| All I can tell you right now is that you have a "time" problem with
respect to MCC. There is a bug in the Time routines where VMS converts
the TDF to a delta time of all zeros, and then complains that it is
an absolute time. This is the only way I know of, of getting this
behavior. If you have a TDF of +1:00, then you should be okay. You
discovered another way to reproduce the same behavior. However, there
is still a 99% probability that the problem is time related. Are
running DTSS (DECdts) on your system? Are you running on a cluster?
VMS supports "private" time formatting. Are you using the VMS DT
formatting logicals?
If push comes to shove, you can try the following steps:
1) Do a
$ show logical SYS$LANGUAGES
$ show logical LIB$*
$ show logical/table=LNM$DT_FORMAT_TABLE
$ show time
2) The last command may display:
LNM$DT_FORMAT_TABLE)
"LIB$DATE_FORMAT_001" = "!DB-!MAAU-!Y4"
"LIB$TIME_FORMAT_001" = "!H04:!M0:!S0.!C2"
If any of the logicals are defined, or if the format table is
different, then this may be the problem. Try removing the
logicals and/or defining the format table to the above.
Just as a sanity check, the current time (date is also important)
should be correct.
3) Install the kit, answering NO to the question of, "Do you have the
BMS2 license installed?".
4) The kitinstal will just move the appropriate files to the correct
places (you will probably get a couple of DCL warning messages about
undefined symbols, which you can ignore).
5) Enter
$ @sys$startup:mcc_startup_dir.com
6) If you get the errors, stop. Wait a day (for the time to straighen
out). Try it again the next day. It should work. If it does not
then something is happening which causes MCC to calculate the UTC
time to be EARLIER than the timestamps in the files.
7) Then enter:
$ @sys$startup:mcc_startup_bms.com
8) If you got errors for the first startup you should *not* get errors
for this startup file.
9) Finally, enter:
$ @sys$startup:mcc_startup_bms2.com
If nothing else seems to work, please contact me:
(TOOK::GUERTIN, dtn:226-5974).
-Matt.
|
| Thanks for the quick reply.
Just to answer a your questions :-
- DECdts is not running on the system
- The system is not clustered
I followed your instructions but the problem still persists (instead of
waiting a day i reset the system time and rebooted). However, I notice
that you mention BMS2 in a couple of places. Perhaps I haven't got the
correct combination of kits because there's no MCC_STARTUP_BMS2.COM
in the kit I've got. As I said in .0, the V1.0 kit I've got is the
pre-SSB one (copied from KEEL::DUA1:[PUBLIC.USERKIT.V1_0.SSB]) and the
V1.1 kit the one as held in KEEL::DUA1:[PUBLIC.USERKIT.V1_1.IFT].
Incidently, I'm assuming I need to install V1.0 before installing
V1.1 IFT, is this actually the case ?
_adrian.
|
| You didn't have to install the V1.0 kit before X1.1.0 kit. Actually, I
think that's why you get these problems. I installed X1.1.0 on a clean
system disk, and I didn't get the kind of problems you are referring
to. I entered some guidance in notes 395, if I remember right.
Bj�rn Olav Haugom
|
| I think I've finally fixed it. It doesn't seem to be necessary to
install BMS V1.0 or the toolkit V1.0 as some notes suggest (and which
the release notes included with V1.1 IFT indicate). However, installing
V1.1 only and specifying a tdf of +1.00 still seems to give the
REPOSNONAME error. I finally got it to install OK by using a clean system
disk, putting V1.1 on and specifying a tdf of -4.00 (I think this is US
EST).
Phew... thanks for the help.
_adrian.
|
|
Some confusion out there might be due to the fact that the ORIGINAL
BMS kit did not include the toolkit components. We rebuilt and updated
the kit area with a BMS that included the tools, making the 1.0
requirement no longer necessary.
Sorry this was so confusing but we were just as confused in house,
it was the first kit readied without a toolkit.
|