| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 317.1 | | CSC32::BUTTERWORTH | Gun Control is a steady hand. | Sat Jul 23 1994 00:00 | 5 |
| Nancy,
I sounds like an oversight in the SETCHAR/SETMODE QIO's. Of course Phil
will probably want *me* to fix it now....;-}
Dan
|
| 317.2 | | OPG::PHILIP | And through the square window... | Sat Jul 23 1994 12:16 | 8 |
| >> I sounds like an oversight in the SETCHAR/SETMODE QIO's. Of course Phil
>> will probably want *me* to fix it now....;-}
Sure you can fix it if you want, stick it on your list of stuff to do.
Cheers,
Phil
|
| 317.3 | | CSC32::BUTTERWORTH | Gun Control is a steady hand. | Mon Jul 25 1994 19:21 | 8 |
| >Sure you can fix it if you want, stick it on your list of stuff to do.
Ahhhhhhhhhhheeeemmmmmmmmm!!
I uh... need source code access to do that ......;-}
Regs,
Dan
|
| 317.4 | | OPG::PHILIP | And through the square window... | Mon Jul 25 1994 19:28 | 9 |
|
I know you need access, thats why I only said stick it on your
list and not my normal response of "I'll test your fix in the morning". ;-)
Cheers,
Phil
Patience, I'm still trying to get the new source system going properly.
|
| 317.5 | Spurious shutdown caused by broadcast message | GMTNET::SYSTEM | SYSINIT-F-NOPAYRISE | Thu Nov 17 1994 11:24 | 39 |
|
Hi,
I'll stick this here cos its relevant I guess....
VAXstation 4000 (GMTSNA) ,OpenVMS6.1,PCM 1.5...
I had a C3 window open,with many console connections open.For some
reason or other,I needed to reboot the PCM workstation.I initiated
the reboot,without shutting down the C3,yeah I know,silly thing
to do,but I did it anyway.Next thing I know,two of my cluster nodes
(MV3800's) had gone unreachable.When I looked,the shutdown braodcast
messages had been "blasted" to all the consoles.Most of them had
just read the message and barfed,but two of the VAXes actually
interpreted the word shutdown <cr> and proceeded to shutdown !!
See below
Any ideas ??
Mick
--------------------------------------------------------------------
$
$ SHUTDOWN message on GMTSNA from user SYSTEM at _GMTSNA$FTA13:
17:06:45
%DCL-W-DEFOVF, too many command procedure parameters - limit to
eight
$ GMTSNA will shut down in 0 minutes; back up LATER. Please log
off node GMTSN
A
.
%DCL-W-IVVERB, unrecognized command verb - check validity and spelling
\GMTSNA\
$ SHUTDOWN
SHUTDOWN -- Perform an Orderly System Shutdown
on node DGMT01
etc etc.....
|
| 317.6 | | OPG::PHILIP | And through the square window... | Thu Nov 17 1994 11:30 | 6 |
|
Hmm, do you have the ECO1 patch kit installed, I put a
fix in for just this problem.
Cheers,
Phil
|
| 317.7 | How does the ECO prevent this from happening? | CX3PST::ANASAZ::J_BECKER | There's no substitute for a good boot | Thu Nov 17 1994 16:58 | 18 |
| phil:
>
> Hmm, do you have the ECO1 patch kit installed, I put a
> fix in for just this problem.
>
It sounds to me the user had the 3xxx logged in and had a symbol for shutdown
defined in their logged in environment, so when the message to shutdown
appeared, the system reacted as requested. Does your patch fix this or
prevent this from happening? How does it do this? I would like to understand
the mechanism.
thanks for your time!
john becker
csc/cs
|
| 317.8 | | OPG::PHILIP | And through the square window... | Thu Nov 17 1994 17:19 | 15 |
| John,
The patch set the characteristics of the terminal line (be
it a Direct, LAT, or Pseudo-Terminal) to NOBROADCAST. So,
Any broadcast messages on the PCM host should not affect
the managed systems.
Now just because the user is logged in to the PCM host and
connected shouldnt mean that the shutdown broadcast message
reaches the remote console as it should only break through
on the users sys$output which we do not accept any input
from, we read and echo sys$input to the managed system.
Cheers,
Phil
|