[slurm-users] When I start slurmctld, there are some errors in log.

[UGI]

When I start slurmctld, there are some errors in log. And the job running
information doesn't store to mysql via slurmdbd.

I set

AccountingStoragePass=/usr/local/munge-munge-0.5.13/var/run/munge/munge.socket.2

AccountingStorageType=accounting_storage/slurmdbd

JobAcctGatherType=jobacct_gather/linux

in slurm.conf.


The following message is the log which slurmctld output.

[2018-06-15T11:05:44.763] Terminate signal (SIGINT or SIGTERM) received

[2018-06-15T11:05:44.807] Saving all slurm state

[2018-06-15T11:05:45.101] error: slurmdbd: Sending fini msg: No error

[2018-06-15T11:05:45.126] layouts: all layouts are now unloaded.

[2018-06-15T11:06:07.761] slurmctld version 17.11.7 started on cluster
myslurm

[2018-06-15T11:06:07.785] error: slurm_persist_conn_open_without_init:
failed to open persistent connection to localhost:6819: Connection refused

[2018-06-15T11:06:07.785] error: slurmdbd: Sending PersistInit msg:
Connection refused

[2018-06-15T11:06:07.785] error: slurmdbd: Sending PersistInit msg:
Connection refused

[2018-06-15T11:06:07.787] layouts: no layout to initialize

[2018-06-15T11:06:07.824] error:
################################################

[2018-06-15T11:06:07.824] error: ###       SEVERE SECURITY VULERABILTY
  ###

[2018-06-15T11:06:07.824] error: ### StateSaveLocation DIRECTORY IS WORLD
WRITABLE ###

[2018-06-15T11:06:07.824] error: ###         CORRECT FILE PERMISSIONS
  ###

[2018-06-15T11:06:07.824] error:
################################################

[2018-06-15T11:06:07.824] layouts: loading entities/relations information

[2018-06-15T11:06:07.824] Recovered state of 1 nodes

[2018-06-15T11:06:07.824] Recovered JobID=12 State=0x3 NodeCnt=0 Assoc=2

[2018-06-15T11:06:07.825] Recovered information about 1 jobs

[2018-06-15T11:06:07.825] cons_res: select_p_node_init

[2018-06-15T11:06:07.825] cons_res: preparing for 1 partitions

[2018-06-15T11:06:07.825] Recovered state of 0 reservations

[2018-06-15T11:06:07.825] _preserve_plugins: backup_controller not specified

[2018-06-15T11:06:07.825] cons_res: select_p_reconfigure

[2018-06-15T11:06:07.825] cons_res: select_p_node_init

[2018-06-15T11:06:07.825] cons_res: preparing for 1 partitions

[2018-06-15T11:06:07.825] Running as primary controller

[2018-06-15T11:06:07.825] Registering slurmctld at port 6817 with slurmdbd.

[2018-06-15T11:06:07.825] error: slurmdbd: Sending PersistInit msg:
Connection refused

[2018-06-15T11:06:07.825] error: slurmdbd: Sending PersistInit msg:
Connection refused

[2018-06-15T11:06:07.826] No parameter for mcs plugin, default values set

[2018-06-15T11:06:07.826] mcs: MCSParameters = (null). ondemand set.

[2018-06-15T11:06:10.829]
SchedulerParameters=default_queue_depth=100,max_rpc_cnt=0,max_sched_time=2,partition_job_depth=0,sched_max_job_start=0,sched_min_interval=2
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.schedmd.com/pipermail/slurm-users/attachments/20180615/11125db0/attachment.html>