[slurm-dev] Re: Slurm configuration problem --Age factor not working @ all
Dear Yogendra, It seems you are missing the PriorityMaxAge parameter. Set it and the Age parameter should start working. Regards, Carles Fenoy On Mon, Nov 11, 2013 at 11:04 AM, wrote: > Hi Team > > > > > > We have enabled mutlifactor priority (PrirityWeightAge & PriorityJobSize) > but PrirityWeightAge is not working, our jobs are always scheduled based on > Jobsize, when we see sprio -l it always show job age = 0. Below are the > priorities which we have configured and we are using slurm 2.6.2, For > testing we are submitting sleep jobs i minutes. > > > > Please help me enabling Age factor also in our configuration. > > > > [root@hpca ~]# sprio -l > JOBID USER PRIORITY AGE FAIRSHAREJOBSIZE PARTITION > QOS NICE > 102 root968 0 0469 > 500 0 0 > 103 root968 0 0469 > 500 0 0 > 104 root968 0 0469 > 500 0 0 > > Below is the part of my slurm.conf > > > > # TIMERS > #BatchStartTimeout=10 > #CompleteWait=0 > #EpilogMsgTime=2000 > #GetEnvTimeout=2 > #HealthCheckInterval=0 > #HealthCheckProgram= > InactiveLimit=0 > KillWait=10 > #MessageTimeout=10 > #ResvOverRun=0 > MinJobAge=2 > #OverTimeLimit=5 > SlurmctldTimeout=120 > SlurmdTimeout=300 > #UnkillableStepTimeout=60 > #VSizeFactor=0 > Waittime=0 > # > # > # SCHEDULING > #DefMemPerCPU=0 > FastSchedule=1 > #MaxMemPerCPU=0 > #SchedulerRootFilter=1 > SchedulerTimeSlice=30 > PreemptMode=cancel > PreemptType=preempt/partition_prio > #PreemptType=preempt/qos > SchedulerType=sched/backfill > SchedulerPort=7321 > #SelectType=select/linear > SelectType=select/cons_res > SelectTypeParameters=CR_Core_Memory > #PreemptMode=GANG > # > # > # JOB PRIORITY > PriorityType=priority/multifactor > #PriorityDecayHalfLife=PriorityMaxAge=10 > #PriorityUsageResetPeriod= > PriorityWeightAge=10 > #PriorityWeightFairshare=1000 > PriorityWeightJobSize=1000 > PriorityWeightPartition=1000 > # > > -- > > Regards, > > Yogendra > #PriorityCalcPeriod= > PriorityFavorSmall=Yes > > * Please do not print this email unless it is absolutely necessary. * > > The information contained in this electronic message and any attachments > to this message are intended for the exclusive use of the addressee(s) and > may contain proprietary, confidential or privileged information. If you are > not the intended recipient, you should not disseminate, distribute or copy > this e-mail. Please notify the sender immediately and destroy all copies of > this message and any attachments. > > WARNING: Computer viruses can be transmitted via email. The recipient > should check this email and any attachments for the presence of viruses. > The company accepts no liability for any damage caused by any virus > transmitted by this email. > > www.wipro.com > -- -- Carles Fenoy
[slurm-dev] Re: Slurm configuration problem --Age factor not working @ all
Dear Carles, Please find attachement of slurm.conf. PriorityMaxAge=10 is already set. Pls let me know if I am missing any other thing. Regards, Yogendra From: Carlos Fenoy [mini...@gmail.com] Sent: Monday, November 11, 2013 3:52 PM To: slurm-dev Cc: Pankaj Sharma (WI01 - GIS) Subject: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all Dear Yogendra, It seems you are missing the PriorityMaxAge parameter. Set it and the Age parameter should start working. Regards, Carles Fenoy On Mon, Nov 11, 2013 at 11:04 AM, mailto:yogendra.shar...