egisin%Math%math.waterloo.edu@RELAY.CS.NET.UUCP (11/21/86)
I can get sysgen to connect vta0 when used interactively, but when it is run at startup from syconfig.com the same command fails with the message `sysgen-e-nounload, driver not unloadable'. A connect noa0 works fine in my syconfig.com. The only unusual thing on this system is that I've redefined sys$system to be a path with sys$sysroot:[sysexe] and a local software directory, but I dont' see how that could affect this. Does anyone know what's wrong?
oberman@LLL-ICDC.ARPA ("Oberman, Kevin") (11/24/86)
>I can get sysgen to connect vta0 when used interactively, but when it is run at >startup from syconfig.com the same command fails with the message >`sysgen-e-nounload, driver not unloadable'. A connect noa0 works fine in my >syconfig.com. The only unusual thing on this system is that I've redefined >sys$system to be a path with sys$sysroot:[sysexe] and a local software >directory, but I dont' see how that could affect this. Does anyone know what's >wrong? I'm not certain, but I suspect that you may not have the right words in SYCONFIG.COM. The VT device is odd in that it has no driver. It is just an alternate entry point in the TTDRIVER. It is therefore important that when it is connected, the following syntax is used: CONNECT VTA0/NOADAPTER/DRIVER=TTDRIVER There must be NO filespec in the /DRIVER qualifier. The TTDRIVER is already in place. I suspect that you may have something like /DRIVER=SYS$SYSTEM:TTDRIVER. This won't work because the TTDRVIER is already loaded and. of course, can't be unloaded. R. Kevin Oberman LLNL arpa: oberman@lll-icdc.arpa (415) 422-6955 ------
LEICHTER-JERRY@YALE.ARPA (11/24/86)
I can get sysgen to connect vta0 when used interactively, but when it is
run at startup from syconfig.com the same command fails with the message
`sysgen-e-nounload, driver not unloadable'.... The only unusual thing on
this system is that I've redefined sys$system to be a path with
sys$sysroot:[sysexe] and a local software directory, but I dont' see how
that could affect this.
I've seen connect's of vta0 work fine in system startup files, so there's
nothing generically wrong here. I'll make a guess: Somehow or another, you
are trying to connect the driver twice, and the second try is crapping out.
One way this MIGHT happen is if you somewhere have a loop that looks for all
command files that match some spec and execute them. If you had copies of
the same command file in both SYS$SYSROOT:[SYSEXE] and the local software
directory, you might find some file twice...but this is guesses on top of
guesses.
Note that F$GETDVI("VTA0:","EXISTS") will be TRUE if and only if the VT driver
has already been connected. Have the command file where the problem occurs
check and tell you if it finds the device already there; then you'll know if
my guess was right.
-- Jerry
-------