| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 2407.1 | More detail - Sigh :-) | IOSG::PYE | Graham - ALL-IN-1 Sorcerer's Apprentice | Fri Mar 12 1993 14:17 | 9 |
| Steve,
None of those ring any bells with me. Can you give us any more
information about this? I'm sure you know we don't care about all the
ones that are specified weak, but are there any other significant
names? I suppose we'll have to see the whole list. I can always delete
the note later, once we've worked out which one was significant!!
Graham
|
| 2407.2 | Give us the STRONG stuff! | AIMTEC::WICKS_A | Oscar the Grouch is an Optimist! | Fri Mar 12 1993 17:36 | 19 |
| Steve,
$MCREA etc are always WEAK REFERENCES - all you ever need to solve
relink problems with undefined symbols are the answers to the
following information
1) %LINK-W-NUDFSYMS, xx undefined symbols - is xx just 1?
2) %LINK-I-UDFSYM, APNDEFS
for all xx
3) %link-w-useundef, undefined symbol APNDEFS referenced
in psect $DSAB PROTOTYPES$ offset %x000034d5 in module
BOB file dev:[allin1.sources]oalibr.olb;
for each xx
It's easy even I can do it!
regards,
Andrew.D.Wicks
|
| 2407.3 | | KERNEL::LOAT | Ahead groove factor 5! Yeah! | Mon Mar 15 1993 10:06 | 14 |
|
re .1
I'll try to get details from the install asap.
re .2
The installation says that there is one undefined symbol. It then gives
lots of %LINK-I-UDFSYM messages, and then one %LINK-W-USEUNDEF for each
of the UDFSYM messages. Now, if I could work out the questions you were
asking in .2, maybe I could get somewhere 8-]
Steve.
|
| 2407.4 | | KERNEL::LOAT | Ahead groove factor 5! Yeah! | Mon Mar 15 1993 17:06 | 21 |
|
Well, the customer has not been telling me the complete truth. He gets
a message saying that there is one undefined symbol, and then lots of
weak references, the *one* %LINK-W-USEUNDEF, undefined symbol message.
The one undefined symbol is FDV$ATERM. Andy was correct. All the others
are weak references.
Given that the first lot of information came from the guy who said that
they'd definitely installed FMS with no problems, I've asked him to
triple-check this one asap.
I thought the normal error for FMS was 68 or 69 undefined symbols. Any
idea what this one is?
Steve.
PS The log file follows in the next note. It's quite large, so
DECwindows noters beware!
|
| 2407.6 | Live from a snowdrift in Alpharetta | AIMTEC::WICKS_A | Oscar the Grouch is an Optimist! | Mon Mar 15 1993 17:29 | 31 |
| Steve,
Here is an edited version of .5 - the first 3 lines below are the
ones I was asking for.
I've deleted .5
....
%LINK-W-NUDFSYMS, 1 undefined symbol:
%LINK-I-UDFSYM, FDV$ATERM
LINK-W-USEUNDEF, undefined symbol FDV$ATERM referenced
in psect $F3_CODE$ offset %X0000062E
in module OAFLO file DUA0:[ALLIN1.SOURCES]OALIBR.OLB;1
....
%VMSINSTAL-E-INSFAIL, The installation of A1 V2.4 has failed.
.
.
%DCL-W-ACTIMAGE, error activating image LIBRTL
-CLI-E-IMGNAME, image file VMI$ROOT:[SYSLIB]LIBRTL.EXE;
-RMS-F-DEV, error in device name or inappropriate device type for
operation
VMSINSTAL procedure done at 23:37
Yes you're right normally you get 68 undefined symbols with FMS but
I'd still try reinstalling that first.
Regards,
Andrew.D.Wicks
|
| 2407.7 | Something more seriously wrong? | IOSG::PYE | Graham - ALL-IN-1 Sorcerer's Apprentice | Tue Mar 16 1993 11:30 | 10 |
| I agree with Andy, check that the FMS images are installed first. Also
that they have the right FMS kit installed, not just the run time kit,
which doesn't have all the images. Also the bit about not being abale
to activate LIBRTL is interesting... Perhaps there's something more
fundamentally wrong with this system? IMAGELIB.OLB sc***ed up, run out
of global thingummies?
Graham
PS I've deleted even more from .6!
|