[slurm-users] SLURMDBD fails trying to talk to MariaDB - Help debugging configuration

Aravindh Sampathkumar aravindh at fastmail.com
Thu Oct 11 14:58:57 MDT 2018


I'm trying to setup a SLURM cluster in a virtual environment before
actually deploying it for serious work. I hit a snag where Slurmdbd
fails soon after starting because of trouble connecting to MariaDB.
SlurmDBD service status:
[root at slmaster ~]# systemctl status slurmdbd

*●* slurmdbd.service - Slurm DBD accounting daemon

   Loaded: loaded (/etc/systemd/system/slurmdbd.service; enabled; vendor
   preset: disabled)   Active: *failed* (Result: timeout) since Thu 2018-10-11 20:34:42
   UTC; 14min ago  Process: 1406 ExecStart=/usr/sbin/slurmdbd $SLURMDBD_OPTIONS
  (code=exited, status=0/SUCCESS)

Oct 11 20:33:11 slmaster systemd[1]: Starting Slurm DBD
accounting daemon...Oct 11 20:33:11 slmaster systemd[1]: PID file /var/run/slurmdbd.pid not
readable (yet?) after start.Oct 11 20:34:42 slmaster systemd[1]: *slurmdbd.service start operation
timed out. Terminating.*Oct 11 20:34:42 slmaster systemd[1]: *Failed to start Slurm DBD
accounting daemon.*Oct 11 20:34:42 slmaster systemd[1]: *Unit slurmdbd.service entered
failed state.*Oct 11 20:34:42 slmaster systemd[1]: *slurmdbd.service failed.*

MariaDB running just fine:
[root at slmaster ~]# systemctl status mariadb

*●* mariadb.service - MariaDB database server

   Loaded: loaded (/usr/lib/systemd/system/mariadb.service; enabled;
   vendor preset: disabled)   Active: *active (running)* since Thu 2018-10-11 20:33:11 UTC; 18min
   ago  Process: 991 ExecStartPost=/usr/libexec/mariadb-wait-ready $MAINPID
  (code=exited, status=0/SUCCESS)  Process: 943 ExecStartPre=/usr/libexec/mariadb-prepare-db-dir %n
  (code=exited, status=0/SUCCESS) Main PID: 989 (mysqld_safe)

   CGroup: /system.slice/mariadb.service

           ├─ 989 /bin/sh /usr/bin/mysqld_safe --basedir=/usr

           └─1265 /usr/libexec/mysqld --basedir=/usr --
           datadir=/var/lib/mysql --plugin-dir=/usr/lib64/mysql/plugin

Oct 11 20:33:07 slmaster systemd[1]: Starting MariaDB database server...Oct 11 20:33:07 slmaster mariadb-prepare-db-dir[943]: Database MariaDB
is probably initialized in /var/lib/mysql already, nothing is done.Oct 11 20:33:08 slmaster mysqld_safe[989]: 181011 20:33:08 mysqld_safe
Logging to '/var/log/mariadb/mariadb.log'.Oct 11 20:33:09 slmaster mysqld_safe[989]: 181011 20:33:09 mysqld_safe
Starting mysqld daemon with databases from /var/lib/mysqlOct 11 20:33:11 slmaster systemd[1]: Started MariaDB database server.
Logfile from Slurmdbd:

[2018-10-11T20:33:11.648] debug:  Log file re-opened

[2018-10-11T20:33:11.720] debug:  Munge authentication plugin loaded

[2018-10-11T20:33:11.749] debug2: mysql_connect() called for db
slurm_acct_db[2018-10-11T20:33:11.785] debug2: innodb_buffer_pool_size: 629145600

[2018-10-11T20:33:11.785] debug2: innodb_log_file_size: 67108864

[2018-10-11T20:33:11.786] debug2: innodb_lock_wait_timeout: 900

[2018-10-11T20:33:11.956] Accounting storage MYSQL plugin loaded

[2018-10-11T20:33:11.958] debug2: ArchiveDir        = /tmp

[2018-10-11T20:33:11.958] debug2: ArchiveScript     = (null)

[2018-10-11T20:33:11.958] debug2: AuthInfo          = (null)

[2018-10-11T20:33:11.958] debug2: AuthType          = auth/munge

[2018-10-11T20:33:11.958] debug2: CommitDelay       = 0

[2018-10-11T20:33:11.958] debug2: DbdAddr           = slmaster

[2018-10-11T20:33:11.958] debug2: DbdBackupHost     = (null)

[2018-10-11T20:33:11.958] debug2: DbdHost           = slmaster

[2018-10-11T20:33:11.958] debug2: DbdPort           = 6819

[2018-10-11T20:33:11.958] debug2: DebugFlags        = (null)

[2018-10-11T20:33:11.958] debug2: DebugLevel        = 6

[2018-10-11T20:33:11.958] debug2: DebugLevelSyslog  = 10

[2018-10-11T20:33:11.958] debug2: DefaultQOS        = (null)

