diff options
Diffstat (limited to 'topics/systems/hpc')
| -rw-r--r-- | topics/systems/hpc/octopus-maintenance.gmi | 69 | ||||
| -rw-r--r-- | topics/systems/hpc/performance.gmi | 4 |
2 files changed, 49 insertions, 24 deletions
diff --git a/topics/systems/hpc/octopus-maintenance.gmi b/topics/systems/hpc/octopus-maintenance.gmi index a0a2f16..d034575 100644 --- a/topics/systems/hpc/octopus-maintenance.gmi +++ b/topics/systems/hpc/octopus-maintenance.gmi @@ -2,10 +2,23 @@ ## Slurm -Status of slurm +Status of slurm (as of 202512) ``` sinfo +workers* up infinite 8 idle octopus[03,05-11] +allnodes up infinite 3 alloc tux[06,08-09] +allnodes up infinite 11 idle octopus[02-03,05-11],tux[05,07] +tux up infinite 3 alloc tux[06,08-09] +tux up infinite 2 idle tux[05,07] +1tbmem up infinite 1 idle octopus02 +headnode up infinite 1 idle octopus01 +highmem up infinite 2 idle octopus[02,11] +386mem up infinite 6 idle octopus[03,06-10] +lowmem up infinite 7 idle octopus[03,05-10] +``` + +``` sinfo -R squeue ``` @@ -29,7 +42,7 @@ UnkillableStepProgram = (null) UnkillableStepTimeout = 60 sec ``` -check valid configuration with `slurmd -C` and update nodes with +check valid configuration with 'slurmd -C' and update nodes with ``` scontrol reconfigure @@ -45,13 +58,13 @@ Basically the root user can copy across. ## Execute binaries on mounted devices -To avoid `./scratch/script.sh: Permission denied` on `device_file`: +To avoid './scratch/script.sh: Permission denied' on 'device_file': -- `sudo bash` -- `ls /scratch -l` to check where `/scratch` is -- `vim /etc/fstab` -- replace `noexec` with `exec` for `device_file` -- `mount -o remount [device_file]` to remount the partition with its new configuration. +- 'sudo bash' +- 'ls /scratch -l' to check where '/scratch' is +- 'vim /etc/fstab' +- replace 'noexec' with 'exec' for 'device_file' +- 'mount -o remount [device_file]' to remount the partition with its new configuration. Some notes: @@ -67,7 +80,7 @@ x-systemd.device-timeout= 10.0.0.110:/export/3T /mnt/3T nfs nofail,x-systemd.automount,x-systemd.requires=network-online.target,x-systemd.device-timeout=10 0 0 -## Installation of `munge` and `slurm` on a new node +## Installation of 'munge' and 'slurm' on a new node Current nodes in the pool have: @@ -78,7 +91,7 @@ sbatch --version slurm-wlm 18.08.5-2 ``` -To install `munge`, go to `octopus01` and run: +To install 'munge', go to 'octopus01' and run: ```shell guix package -i munge@0.5.14 -p /export/octopus01/guix-profiles/slurm @@ -86,7 +99,7 @@ guix package -i munge@0.5.14 -p /export/octopus01/guix-profiles/slurm systemctl status munge # to check if the service is running and where its service file is ``` -We need to setup the rights for `munge`: +We need to setup the rights for 'munge': ```shell sudo bash @@ -100,7 +113,7 @@ mkdir -p /var/lib/munge chown munge:munge /var/lib/munge/ mkdir -p /etc/munge -# copy `munge.key` (from a working node) to `/etc/munge/munge.key` +# copy 'munge.key' (from a working node) to '/etc/munge/munge.key' chown -R munge:munge /etc/munge mkdir -p /run/munge @@ -112,7 +125,7 @@ chown munge:munge /var/log/munge mkdir -p /var/run/munge # todo: not sure why it needs such a folder chown munge:munge /var/run/munge -# copy `munge.service` (from a working node) to `/etc/systemd/system/munge.service` +# copy 'munge.service' (from a working node) to '/etc/systemd/system/munge.service' systemctl daemon-reload systemctl enable munge @@ -120,25 +133,25 @@ systemctl start munge systemctl status munge ``` -To test the new installation, go to `octopus01` and then: +To test the new installation, go to 'octopus01' and then: ```shell munge -n | ssh tux08 /export/octopus01/guix-profiles/slurm-2-link/bin/unmunge ``` -If you get `STATUS: Rewound credential (16)`, it means that there is a difference between the encoding and decoding times. To fix it, go into the new machine and fix the time with +If you get 'STATUS: Rewound credential (16)', it means that there is a difference between the encoding and decoding times. To fix it, go into the new machine and fix the time with ```shell sudo date MMDDhhmmYYYY.ss ``` -To install `slurm`, go to `octopus01` and run: +To install 'slurm', go to 'octopus01' and run: ```shell guix package -i slurm@18.08.9 -p /export/octopus01/guix-profiles/slurm ``` -We need to setup the rights for `slurm`: +We need to setup the rights for 'slurm': ```shell sudo bash @@ -152,8 +165,8 @@ mkdir -p /var/lib/slurm chown munge:munge /var/lib/slurm/ mkdir -p /etc/slurm -# copy `slurm.conf` to `/etc/slurm/slurm.conf` -# copy `cgroup.conf` to `/etc/slurm/cgroup.conf` +# copy 'slurm.conf' to '/etc/slurm/slurm.conf' +# copy 'cgroup.conf' to '/etc/slurm/cgroup.conf' chown -R slurm:slurm /etc/slurm @@ -163,7 +176,7 @@ chown slurm:slurm /run/slurm mkdir -p /var/log/slurm chown slurm:slurm /var/log/slurm -# copy `slurm.service` to `/etc/systemd/system/slurm.service` +# copy 'slurm.service' to '/etc/systemd/system/slurm.service' /export/octopus01/guix-profiles/slurm-2-link/sbin/slurmd -f /etc/slurm/slurm.conf -C | head -n 1 >> /etc/slurm/slurm.conf # add node configuration information @@ -173,12 +186,24 @@ systemctl start slurm systemctl status slurm ``` -On `octopus01` (the master): +On 'octopus01' (the master): ```shell sudo bash -# add the new node to `/etc/slurm/slurm.conf` +# add the new node to '/etc/slurm/slurm.conf' systemctl restart slurmctld # after editing /etc/slurm/slurm.conf on the master ``` + + +# Removing a node + +We are removing o3 so it can become the new head node: + +``` +scontrol update nodename=octopus03 state=drain reason="removing" +scontrol show node octopus03 | grep State +scontrol update nodename=octopus03 state=down reason="removed" + State=DOWN+DRAIN ThreadsPerCore=2 TmpDisk=0 Weight=1 Owner=N/A MCS_label=N/A +``` diff --git a/topics/systems/hpc/performance.gmi b/topics/systems/hpc/performance.gmi index ac5e861..ee604b5 100644 --- a/topics/systems/hpc/performance.gmi +++ b/topics/systems/hpc/performance.gmi @@ -14,13 +14,13 @@ hdparm -Ttv /dev/sdc1 Cheap and cheerful: -Read test: +Write test: ``` dd if=/dev/zero of=./test bs=512k count=2048 oflag=direct ``` -Write test: +Read test: ``` /sbin/sysctl -w vm.drop_caches=3 |