Provide tests for SIG/HPC slurm packages

This MR provides a very, very basic test suite for the Slurm packages
built by the HPC SIG. It checks the following:

- Necessary packages for a single-node Slurm instance install
  successfully from the SIG/HPC repository
- A job can be scheduled and executed to completion
- A job can be scheduled and then cancelled

```
./fifloader.py --clean --load templates.fif.json
openqa-cli api -X POST isos ISO=Rocky-8.8-x86_64-dvd.iso ARCH=x86_64 DISTRI=rocky FLAVOR=dvd-iso VERSION=8.8 CURRREL=8 BUILD=-${date +%Y%d%m}.0-slurm-8.8 TEST=slurm22,slurm23
openqa-cli api -X POST isos ISO=Rocky-9.2-x86_64-dvd.iso ARCH=x86_64 DISTRI=rocky FLAVOR=dvd-iso VERSION=9.2 CURRREL=9 BUILD=-${date +%Y%d%m}.0-slurm-9.2 TEST=slurm22,slurm23

```

- [x] My code follows the style guidelines of this project
- [x] I have performed a self-review of my own code
- [x] I have commented my code, particularly in hard-to-understand areas
- [ ] I have made corresponding changes to the documentation
- [x] My changes generate no new warnings
- [x] Any dependent changes have been merged and published in downstream modules

templates.fif.json

@@ -1418,6 +1418,36 @@
                 "WORKER_CLASS": "tap"
             }
         },
+        "slurm22": {
+            "profiles": {
+                "rocky-dvd-iso-aarch64-*-aarch64": 10,
+                "rocky-dvd-iso-x86_64-*-64bit": 10
+            },
+            "settings": {
+                "BOOTFROM": "c",
+                "HDD_1": "disk_%FLAVOR%_%MACHINE%.qcow2",
+                "POSTINSTALL": "slurm",
+                "START_AFTER_TEST": "%DEPLOY_UPLOAD_TEST%",
+                "ROOT_PASSWORD": "weakpassword",
+                "USER_LOGIN": "false",
+                "SLURM_VERSION": "22"
+            }
+        },
+        "slurm23": {
+            "profiles": {
+                "rocky-dvd-iso-aarch64-*-aarch64": 10,
+                "rocky-dvd-iso-x86_64-*-64bit": 10
+            },
+            "settings": {
+                "BOOTFROM": "c",
+                "HDD_1": "disk_%FLAVOR%_%MACHINE%.qcow2",
+                "POSTINSTALL": "slurm",
+                "START_AFTER_TEST": "%DEPLOY_UPLOAD_TEST%",
+                "ROOT_PASSWORD": "weakpassword",
+                "USER_LOGIN": "false",
+                "SLURM_VERSION": "23"
+            }
+        },
         "support_server": {
             "profiles": {
                 "rocky-dvd-iso-aarch64-*-aarch64": 10,

tests/slurm.pm

@@ -0,0 +1,65 @@
+use base "installedtest";
+use strict;
+use testapi;
+use utils;
+
+sub slurm_setup {
+    # install HPC repository
+    my $version = get_var("SLURM_VERSION");
+    assert_script_run "dnf -y install rocky-release-hpc", 120;
+
+    # Set up munge
+    assert_script_run "dnf -y install munge", 120;
+    assert_script_run "dd if=/dev/urandom bs=1 count=1024 >/etc/munge/munge.key";
+    assert_script_run "chmod 400 /etc/munge/munge.key";
+    assert_script_run "chown munge.munge /etc/munge/munge.key";
+    assert_script_run "systemctl enable --now munge.service";
+
+    # install slurm
+    if (get_var("CURRREL") eq '8') {
+        assert_script_run "dnf config-manager --set-enabled powertools";
+    }
+    assert_script_run "dnf install -y slurm$version-slurmdbd slurm$version-slurmrestd slurm$version-slurmctld slurm$version-slurmd";
+
+    # Since this is a single node system, we don't have to modify the conf files. We will for larger multi-node tests.
+    # start services
+    assert_script_run "systemctl enable --now slurmctld slurmdbd slurmrestd slurmd";
+}
+
+sub run {
+    my $self = shift;
+
+    # do all the install stuff
+    slurm_setup();
+
+    # if everything is configured right, sinfo should show the following output
+    # $ sinfo
+    #   PARTITION AVAIL  TIMELIMIT  NODES  STATE NODELIST
+    #   debug*       up   infinite      1   idle localhost
+    validate_script_output "sinfo", sub { m/debug.*localhost/ };
+
+    # write a boring job script
+    assert_script_run "echo '#!/bin/bash\n#SBATCH --job-name=antarctica_time\nsleep 120\nTZ=NZST date' > job.sh";
+
+    ## schedule a job and run it to completion
+    assert_script_run "sbatch job.sh";
+    validate_script_output "squeue", sub { m/antar/ };
+    sleep 121;
+    # after 121 seconds, job should have completed and no longer exist in the queue
+    validate_script_output "squeue", sub { $_ !~ m/antar/ };
+
+    ## cancel a job
+    assert_script_run "sbatch job.sh";
+    validate_script_output "squeue", sub { m/antar/ };
+    assert_script_run "scancel 2";
+    # job should no longer be in the queue
+    validate_script_output "squeue", sub { $_ !~ m/antar/ };
+}
+
+sub test_flags {
+    return {fatal => 1};
+}
+
+1;
+
+# vim: set sw=4 et: