ldd and nfs question
Re ldd and nfs, segfaults on nfs mounted drive
We are upgrading kernels from 2.0 to 2.1 series and glibc. All of
our machines have glibc now (debian/slink).
I have binaries that work locally but segfault when run from
an nfs mounted drive. This does not happen to all binaries,
but only some, both libc6 and libc5 binaries, some compiled here
and some prepackaged debs.
/admin/bin/ is exported from a libc5 machine with 2.0.3? kernel.
It has a glibc6 binary, lpr(ng) for example:
ldd /usr/bin/lpr
libnsl.so.1 => /lib/libnsl.so.1 (0x4000e000)
libncurses.so.3.4 => /lib/libncurses.so.3.4 (0x40014000)
libc.so.6 => /lib/libc.so.6 (0x40059000)
/lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)
On one admin machine, lprng works fine from /usr/bin. When I
cp -ap to /admin/bin, on the network drive it fails.
ldd /admin/bin/lpr
ldd: /admin/bin/lpr exited with signal 11
But it works on the network drive from other machines, both with
2.0 kernels and 2.1.122.
Can anyone share a clue?
cfm
If I run ldd from the machine exporting /admin/bin it's OK:
ldd /admin/bin/lpr
libnsl.so.1 => /lib/libnsl.so.1 (0x4000e000)
libncurses.so.3.4 => /lib/libncurses.so.3.4 (0x40014000)
libc.so.6 => /lib/libc.so.6 (0x40059000)
/lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)
If I run ldd from another libc5 2.0.3? machine it's OK:
ldd /admin/bin/lpr
libnsl.so.1 => /lib/libnsl.so.1 (0x4000d000)
libncurses.so.3.4 => /lib/libncurses.so.3.4 (0x40013000)
libc.so.6 => /lib/libc.so.6 (0x40058000)
/lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)
If I run ldd from another glibc6 2.1.122 machine it's OK:
ldd /admin/bin/lpr
libnsl.so.1 => /lib/libnsl.so.1 (0x4000e000)
libncurses.so.3.4 => /lib/libncurses.so.3.4 (0x40014000)
libc.so.6 => /lib/libc.so.6 (0x40059000)
/lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)
--
Christopher F. Miller, Publisher cfm@maine.com
MaineStreet Communications, Inc 208 Portland Road, Gray, ME 04039
1.207.657.5078 (MTRF 3-5pm) http://www.maine.com/
Reply to: