Ticket 14595

Hi Ali,

At first glance this looks like it has to do with x11.  Can you list your installed packages and grep for x11 related ones?  I'm curious if you add the '--disable-x11' flag to your configure line if you're able to build?  You don't have to leave it disabled, but that would let us confirm that it is related to the x11 packages.

Thanks,
Ben

Hi Ben,

I ran:

./configure \
    --prefix=/opt/slurm \
    --sysconfdir=/opt/slurm/etc/slurm \
    --without-shared-libslurm \
    --enable-pam \
    --with-pam_dir=/lib/x86_64-linux-gnu/security \
    --with-pmix=/usr/lib/x86_64-linux-gnu/pmix \
    --disable-x11
make

and it appears the result is the same. I will attach the build output and the list of x11-related packages.

Created attachment 25998 [details]
slurm 22.05 build error --disable-x11

Make output with --disable-x11.

Created attachment 25999 [details]
dpkg -l | grep -i x11

List of X11-related packages on build image.

Is there any more information I can provide to help with this? I am happy to run more builds or debugging as needed.

(In reply to Ali Nikkhah from comment #9)
> Is there any more information I can provide to help with this? I am happy to
> run more builds or debugging as needed.

We have recreated the issue locally and are working on a patch set to correct the issue.

(In reply to Nate Rini from comment #10)
> (In reply to Ali Nikkhah from comment #9)
> > Is there any more information I can provide to help with this? I am happy to
> > run more builds or debugging as needed.
> 
> We have recreated the issue locally and are working on a patch set to
> correct the issue.

Thanks!

Created attachment 26068 [details]
patch for 2205 (v2)

(In reply to Ali Nikkhah from comment #11)
> (In reply to Nate Rini from comment #10)
> > (In reply to Ali Nikkhah from comment #9)
> > > Is there any more information I can provide to help with this? I am happy to
> > > run more builds or debugging as needed.
> > 
> > We have recreated the issue locally and are working on a patch set to
> > correct the issue.
> 
> Thanks!

Please try this patch to see if it resolves your issue.

Hi Nate,

The patch does not resolve the issue for 22.05.2. I modified the patch slightly so it would apply cleanly (just removed the NEWS modification) to 22.05.2.

patch --verbose -p1 < ~/git/slurm-package/src/patches/bug14595.2205.v2.patch
./configure \
    --prefix=/opt/slurm \
    --sysconfdir=/opt/slurm/etc/slurm \
    --without-shared-libslurm \
    --enable-pam \
    --with-pam_dir=/lib/x86_64-linux-gnu/security \
    --with-pmix=/usr/lib/x86_64-linux-gnu/pmix
make

Error:

/bin/sh ../../../libtool  --tag=CC   --mode=link gcc  -DNUMA_VERSION1_COMPATIBILITY -g -O2 -fno-omit-frame-pointer -pthread -ggdb3 -Wall -g -O1 -fno-strict-aliasing -export-dynamic  -Wl,-rpath -Wl,/usr/lib64 -L/usr/lib64  -o slurmstepd container.o slurmstepd.o mgr.o task.o slurmstepd_job.o io.o ulimits.o pdebug.o pam_ses.o read_oci_conf.o req.o multi_prog.o step_terminate_monitor.o x11_forwarding.o ../common/libslurmd_common.o ../../../src/api/libslurm.o -ldl -lhwloc -lpam -lpam_misc -lutil  -lpthread -lm -lresolv 
libtool: link: gcc -DNUMA_VERSION1_COMPATIBILITY -g -O2 -fno-omit-frame-pointer -pthread -ggdb3 -Wall -g -O1 -fno-strict-aliasing -Wl,-rpath -Wl,/usr/lib64 -o slurmstepd container.o slurmstepd.o mgr.o task.o slurmstepd_job.o io.o ulimits.o pdebug.o pam_ses.o read_oci_conf.o req.o multi_prog.o step_terminate_monitor.o x11_forwarding.o ../common/libslurmd_common.o ../../../src/api/libslurm.o -Wl,--export-dynamic  -L/usr/lib64 -ldl -lhwloc -lpam -lpam_misc -lutil -lpthread -lm -lresolv -pthread
/usr/bin/ld: ../../../src/api/libslurm.o: in function `x11_str2flags':
/home/ali/slurm/slurm-git/src/common/x11_util.c:66: multiple definition of `x11_str2flags'; mgr.o:/home/ali/slurm/slurm-git/src/slurmd/slurmstepd/../../../src/common/x11_util.c:66: first defined here
/usr/bin/ld: ../../../src/api/libslurm.o: in function `x11_delete_xauth':
/home/ali/slurm/slurm-git/src/common/x11_util.c:275: multiple definition of `x11_delete_xauth'; mgr.o:/home/ali/slurm/slurm-git/src/slurmd/slurmstepd/../../../src/common/x11_util.c:275: first defined here
/usr/bin/ld: ../../../src/api/libslurm.o: in function `x11_flags2str':
/home/ali/slurm/slurm-git/src/common/x11_util.c:82: multiple definition of `x11_flags2str'; mgr.o:/home/ali/slurm/slurm-git/src/slurmd/slurmstepd/../../../src/common/x11_util.c:82: first defined here
/usr/bin/ld: ../../../src/api/libslurm.o: in function `x11_get_display':
/home/ali/slurm/slurm-git/src/common/x11_util.c:103: multiple definition of `x11_get_display'; mgr.o:/home/ali/slurm/slurm-git/src/slurmd/slurmstepd/../../../src/common/x11_util.c:103: first defined here
/usr/bin/ld: ../../../src/api/libslurm.o: in function `x11_get_xauth':
/home/ali/slurm/slurm-git/src/common/x11_util.c:156: multiple definition of `x11_get_xauth'; mgr.o:/home/ali/slurm/slurm-git/src/slurmd/slurmstepd/../../../src/common/x11_util.c:156: first defined here
/usr/bin/ld: ../../../src/api/libslurm.o: in function `x11_set_xauth':
/home/ali/slurm/slurm-git/src/common/x11_util.c:219: multiple definition of `x11_set_xauth'; mgr.o:/home/ali/slurm/slurm-git/src/slurmd/slurmstepd/../../../src/common/x11_util.c:219: first defined here
collect2: error: ld returned 1 exit status
make[4]: *** [Makefile:622: slurmstepd] Error 1
make[4]: Leaving directory '/home/ali/slurm/slurm-git/src/slurmd/slurmstepd'
make[3]: *** [Makefile:515: all-recursive] Error 1
make[3]: Leaving directory '/home/ali/slurm/slurm-git/src/slurmd'
make[2]: *** [Makefile:545: all-recursive] Error 1
make[2]: Leaving directory '/home/ali/slurm/slurm-git/src'
make[1]: *** [Makefile:624: all-recursive] Error 1
make[1]: Leaving directory '/home/ali/slurm/slurm-git'
make: *** [Makefile:523: all] Error 2


