[Search for users]
[Overall Top Noters]
[List of all Conferences]
[Download this site]
Title: | POLYCENTER Console Manager |
Notice: | Kits, Scans, Docs on CSC32:: as PCM$KITS:,PCM$DOCS:, PCM$SCANS: |
Moderator: | CSC32::BUTTERWORTH |
|
Created: | Thu Aug 06 1992 |
Last Modified: | Fri Jun 06 1997 |
Last Successful Update: | Fri Jun 06 1997 |
Number of topics: | 1541 |
Total number of notes: | 6564 |
1434.0. "Memory leak with 1.6-309 + UCX REXEC" by WOTVAX::WATSONR (Lambs... so cute... but so tasty !) Fri Nov 29 1996 07:42
Now... here's an odd one !
I seem to have found a consistent memory leak with version 1.6-309. The
setup is as follows:
OpenVMS V6.2-1H3 (Alpha)
PCM V1.6-309
UCX V4.0-10 (server end)
TCPWARE V5.1-4 (client end)
The situation is this. The user runs rsh to our pcm system, where a command
file does a 'set display/tranport=decnet' and fires up 'console c3'. This
process, running CONSOLE$C3.EXE, consumes approx 1k of virtual address space
every hour. This memory consumption is constant - if I plot a graph using PSPA,
then the VASIZE for the user or that image, is a straight line, climbing at
a constant 1k per hour.
If I put back the 1.6-302 CONSOLE$C3, then there is no memory consumption,
and, oddly enough, if the user connects via a 'set host', and the login.com
does the 'set display', then there is no memory consumption.
The UCX$REXECD_STARTUP.LOG file in the user's directory shows nothing odd,
and there seem to be no other side-effects. I do, however, find it extremely
unlikely that there are a constant number of events occuring, that effect the
C3 display.
I have, for the time being, put back just the 302 C3 image. There seems no
problems running like that (ie split PCM versions).
Ross
T.R | Title | User | Personal Name | Date | Lines |
---|
1434.1 | Clear as mud ? | WOTVAX::WATSONR | Lambs... so cute... but so tasty ! | Fri Nov 29 1996 07:44 | 5 |
| Sorry... that wasn't clear.
The user's box is the Alpha, the PCM box is a V6.1 VAX
Ross
|