[slurm-users] Fwd: srun: error: Unable to allocate resources: Invalid partition name specified

[Brian Andrus]

You show you still have more that one partition with Default=YES.

There should one and only one that is set to YES.
That is the one partition that is used if it is not specified.

Brian Andrus


On 7/27/2018 6:34 AM, valeriana at cbpf.br wrote:
> Hi Merlin
>
>> Do you accidentally have more than one partition with Default=YES?
> It was. I changed to NO and I continue with the same error.
>
> [root at master ~]# scontrol show partition
> PartitionName=course
>    AllowGroups=courseit AllowAccounts=ALL AllowQos=ALL
>    AllocNodes=ALL Default=NO QoS=N/A
>    DefaultTime=NONE DisableRootJobs=NO ExclusiveUser=NO GraceTime=0 
> Hidden=NO
>    MaxNodes=UNLIMITED MaxTime=UNLIMITED MinNodes=1 LLN=NO 
> MaxCPUsPerNode=UNLIMITED
>    Nodes=node[02-04,06,09-12]
>    PriorityJobFactor=1 PriorityTier=1 RootOnly=NO ReqResv=NO 
> OverSubscribe=NO
>    OverTimeLimit=NONE PreemptMode=OFF
>    State=UP TotalCPUs=64 TotalNodes=8 SelectTypeParameters=NONE
>    DefMemPerNode=UNLIMITED MaxMemPerNode=UNLIMITED
>
> PartitionName=test
>    AllowGroups=testcluster AllowAccounts=ALL AllowQos=ALL
>    AllocNodes=ALL Default=YES QoS=N/A
>    DefaultTime=NONE DisableRootJobs=NO ExclusiveUser=NO GraceTime=0 
> Hidden=NO
>    MaxNodes=UNLIMITED MaxTime=UNLIMITED MinNodes=1 LLN=NO 
> MaxCPUsPerNode=UNLIMITED
>    Nodes=node[01,05,07,08]
>    PriorityJobFactor=1 PriorityTier=1 RootOnly=NO ReqResv=NO 
> OverSubscribe=NO
>    OverTimeLimit=NONE PreemptMode=OFF
>    State=UP TotalCPUs=32 TotalNodes=4 SelectTypeParameters=NONE
>    DefMemPerNode=UNLIMITED MaxMemPerNode=UNLIMITED
>
> /etc/slurm/slurm.conf
> PartitionName=course Nodes=node[02-04,06,09-12] AllowGroups=curseit 
> Default=NO MaxTime=INFINITE State=UP
> PartitionName=test Nodes=node[01,05,07,08] AllowGroups=testcluster 
> Default=YES MaxTime=INFINITE State=UP
>
> If I have a lot of partitions, how can I set a default partition to a 
> distinct groups?
>
> In my slurm.conf file, I think that I have to set Default=YES to all 
> main partition to my all distinct partitions
>
> For example:
>
> /etc/slurm/slurm.conf
>
> PartitionName=course Nodes=node[02-04,06,09-12] AllowGroups=curseit 
> Default=YES MaxTime=INFINITE State=UP
> PartitionName=courset Nodes=node[13-20] AllowGroups=curseit Default=NO 
> MaxTime=INFINITE State=UP
>
> PartitionName=test Nodes=node[01,05,07,08] AllowGroups=testcluster 
> Default=YES MaxTime=INFINITE State=UP
> PartitionName=testc Nodes=node[21-30] AllowGroups=testcluster 
> Default=NO MaxTime=INFINITE State=UP
>
> Thanks!!!
>
> Valeriana
>
> Citando Merlin Hartley <merlin at mrc-mbu.cam.ac.uk>:
>
>> Do you accidentally have more than one partition with Default=YES?
>>
>>
>> -- 
>> Merlin Hartley
>> Computer Officer
>> MRC Mitochondrial Biology Unit
>> University of Cambridge
>> Cambridge, CB2 0XY
>> United Kingdom
>>
>>> On 26 Jul 2018, at 16:57, valeriana at cbpf.br wrote:
>>>
>>> Hi all,
>>>
>>> I dont´t understand why its occurs!
>>>
>>> user: john
>>> group: courseit
>>> partition: course
>>>
>>> [john at master ~]$ sinfo
>>> PARTITION AVAIL  TIMELIMIT  NODES  STATE NODELIST
>>> course        up   infinite      8   idle node[02-04,06,09-12]
>>>
>>> /etc/group
>>> courseit:x:1002:john
>>>
>>> /etc/passwd
>>> john:x:1001:1002::/home/john:/bin/bash
>>>
>>> /etc/slurm/slurm.conf
>>> PartitionName=course Nodes=node[02-04,06,09-12] AllowGroups=courseit 
>>> Default=YES MaxTime=INFINITE State=UP
>>>
>>>
>>> [john at master ~]$ srun -N3 -l /bin/hostname
>>> srun: error: Unable to allocate resources: User's group not 
>>> permitted to use this partition
>>>
>>> And if I put -p course, it´s ok
>>>
>>> [john at master ~]$ srun -p course -N3 -l /bin/hostname
>>> 2: node04
>>> 1: node03
>>> 0: node02
>>>
>>> Can someone has an idea?
>>>
>>> Thanks in advance!
>>> Valeriana
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>
> ----- Final da mensagem encaminhada -----
>
>