Felix,

Can you ping the nodes from the controller and vise versa?

The snippet below looks like a potential firewall issue:

[2015-03-16T15:40:02.845] debug2: Error connecting slurm stream socket at
***.***.***.***52:6818: Connection timed out

Try telnet'ing from the controller to each node on port 6818 and then
telnet'ing from each node to the controller on port 6817.

John DeSantis


2015-03-17 11:23 GMT-04:00 Yann Sagon <[email protected]>:

>
>
> 2015-03-17 13:31 GMT+01:00 Felix Willenborg <
> [email protected]>:
>
>>
>> Hi there,
>>
>> first of all, i'm kinda new to slurm, so hopefully i may have missed
>> something very basic here.
>>
>>
>> slurmctld.log
>> ------------------------------------------------------------
>> ------------------------------------------------------------
>> ------------------------------------
>> [2015-03-16T15:39:54.813] debug:  sched: slurmctld starting
>> [2015-03-16T15:39:54.817] error: Configured MailProg is invalid
>> [2015-03-16T15:39:54.817] debug3: Trying to load plugin
>> /usr/lib/slurm/accounting_storage_filetxt.so
>> [2015-03-16T15:39:54.817] debug2: slurmdb_init() called
>> [2015-03-16T15:39:54.817] Accounting storage FileTxt plugin loaded
>> [2015-03-16T15:39:54.818] debug3: Success.
>> [2015-03-16T15:39:54.818] debug3: not enforcing associations and no list
>> was given so we are giving a blank list
>> [2015-03-16T15:39:54.818] debug3: Version in assoc_mgr_state header is 1
>> [2015-03-16T15:39:54.818] slurmctld version 2.6.5 started on cluster
>> cluster
>>
>
> As you are new to slurm, I would as first step suggest to try with latest
> slurm version. (14.11.4)
>
>
>

Reply via email to