[GE users] How to delete a queue that does/does not(?) exist

templedf dan.templeton at sun.com
Fri May 15 18:34:30 BST 2009


It won't show up.  It's effectively deleted.  The short version of the 
longer discussion is that all.q is a cluster queue that contains queue 
instances, e.g. all.q at corke010.  The cluster queue defines the 
attributes inherited by the queue instances, unless they specificly 
override some attributes.  The qmaster "knows" that there is a queue 
instance for every host in the cluster queue's hostlist, but beyond 
what's in the host list and/or the override of cluster queue attributes, 
queue instances don't exist in any concrete form.  Every time you modify 
the hostlist, you change the list of queue instances.  Removing a host 
from the host list removes the queue instance of that cluster queue on 
that host.

Daniel

jcd wrote:
> Dan-
> Thanks for the quick reply.
> However I can foresee users telling me about that queue (and other 
> queues as a matter of fact) as it shows up in the output like qstat -f
> Any simple way to force the deletion?
> Thanks again
> JC
>
>
> templedf wrote:
>   
>> Then the queue is effectively deleted.  There's a longer discussion we 
>> could have about how queues are actually represented by the master, but 
>> suffice it to say that as long as the host in not in the hostlist and 
>> there is no mention of it in another attribute, then the queue doesn't 
>> exist.
>>
>> Daniel
>>
>> jcd wrote:
>>     
>>> No mention of it:
>>> $ qconf -sq all.q
>>> qname                 all.q
>>> hostlist              @allhosts
>>> seq_no                0
>>> load_thresholds       np_load_avg=1.75
>>> suspend_thresholds    NONE
>>> nsuspend              1
>>> suspend_interval      00:05:00
>>> priority              0
>>> min_cpu_interval      00:05:00
>>> processors            UNDEFINED
>>> qtype                 BATCH
>>> ckpt_list             NONE
>>> pe_list               mpich1 smp
>>> rerun                 FALSE
>>> slots                 4
>>> tmpdir                /tmp
>>> shell                 /bin/csh
>>> prolog                NONE
>>> epilog                NONE
>>> shell_start_mode      unix_behavior
>>> starter_method        NONE
>>> suspend_method        NONE
>>> resume_method         NONE
>>> terminate_method      NONE
>>> notify                00:00:60
>>> owner_list            NONE
>>> user_lists            NONE
>>> xuser_lists           NONE
>>> subordinate_list      NONE
>>> complex_values        NONE
>>> projects              NONE
>>> xprojects             NONE
>>> calendar              NONE
>>> initial_state         default
>>> s_rt                  INFINITY
>>> h_rt                  INFINITY
>>> s_cpu                 INFINITY
>>> h_cpu                 INFINITY
>>> s_fsize               INFINITY
>>> h_fsize               INFINITY
>>> s_data                INFINITY
>>> h_data                INFINITY
>>> s_stack               INFINITY
>>> h_stack               INFINITY
>>> s_core                INFINITY
>>> h_core                INFINITY
>>> s_rss                 INFINITY
>>> h_rss                 INFINITY
>>> s_vmem                INFINITY
>>> h_vmem                INFINITY
>>>
>>>
>>> I deleted that hosts earlier from @allhosts
>>> JC
>>>
>>>
>>>
>>> templedf wrote:
>>>   
>>>       
>>>> You just want to delete it just on corke010?  The easiest thing to do is 
>>>> qconf -mq all.q and delete anything that says [corke010=...].
>>>>
>>>> Daniel
>>>>
>>>> jcd wrote:
>>>>     
>>>>         
>>>>> I'd like to delete the following queue
>>>>>
>>>>> # qconf -sq all.q at corke010
>>>>> qname                 all.q
>>>>> hostname              corke010
>>>>> seq_no                0
>>>>> load_thresholds       np_load_avg=1.75
>>>>> suspend_thresholds    NONE
>>>>> nsuspend              1
>>>>> suspend_interval      00:05:00
>>>>> priority              0
>>>>> min_cpu_interval      00:05:00
>>>>> processors            UNDEFINED
>>>>> qtype                 BATCH
>>>>> ckpt_list             NONE
>>>>> pe_list               mpich1 smp
>>>>> rerun                 FALSE
>>>>> slots                 4
>>>>> tmpdir                /tmp
>>>>> shell                 /bin/csh
>>>>> prolog                NONE
>>>>> epilog                NONE
>>>>> shell_start_mode      unix_behavior
>>>>> starter_method        NONE
>>>>> suspend_method        NONE
>>>>> resume_method         NONE
>>>>> terminate_method      NONE
>>>>> notify                00:00:60
>>>>> owner_list            NONE
>>>>> user_lists            NONE
>>>>> xuser_lists           NONE
>>>>> subordinate_list      NONE
>>>>> complex_values        NONE
>>>>> projects              NONE
>>>>> xprojects             NONE
>>>>> calendar              NONE
>>>>> initial_state         default
>>>>> s_rt                  INFINITY
>>>>> h_rt                  INFINITY
>>>>> s_cpu                 INFINITY
>>>>> h_cpu                 INFINITY
>>>>> s_fsize               INFINITY
>>>>> h_fsize               INFINITY
>>>>> s_data                INFINITY
>>>>> h_data                INFINITY
>>>>> s_stack               INFINITY
>>>>> h_stack               INFINITY
>>>>> s_core                INFINITY
>>>>> h_core                INFINITY
>>>>> s_rss                 INFINITY
>>>>> h_rss                 INFINITY
>>>>> s_vmem                INFINITY
>>>>> h_vmem                INFINITY
>>>>>
>>>>> However when I try to delete it I get the following error message:
>>>>> # qconf -dq all.q at corke010
>>>>> denied: cluster queue "all.q at corke010" does not exist
>>>>>
>>>>> What am I missing? So how do I delete it?
>>>>>
>>>>> Thanks
>>>>> JC
>>>>>
>>>>> ------------------------------------------------------
>>>>> http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=195623
>>>>>
>>>>> To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net].
>>>>>
>>>>>       
>>>>>           
>>>> ------------------------------------------------------
>>>> http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=195646
>>>>
>>>> To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net].
>>>>
>>>>     
>>>>         
>>> ------------------------------------------------------
>>> http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=195932
>>>
>>> To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net].
>>>
>>>       
>> ------------------------------------------------------
>> http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=195960
>>
>> To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net].
>>
>>     
>
> ------------------------------------------------------
> http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=195963
>
> To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net].
>

------------------------------------------------------
http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=195981

To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net].



More information about the gridengine-users mailing list