I built from the slurm-22.05 git branch just to double-check myself and it fails without the patch, but succeeds with it. However, the build failure is different in the git branch than 22.05.2:

/bin/sh ../../libtool  --tag=CC   --mode=link gcc  -DNUMA_VERSION1_COMPATIBILITY -g -O2 -fno-omit-frame-pointer -pthread -ggdb3 -Wall -g -O1 -fno-strict-aliasing -export-dynamic   -o slurmrestd http.o operations.o slurmrestd.o rest_auth.o ../../src/api/libslurm.o -ldl -L/usr/lib64 -lhttp_parser libslurmrest_ref.la -lpthread -lm -lresolv 
libtool: link: gcc -DNUMA_VERSION1_COMPATIBILITY -g -O2 -fno-omit-frame-pointer -pthread -ggdb3 -Wall -g -O1 -fno-strict-aliasing -o slurmrestd http.o operations.o slurmrestd.o rest_auth.o ../../src/api/libslurm.o -Wl,--export-dynamic  -ldl -L/usr/lib64 -lhttp_parser ./.libs/libslurmrest_ref.a -lpthread -lm -lresolv -pthread
/usr/bin/ld: ../../src/api/libslurm.o: in function `parse_host_port':
/home/ali/slurm/slurm-git/src/common/http.c:308: multiple definition of `parse_host_port'; http.o:/home/ali/slurm/slurm-git/src/slurmrestd/http.c:803: first defined here
/usr/bin/ld: ../../src/api/libslurm.o: in function `free_parse_host_port':
/home/ali/slurm/slurm-git/src/common/http.c:314: multiple definition of `free_parse_host_port'; http.o:/home/ali/slurm/slurm-git/src/slurmrestd/http.c:856: first defined here
collect2: error: ld returned 1 exit status
make[4]: *** [Makefile:642: slurmrestd] Error 1
make[4]: Leaving directory '/home/ali/slurm/slurm-git/src/slurmrestd'
make[3]: *** [Makefile:695: all-recursive] Error 1
make[3]: Leaving directory '/home/ali/slurm/slurm-git/src/slurmrestd'
make[2]: *** [Makefile:545: all-recursive] Error 1
make[2]: Leaving directory '/home/ali/slurm/slurm-git/src'
make[1]: *** [Makefile:624: all-recursive] Error 1
make[1]: Leaving directory '/home/ali/slurm/slurm-git'
make: *** [Makefile:523: all] Error 2


So it appears the x11 errors in 22.05.2 are fixed in the slurm-22.05 branch, but the http errors were introduced (which your patch fixes).

(In reply to Ali Nikkhah from comment #14)
> The patch does not resolve the issue for 22.05.2. I modified the patch
> slightly so it would apply cleanly (just removed the NEWS modification) to
> 22.05.2.

Please attach the config.log from this run.

> I built from the slurm-22.05 git branch just to double-check myself and it
> fails without the patch, but succeeds with it. However, the build failure is
> different in the git branch than 22.05.2:

Patch is not upstream yet. These errors look like what I saw. Does the slurm-22.05 branch with the patch in comment#13 work?

Created attachment 26073 [details]
config.log 22.05.2 patched

Here is config.log for the failed 22.05.2 patched build.

(In reply to Nate Rini from comment #15)
> Patch is not upstream yet. These errors look like what I saw. Does the
> slurm-22.05 branch with the patch in comment#13 work?

Yes, the slurm-22.05 branch builds with the patch.

(In reply to Ali Nikkhah from comment #17)
> (In reply to Nate Rini from comment #15)
> > Patch is not upstream yet. These errors look like what I saw. Does the
> > slurm-22.05 branch with the patch in comment#13 work?
> 
> Yes, the slurm-22.05 branch builds with the patch.

Great, there is no reason to debug why 22.05.2 doesn't build as the issue has already been resolved. We will follow the normal QA process. With any luck, the patch will get included with 22.05.3.

Sounds good, thanks for your help! I found the commit that fixes the 22.05.2 build error in the git history.

(In reply to Ali Nikkhah from comment #20)
> Sounds good, thanks for your help! I found the commit that fixes the 22.05.2
> build error in the git history.

This is now fixed upstream for the pending slurm-22.05.4 release:
> https://github.com/SchedMD/slurm/commit/e6f28f51b8da60a1a5ce86fcc6bf8275bbf732d9