Recommended update for pdsh, pmix, slurm 20.02

Announcement ID: SUSE-RU-2020:1397-1
Rating: moderate
References:
Affected Products:
  • HPC Module 15-SP1
  • SUSE Linux Enterprise High Performance Computing 15 SP1

An update that has two fixes can now be installed.

Description:

This update for pdsh fixes the following issues:

The slurm version 20.02 was shipped by this update.

  • Improve job state reason for jobs hitting partition_job_depth.
  • Speed up testing of singleton dependencies.
  • Fix negative loop bound in cons_tres.
  • srun - capture the MPI plugin return code from mpi_hook_client_fini() and use as final return code for step failure.
  • Fix segfault in cli_filter/lua.
  • Fix --gpu-bind=map_gpu reusability if tasks > elements.
  • Make sure config_flags on a gres are sent to the slurmctld on node registration.
  • Prolog/Epilog - Fix missing GPU information.
  • Fix segfault when using config parser for expanded lines.
  • Fix bit overlap test function.
  • Don't accrue time if job begin time is in the future.
  • Remove accrue time when updating a job start/eligible time to the future.
  • Fix regression in 20.02.0 that broke --depend=expand.
  • Reset begin time on job release if it's not in the future.
  • Fix for recovering burst buffers when using high-availability.
  • Fix invalid read due to freeing an incorrectly allocated env array.
  • Update slurmctld -i message to warn about losing data.
  • Fix scontrol cancel_reboot so it clears the DRAIN flag and node reason for a pending ASAP reboot.

Update to version pdsh-2.34 (2020-01-07) (bsc#1167704)

  • Fix for output corruption with no newlines
  • pipecmd: fix result check error handling
  • slurm: workaround slurm export of internal List interfaces
  • readline: add application name as argv[0]

  • Fix errors from lgtm.com scan

  • Change 'typedef List' to 'typedef xList'.

Since SLURM exposes a type List in its public API without proper namespacing we need to change this type in pdsh. To remain compatible with older versions of SLURM whose type List was identical to pdsh's we need to rename not only the struct.

pmix 3.1.5 was shipped as dependency of slurm 20.02. (bsc#1170613)

  • pmix is a process management interface for MPI.

Patch Instructions:

To install this SUSE update use the SUSE recommended installation methods like YaST online_update or "zypper patch".
Alternatively you can run the command listed for your product:

  • HPC Module 15-SP1
    zypper in -t patch SUSE-SLE-Module-HPC-15-SP1-2020-1397=1

Package List:

  • HPC Module 15-SP1 (aarch64 x86_64)
    • pdsh-netgroup-2.34-10.5.7
    • slurm_20_02-auth-none-20.02.1-3.5.3
    • pdsh-dshgroup-2.34-10.5.7
    • libmca_common_dstore1-3.1.5-1.3.4
    • slurm_20_02-plugins-debuginfo-20.02.1-3.5.3
    • pmix-devel-3.1.5-1.3.4
    • pdsh-debugsource-2.34-10.5.7
    • pdsh-dshgroup-debuginfo-2.34-10.5.7
    • pmix-debugsource-3.1.5-1.3.4
    • slurm_20_02-config-20.02.1-3.5.3
    • perl-slurm_20_02-20.02.1-3.5.3
    • pmix-plugins-debuginfo-3.1.5-1.3.4
    • pmix-plugins-3.1.5-1.3.4
    • perl-slurm_20_02-debuginfo-20.02.1-3.5.3
    • slurm_20_02-auth-none-debuginfo-20.02.1-3.5.3
    • slurm_20_02-slurmdbd-debuginfo-20.02.1-3.5.3
    • libslurm35-20.02.1-3.5.3
    • slurm_20_02-webdoc-20.02.1-3.5.3
    • slurm_20_02-sql-20.02.1-3.5.3
    • pdsh-debuginfo-2.34-10.5.7
    • pdsh-slurm_20_02-debuginfo-2.34-10.5.3
    • slurm_20_02-debugsource-20.02.1-3.5.3
    • slurm_20_02-sview-20.02.1-3.5.3
    • pdsh-slurm_18_08-2.34-10.5.7
    • slurm_20_02-node-20.02.1-3.5.3
    • pdsh-machines-debuginfo-2.34-10.5.7
    • slurm_20_02-doc-20.02.1-3.5.3
    • slurm_20_02-devel-20.02.1-3.5.3
    • slurm_20_02-slurmdbd-20.02.1-3.5.3
    • slurm_20_02-torque-debuginfo-20.02.1-3.5.3
    • pdsh-netgroup-debuginfo-2.34-10.5.7
    • slurm_20_02-munge-20.02.1-3.5.3
    • slurm_20_02-sview-debuginfo-20.02.1-3.5.3
    • pdsh-slurm-2.34-10.5.7
    • slurm_20_02-lua-debuginfo-20.02.1-3.5.3
    • libslurm35-debuginfo-20.02.1-3.5.3
    • pdsh-slurm-debuginfo-2.34-10.5.7
    • slurm_20_02-node-debuginfo-20.02.1-3.5.3
    • libnss_slurm2-20.02.1-3.5.3
    • pmix-3.1.5-1.3.4
    • slurm_20_02-pam_slurm-20.02.1-3.5.3
    • slurm_20_02-plugins-20.02.1-3.5.3
    • slurm_20_02-torque-20.02.1-3.5.3
    • libpmi0_20_02-debuginfo-20.02.1-3.5.3
    • slurm_20_02-sql-debuginfo-20.02.1-3.5.3
    • slurm_20_02-20.02.1-3.5.3
    • slurm_20_02-lua-20.02.1-3.5.3
    • pdsh-2.34-10.5.7
    • pdsh-machines-2.34-10.5.7
    • libnss_slurm2-debuginfo-20.02.1-3.5.3
    • libmca_common_dstore1-debuginfo-3.1.5-1.3.4
    • libpmi0_20_02-20.02.1-3.5.3
    • slurm_20_02-munge-debuginfo-20.02.1-3.5.3
    • slurm_20_02-pam_slurm-debuginfo-20.02.1-3.5.3
    • slurm_20_02-config-man-20.02.1-3.5.3
    • pdsh-slurm_18_08-debuginfo-2.34-10.5.7
    • pdsh-slurm_20_02-2.34-10.5.3
    • libpmix2-3.1.5-1.3.4
    • pmix-debuginfo-3.1.5-1.3.4
    • libpmix2-debuginfo-3.1.5-1.3.4
    • pdsh-genders-debuginfo-2.34-10.5.7
    • slurm_20_02-debuginfo-20.02.1-3.5.3
    • pdsh-genders-2.34-10.5.7
  • HPC Module 15-SP1 (noarch)
    • pmix-mca-params-3.1.5-1.3.4
    • pmix-headers-3.1.5-1.3.4

References: