1 |
tdb |
1.1 |
Minutes of meeting 22/01/2001 @ 2pm |
2 |
|
|
Location: UKC Computer Science Meeting Room |
3 |
|
|
|
4 |
|
|
Present: ab11, ajm4, pjm2, tdb1 |
5 |
|
|
Absent: None |
6 |
|
|
|
7 |
|
|
The meeting began, as usual, with a review of the current |
8 |
|
|
status. AJ told the group about the rework of the swing |
9 |
|
|
client, and the work he and Tim had done on the server to |
10 |
|
|
client protocol. Tim reported that the client interface of |
11 |
|
|
the server now conformed to this protocol, but that more |
12 |
|
|
work would need to be done to meet the full requirements. |
13 |
|
|
|
14 |
|
|
Tim also added that the bug found last week, of the server |
15 |
|
|
crashing, had be semi-solved by the use of Queue's. However, |
16 |
|
|
it was noted that the Queue's would still fill up and |
17 |
|
|
therefore cause the same problems. The group made a decision |
18 |
|
|
that when a Queue becomes full, data should be dropped. This |
19 |
|
|
has led to the obvious need for monitoring the state of the |
20 |
|
|
Queue's throughout the server, and this will provide |
21 |
|
|
valuable statistical information. |
22 |
|
|
|
23 |
|
|
Ash said the C++ host was nearing completion, and the group |
24 |
|
|
clarified some of the issues with regard to sampling and |
25 |
|
|
averaging of data. Paul added that the Perl script for |
26 |
|
|
grabbing the system information was now in place, and ready |
27 |
|
|
to be utilitised. Tim was quick to add that the Java host |
28 |
|
|
already used and supported it. |
29 |
|
|
|
30 |
|
|
The group made a decision, last week, to use a Perl script |
31 |
|
|
as a stopgap/solution to the C++ problem of getting system |
32 |
|
|
information. The overhead is negligable, and the code is |
33 |
|
|
portable. It was considered acceptable for now, and in the |
34 |
|
|
long run, could be perfectly usable in a working system. |
35 |
|
|
|
36 |
|
|
Paul made a request for the database to be populated with |
37 |
|
|
regard to demo'ing to Ian on Wednesday and data for |
38 |
|
|
graphing. It was decided to run the server and hosts for a |
39 |
|
|
period of around two days to provide a good supply of data. |
40 |
|
|
Tim added that permission had been granted by John Cinnamond |
41 |
|
|
to run the host on all the Rocks, providing that the load |
42 |
|
|
was not detrimental to other users. |
43 |
|
|
|
44 |
|
|
The group discussed what would be shown to Ian, and made |
45 |
|
|
plans to have things done by then. |
46 |
|
|
|
47 |
|
|
The next meeting will be on Wednesday afternoon after |
48 |
|
|
meeting with Ian - presumably he will have some ideas :) The |
49 |
|
|
meeting room is booked from 3pm to 5pm. |