* Help understanding error message?
@ 2003-10-23 21:59 Steve Wampler
2003-10-28 1:43 ` Neil Brown
0 siblings, 1 reply; 3+ messages in thread
From: Steve Wampler @ 2003-10-23 21:59 UTC (permalink / raw)
To: nfs
System(s): 11 identical 2-cpu Althon boxes
RH 8.0, kernel 2.4.18-24.8.0/smp
nfs-utils-1.0.1-2
We're trying to track down a problem and need help
understanding what an error message means (and what
might be causing it...)
Background: We have 11 identical Linux RH 8.0 boxes
(dual-cpu Athlons). One of them exports a filesystem
via NFS to the other 10 (and to 5 other [different]
Linux boxes as well).
Running a specific program on each of the 10 machines
works sometimes and not others. The machines on which
it fails changes, but once it fails it continues to
do so until we remount (unmount+mount, not with
"-o remount") the NFS directory. After the remount
the program functions properly.
Here is the error message (split to fit):
jboc-03:18> runwip
../../../Bin/Linux/DHS/vsmWip:
error while loading shared libraries:
/opt/solis/devel/release/Lib/Linux/DHS/libprocess.so:
unexpected PLTreloc type 0xec
Both the program (vsmWip) and the offending shared
library libprocess.so are located on the NFS-mounted FS.
I couldn't find a reference to 'unexpected PLTreloc'
in mailing list archive search.
[Also, I've just discovered that ldd gives a segmentation
fault:
->ldd vsmWip
/usr/bin/ldd: line 1: 15403 Segmentation fault
LD_TRACE_LOADED_OBJECTS=1 LD_WARN= LD_BIND_NOW=
LD_LIBRARY_VERSION=$verify_out LD_VERBOSE= "$file"
->
if LD_LIBRARY_PATH is set to include the library
directory on the NFS mount. (Without that directory,
ldd [correctly] simply identifies the shared libraries
as missing. ldd works properly in either case after
a remount of the NFS mount.]
*ANY* help is appreciated. Thanks in advance!
This is fairly new behavior, though we're now using the
system more heavily than previously.
-Steve
-------------------------------------------------------
This SF.net email is sponsored by: The SF.net Donation Program.
Do you like what SourceForge.net is doing for the Open
Source Community? Make a contribution, and help us add new
features and functionality. Click here: http://sourceforge.net/donate/
_______________________________________________
NFS maillist - NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Help understanding error message?
2003-10-23 21:59 Help understanding error message? Steve Wampler
@ 2003-10-28 1:43 ` Neil Brown
2003-10-28 13:54 ` Steve Wampler
0 siblings, 1 reply; 3+ messages in thread
From: Neil Brown @ 2003-10-28 1:43 UTC (permalink / raw)
To: Steve Wampler; +Cc: nfs
On Thursday October 23, swampler@noao.edu wrote:
>
> System(s): 11 identical 2-cpu Althon boxes
> RH 8.0, kernel 2.4.18-24.8.0/smp
> nfs-utils-1.0.1-2
>
> We're trying to track down a problem and need help
> understanding what an error message means (and what
> might be causing it...)
>
> Background: We have 11 identical Linux RH 8.0 boxes
> (dual-cpu Athlons). One of them exports a filesystem
> via NFS to the other 10 (and to 5 other [different]
> Linux boxes as well).
>
> Running a specific program on each of the 10 machines
> works sometimes and not others. The machines on which
> it fails changes, but once it fails it continues to
> do so until we remount (unmount+mount, not with
> "-o remount") the NFS directory. After the remount
> the program functions properly.
You aren't using "soft" mounts are you? Please tell me you aren't.
>
> Here is the error message (split to fit):
>
> jboc-03:18> runwip
> ../../../Bin/Linux/DHS/vsmWip:
> error while loading shared libraries:
> /opt/solis/devel/release/Lib/Linux/DHS/libprocess.so:
> unexpected PLTreloc type 0xec
>
> Both the program (vsmWip) and the offending shared
> library libprocess.so are located on the NFS-mounted FS.
>
> I couldn't find a reference to 'unexpected PLTreloc'
> in mailing list archive search.
>
> [Also, I've just discovered that ldd gives a segmentation
> fault:
>
> ->ldd vsmWip
> /usr/bin/ldd: line 1: 15403 Segmentation fault
> LD_TRACE_LOADED_OBJECTS=1 LD_WARN= LD_BIND_NOW=
> LD_LIBRARY_VERSION=$verify_out LD_VERBOSE= "$file"
> ->
>
> if LD_LIBRARY_PATH is set to include the library
> directory on the NFS mount. (Without that directory,
> ldd [correctly] simply identifies the shared libraries
> as missing. ldd works properly in either case after
> a remount of the NFS mount.]
>
I don't understand why you say "[correctly]".
Can you copy the library file to /tmp when it is working, and then
copy again when it is failing, and do a
cmp -l
of the two files?
NeilBrown
-------------------------------------------------------
This SF.net email is sponsored by: The SF.net Donation Program.
Do you like what SourceForge.net is doing for the Open
Source Community? Make a contribution, and help us add new
features and functionality. Click here: http://sourceforge.net/donate/
_______________________________________________
NFS maillist - NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Help understanding error message?
2003-10-28 1:43 ` Neil Brown
@ 2003-10-28 13:54 ` Steve Wampler
0 siblings, 0 replies; 3+ messages in thread
From: Steve Wampler @ 2003-10-28 13:54 UTC (permalink / raw)
To: Neil Brown; +Cc: nfs
On Mon, 2003-10-27 at 18:43, Neil Brown wrote:
>
> You aren't using "soft" mounts are you? Please tell me you aren't.
You're right, we're not.
> > [Also, I've just discovered that ldd gives a segmentation
> > fault:
> >
> > ->ldd vsmWip
> > /usr/bin/ldd: line 1: 15403 Segmentation fault
> > LD_TRACE_LOADED_OBJECTS=1 LD_WARN= LD_BIND_NOW=
> > LD_LIBRARY_VERSION=$verify_out LD_VERBOSE= "$file"
> > ->
> >
> > if LD_LIBRARY_PATH is set to include the library
> > directory on the NFS mount. (Without that directory,
> > ldd [correctly] simply identifies the shared libraries
> > as missing. ldd works properly in either case after
> > a remount of the NFS mount.]
> >
>
> I don't understand why you say "[correctly]".
Because not finding the shared libraries is the proper
behavior when LD_LIBRARY_PATH doesn't include the library dir
(as apposed to ldd seg-faulting when LD_LIBRARY_PATH does
include the library dir).
> Can you copy the library file to /tmp when it is working, and then
> copy again when it is failing, and do a
> cmp -l
> of the two files?
Thanks Neil - that's an excellent suggestion! I'll see what the
produces.
--
Steve Wampler -- swampler@noao.edu
Quantum materiae materietur marmota monax si marmota
monax materiam possit materiari?
-------------------------------------------------------
This SF.net email is sponsored by: The SF.net Donation Program.
Do you like what SourceForge.net is doing for the Open
Source Community? Make a contribution, and help us add new
features and functionality. Click here: http://sourceforge.net/donate/
_______________________________________________
NFS maillist - NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2003-10-28 13:58 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2003-10-23 21:59 Help understanding error message? Steve Wampler
2003-10-28 1:43 ` Neil Brown
2003-10-28 13:54 ` Steve Wampler
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.