX error as before. |
| 37 |
|
| 38 |
Tim was pleased to inform the group that the server seemed |
| 39 |
to be fixed. He and AJ had spent a long night in the |
| 40 |
computer room debugging and fixing parts of the server. In |
| 41 |
the end it seems it was a deadlock (or livelock?) issue in |
| 42 |
the Queue. The system had been running for nearly 48 hours |
| 43 |
without any hassle and was looking good. |
| 44 |
Tim also added that the server was essentially complete, |
| 45 |
except for the LocalClient which still needed a few minor |
| 46 |
points ironing out with alerting. |
| 47 |
|
| 48 |
It was also noted that statgrab and ihost were as good as |
| 49 |
finished, although a few minor issues still needed solving. |
| 50 |
These would be solved ASAP. |
| 51 |
|
| 52 |
Ash informed the group that he had finally solved the 1024kb |
| 53 |
UDP packet size problem on the C++ host. It required |
| 54 |
changing the socket library, which was not satisifactory, |
| 55 |
but it at least worked. He also added that documentation on |
| 56 |
both the C++ host and the Java host had begun. |
| 57 |
|
| 58 |
The group noted the urgency to complete documentation. With |
| 59 |
little over a week to go, and a lot to write, things were |
| 60 |
getting very close. As coding had virtually completed now, |
| 61 |
documentation could be done throughout the week. As luck |
| 62 |
would have it, everyone had a databases assignment due in |
| 63 |
midweek, which may take up valuable time. |
| 64 |
|
| 65 |
There will be a meeting with Ian at 12pm on Thursday, and |
| 66 |
the usual meeting on Monday at 2.15pm. |
