Slurm in Docker issueshttps://codebase.helmholtz.cloud/fwcc/slurm-in-docker/-/issues2023-02-03T08:26:57+01:00https://codebase.helmholtz.cloud/fwcc/slurm-in-docker/-/issues/1sudo: setrlimit(RLIMIT_CORE): Operation not permitted2023-02-03T08:26:57+01:00David Paped.pape@hzdr.desudo: setrlimit(RLIMIT_CORE): Operation not permittedProbably caused by [this bug](https://bugzilla.redhat.com/show_bug.cgi?id=1773148).Probably caused by [this bug](https://bugzilla.redhat.com/show_bug.cgi?id=1773148).https://codebase.helmholtz.cloud/fwcc/slurm-in-docker/-/issues/2Set up Slurm DataBase Daemon2020-01-08T10:41:41+01:00David Paped.pape@hzdr.deSet up Slurm DataBase Daemon... to be able to run tests using accounting information.... to be able to run tests using accounting information.https://codebase.helmholtz.cloud/fwcc/slurm-in-docker/-/issues/3running slurm as docker2019-12-18T10:02:55+01:00Steinbach, Dr. rer. nat. Peter (FWCC) - 120595running slurm as dockerJust saw this repo and thought, maybe the following might help:
https://github.com/giovtorres/docker-centos7-slurm
this repo set's up several docker containers with docker-swarm and connects them as if they would be compute nodes of a cl...Just saw this repo and thought, maybe the following might help:
https://github.com/giovtorres/docker-centos7-slurm
this repo set's up several docker containers with docker-swarm and connects them as if they would be compute nodes of a cluster. It then set's up slurm to send jobs to each node.
If I am missing the point completely, feel free to close this issue. ;)https://codebase.helmholtz.cloud/fwcc/slurm-in-docker/-/issues/4ERROR 1396 (HY000) at line 1: Operation CREATE USER failed for 'slurm'@'local...2023-02-03T08:27:32+01:00David Paped.pape@hzdr.deERROR 1396 (HY000) at line 1: Operation CREATE USER failed for 'slurm'@'localhost'When using the image in a GitLab CI job, this error is printed _after_ the job has successfully finished.
Executing `mysql -uslurm -ppassword < script.sql` with script
```sql
use slurm_acct_db;
show tables;
```
works and yields
```
T...When using the image in a GitLab CI job, this error is printed _after_ the job has successfully finished.
Executing `mysql -uslurm -ppassword < script.sql` with script
```sql
use slurm_acct_db;
show tables;
```
works and yields
```
Tables_in_slurm_acct_db
acct_coord_table
acct_table
clus_res_table
cluster_table
convert_version_table
federation_table
qos_table
res_table
sid_assoc_table
sid_assoc_usage_day_table
sid_assoc_usage_hour_table
sid_assoc_usage_month_table
sid_event_table
sid_job_table
sid_last_ran_table
sid_resv_table
sid_step_table
sid_suspend_table
sid_usage_day_table
sid_usage_hour_table
sid_usage_month_table
sid_wckey_table
sid_wckey_usage_day_table
sid_wckey_usage_hour_table
sid_wckey_usage_month_table
table_defs_table
tres_table
txn_table
user_table
```
so everything should be alright.https://codebase.helmholtz.cloud/fwcc/slurm-in-docker/-/issues/5Always run container as janedoe?2023-03-09T15:38:41+01:00David Paped.pape@hzdr.deAlways run container as janedoe?At the moment, the default command logs the user in as _janedoe_. Since GitLab CI passes a script to the container, the default command is overridden, resulting in the job running as root.
This can be useful when trying out admin tasks ...At the moment, the default command logs the user in as _janedoe_. Since GitLab CI passes a script to the container, the default command is overridden, resulting in the job running as root.
This can be useful when trying out admin tasks but is not really necessary. Maybe this behaviour could be changed in the future.