T.R | Title | User | Personal Name | Date | Lines |
---|
581.1 | | CSC32::BUTTERWORTH | Gun Control is a steady hand. | Thu Jan 26 1995 21:25 | 18 |
| >I'm using V1.5-006 on OpenVMS VAX. When you exit from a multi-line
>event list window, a message is generated by the audit server.
More than likely you are setup to log unsuccessful object deletions
>Apparently CONSOLE$ENS_DAEMON.EXE tries to delete the log file, while
>this file is still open for write.
Correct.
Regs,
Dan
|
581.2 | The exit makes the window re-appear ... | KETJE::STAES | Topless = No brains at all | Sat Jan 28 1995 11:23 | 19 |
| > More than likely you are setup to log unsuccessful object deletions
Yes. That's what security audit/alarms are used for.
> >Apparently CONSOLE$ENS_DAEMON.EXE tries to delete the log file, while
> >this file is still open for write.
>
> Correct.
Eh? Correct, but probably not what you wanted. The alarm triggers the
re-creation of the security message window which you just tried to exit
from... Maybe time for a purge/delete option via a (startup) logical?
Nand.
(Who's analysing the security logfiles every day, and has an ENS security
window under his nose. <-) )
|
581.3 | | KETJE::MICHIELS | OpenVMS: No compromise computing! | Sun Jan 29 1995 11:44 | 19 |
| re .1
Dan,
What I wanted to point out it the fact that CONSOLE$ENS_DAEMON.EXE tries
to delete a file that is still open for write. As this is not possible,
the files remains in CONSOLE$TMP.
I've got other PCM action routines where the log file is deleted upon
successfull completion (unless debug is on). I believe this is the wanted
behaviour, but for one reason or another this fails for the multi-line
event list window.
It's indeed a detail; I even don't want to call this a bug.
Probably easy to fix this.
Johan
|
581.4 | | CSC32::BUTTERWORTH | Gun Control is a steady hand. | Fri Feb 03 1995 16:34 | 17 |
| >Yes. That's what security audit/alarms are used for.
I guess I'm supposed to be humbled by this remark ...
EH!!!??
Re -1/
Yes, it should be easy to fix. My reponse in .1 was simply designed to
confirm your suspicions. In no way was I implying that it was the
correct behavior.
Regards,
Dan
|