[2018-10-11T20:33:11.958] debug2: LogFile           =
/var/log/slurm/slurmdbd.log[2018-10-11T20:33:11.958] debug2: MessageTimeout    = 10

[2018-10-11T20:33:11.958] debug2: Parameters        = (null)

[2018-10-11T20:33:11.958] debug2: PidFile           =
/var/spool/slurm/slurmdbd.pid[2018-10-11T20:33:11.958] debug2: PluginDir         = /usr/lib64/slurm[2018-10-11T20:33:11.958] debug2: PrivateData       = none

[2018-10-11T20:33:11.958] debug2: PurgeEventAfter   = NONE
[2018-10-11T20:33:11.958] debug2: PurgeJobAfter     = NONE

[2018-10-11T20:33:11.958] debug2: PurgeResvAfter    = NONE

[2018-10-11T20:33:11.958] debug2: PurgeStepAfter    = NONE

[2018-10-11T20:33:11.958] debug2: PurgeSuspendAfter = NONE

[2018-10-11T20:33:11.958] debug2: PurgeTXNAfter = NONE

[2018-10-11T20:33:11.958] debug2: PurgeUsageAfter = NONE

[2018-10-11T20:33:11.958] debug2: SlurmUser         = slurm(982)

[2018-10-11T20:33:11.958] debug2: StorageBackupHost = (null)

[2018-10-11T20:33:11.958] debug2: StorageHost       = localhost

[2018-10-11T20:33:11.958] debug2: StorageLoc        = slurm_acct_db

[2018-10-11T20:33:11.958] debug2: StoragePort       = 3306

[2018-10-11T20:33:11.958] debug2: StorageType       =
accounting_storage/mysql[2018-10-11T20:33:11.958] debug2: StorageUser       = slurm

[2018-10-11T20:33:11.958] debug2: TCPTimeout        = 2

[2018-10-11T20:33:11.958] debug2: TrackWCKey        = 0

[2018-10-11T20:33:11.958] debug2: TrackSlurmctldDown= 0

[2018-10-11T20:33:11.958] debug2: acct_storage_p_get_connection: request
new connection 1[2018-10-11T20:33:11.974] slurmdbd version 18.08.1 started

[2018-10-11T20:33:11.986] debug2: running rollup at Thu Oct 11
20:33:11 2018[2018-10-11T20:33:11.986] debug2: Everything rolled up

[2018-10-11T20:34:42.968] Terminate signal (SIGINT or SIGTERM) received[2018-10-11T20:34:42.969] debug:  rpc_mgr shutting down

sacctmgr says:
[root at slmaster ~]# sacctmgr -vvvv

sacctmgr: Accounting storage SLURMDBD plugin loaded

sacctmgr: debug2: slurm_connect failed: Connection refused

sacctmgr: debug2: Error connecting slurm stream socket at Connection refusedsacctmgr: error: slurm_persist_conn_open_without_init: failed to open persistent connection to slmaster:6819: Connection refusedsacctmgr: error: slurmdbd: Sending PersistInit msg: Connection refusedsacctmgr: error: Problem talking to the database: Connection refused

I am able to connect to MariaDB locally using the above settings...

[root at slmaster ~]# mysql -u slurm -h localhost -P 3306 -p

Enter password: 

*Welcome to the MariaDB monitor.  Commands end with ; or \g.*

*Your MariaDB connection id is 7*

*Server version: 5.5.60-MariaDB MariaDB Server*

*Copyright (c) 2000, 2018, Oracle, MariaDB Corporation Ab and others.*

*Type 'help;' or '\h' for help. Type '\c' to clear the current input

MariaDB [(none)]> 

Config files are all attached. slurm.conf, slurmdbd.conf

Mariadb configuration was not changed..



# Disabling symbolic-links is recommended to prevent assorted
# security riskssymbolic-links=0

# Settings user and group are ignored when systemd is used.

# If you need to run mysqld under a different user or group,

# customize your systemd unit file for mariadb according to the

# instructions in http://fedoraproject.org/wiki/Systemd





# include all files from the config directory


!includedir /etc/my.cnf.d

Appreciate any help troubleshooting the "Connection refused" error.. 

  Aravindh Sampathkumar
  aravindh at fastmail.com

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.schedmd.com/pipermail/slurm-users/attachments/20181011/b54742c8/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: slurm.conf
Type: application/octet-stream
Size: 3339 bytes
Desc: not available
URL: <http://lists.schedmd.com/pipermail/slurm-users/attachments/20181011/b54742c8/attachment-0002.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: slurmdbd.conf
Type: application/octet-stream
Size: 742 bytes
Desc: not available
URL: <http://lists.schedmd.com/pipermail/slurm-users/attachments/20181011/b54742c8/attachment-0003.obj>

More information about the slurm-users mailing list