Issue when trying to run a certain singularity image
Hi @BAQ0506 ,
this issue is not strictly related to the build script, but since it uses the environment, I thought here might still be a decent place to ask for help :)
I am currently trying to run a singularity image on our environment on the fat nodes which contains software to process LOFAR data for the next few months. The same image works on many other clusters as well as on the workstations we have at the observatory, however, on LGC2 I get the following error:
p1uy068@node34:~/data/container$ singularity run -B/home/p1uy068:/home/p1uy068 -B/localwork/henrik:/home/p1uy068/nodedata ~/node/lofar_sksp_fedora27_ddf_cpuinfofix.sif
INFO: Convert SIF file to sandbox...
FATAL: while extracting /home/p1uy068/node/lofar_sksp_fedora27_ddf_cpuinfofix.sif: root filesystem extraction failed: extract command failed:
write_xattr: could not write xattr security.selinux for file /tmp/p1uy068/singularity/rootfs-305171928/.exec because you're not superuser!
write_xattr: to avoid this error message, either specify -user-xattrs, -no-xattrs, or run as superuser!
Further error messages of this type are suppressed!
Parallel unsquashfs: Using 64 processors
108332 inodes (142015 blocks) to write
[======================================\ ] 95900/142015 67%
create_inode: socket /tmp/p1uy068/singularity/rootfs-305171928/var/cache/dnf/intel-mkl-repo-92b021b91e7ff902/pubring/S.gpg-agent ignored
create_inode: socket /tmp/p1uy068/singularity/rootfs-305171928/var/cache/dnf/intel-mkl-repo-92b021b91e7ff902/pubring/S.gpg-agent.browser ignored
create_inode: socket /tmp/p1uy068/singularity/rootfs-305171928/var/cache/dnf/intel-mkl-repo-92b021b91e7ff902/pubring/S.gpg-agent.extra ignored
create_inode: socket /tmp/p1uy068/singularity/rootfs-305171928/var/cache/dnf/intel-mkl-repo-92b021b91e7ff902/pubring/S.gpg-agent.ssh ignored
created 97114 files
created 10378 directories
created 4953 symlinks
created 0 devices
created 0 fifos
: exit status 1
A while a go another phd student had the same issue on lgc2, he could not solve the problem and migrated to an observatory workstation - this is not a viable solution here, since I have to process quite a lot of data. Also the author of the singularity image does not know what could cause the error or how to solve it.
Since it is working fine on other clusters, I wonder if it is specific to either some configuration of our cluster (filesystem?) or the singularity version. We are running 3.5.2, on 3.7 the image definitely works, for older versions I have to check.
Any idea how this could be solved or how I could further investigate the cause?
Cheers, -Henrik