@wipro.com>> wrote: Hi Team We have enabled mutlifactor priority (PrirityWeightAge & PriorityJobSize) but PrirityWeightAge is not working, our jobs are always scheduled based on Jobsize, when we see sprio -l it always show job age = 0. Below are the priorities which we have configured and we are using slurm 2.6.2, For testing we are submitting sleep jobs i minutes. Please help me enabling Age factor also in our configuration. [root@hpca ~]# sprio -l JOBID USER PRIORITY AGE FAIRSHAREJOBSIZE PARTITIONQOS NICE 102 root968 0 0469500 0 0 103 root968 0 0469500 0 0 104 root968 0 0469500 0 0 Below is the part of my slurm.conf # TIMERS #BatchStartTimeout=10 #CompleteWait=0 #EpilogMsgTime=2000 #GetEnvTimeout=2 #HealthCheckInterval=0 #HealthCheckProgram= InactiveLimit=0 KillWait=10 #MessageTimeout=10 #ResvOverRun=0 MinJobAge=2 #OverTimeLimit=5 SlurmctldTimeout=120 SlurmdTimeout=300 #UnkillableStepTimeout=60 #VSizeFactor=0 Waittime=0 # # # SCHEDULING #DefMemPerCPU=0 FastSchedule=1 #MaxMemPerCPU=0 #SchedulerRootFilter=1 SchedulerTimeSlice=30 PreemptMode=cancel PreemptType=preempt/partition_prio #PreemptType=preempt/qos SchedulerType=sched/backfill SchedulerPort=7321 #SelectType=select/linear SelectType=select/cons_res SelectTypeParameters=CR_Core_Memory #PreemptMode=GANG # # # JOB PRIORITY PriorityType=priority/multifactor #PriorityDecayHalfLife=PriorityMaxAge=10 #PriorityUsageResetPeriod= PriorityWeightAge=10 #PriorityWeightFairshare=1000 PriorityWeightJobSize=1000 PriorityWeightPartition=1000 # -- Regards, Yogendra #PriorityCalcPeriod= PriorityFavorSmall=Yes Please do not print this email unless it is absolutely necessary. The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately and destroy all copies of this message and any attachments. WARNING: Computer viruses can be transmitted via email. The recipient should check this email and any attachments for the presence of viruses. The company accepts no liability for any damage caused by any virus transmitted by this email. www.wipro.com<http://www.wipro.com> -- -- Carles Fenoy Please do not print this email unless it is absolutely necessary. The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately and destroy all copies of this message and any attachments. WARNING: Computer viruses can be transmitted via email. The recipient should check this email and any attachments for the presence of viruses. The company accepts no liability for any damage caused by any virus transmitted by this email. www.wipro.com # slurm.conf file generated by configurator.html. # Put this file on all nodes of your cluster. # See the slurm.conf man page for more information. # ControlMachine=hpca #ControlAddr= #BackupController= #BackupAddr= # AuthType=auth/munge CacheGroups=0 #CheckpointType=checkpoint/none #CryptoType=crypto/openssl #DisableRootJobs=NO #EnforcePartLimits=NO #Epilog=/opt/epilog #EpilogSlurmctld=logout #FirstJobId=1 #MaxJobId=99 #GresTypes= #GroupUpdateForce=0 #GroupUpdateTime=600 #JobCheckpointDir=/var/slurm/checkpoint #JobCredentialPrivateKey= #JobCredentialPublicCertificate= #JobFileAppend=0 JobRequeue=1 #JobSubmitPlugins=1 #KillOnBadExit=0 #Licenses=foo*4,bar #MailProg=/bin/mail #MaxJobCount=5000 #MaxStepCount=4 #MaxTasksPerNode=128 MpiDefault=none #MpiParams=ports=#-# #PluginDir= #PlugStackConfig= #PrivateData=jobs ProctrackType=proctrack/pgid #Prolog= #PrologSlurmctld= #PropagatePrioProcess=0 #PropagateResourceLimits= #PropagateReso
[slurm-dev] Re: Slurm configuration problem --Age factor not working @ all
Hi All, Plz help me with below issue. Is this a bug in slurm-2.6.2 because of which age factor is not contributing to job priority even little. We have set up everything fine (we believe). -- Regards, Yogendra From: Yogendra Kumar Sharma (WI01 - GIS - ENU) Sent: Monday, November 11, 2013 4:35 PM To: slurm-dev Cc: Pankaj Sharma (WI01 - GIS); mini...@gmail.com Subject: RE: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all Dear Carles, Please find attachement of slurm.conf. PriorityMaxAge=10 is already set. Pls let me know if I am missing any other thing. Regards, Yogendra From: Carlos Fenoy [mini...@gmail.com] Sent: Monday, November 11, 2013 3:52 PM To: slurm-dev Cc: Pankaj Sharma (WI01 - GIS) Subject: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all Dear Yogendra, It seems you are missing the PriorityMaxAge parameter. Set it and the Age parameter should start working. Regards, Carles Fenoy On Mon, Nov 11, 2013 at 11:04 AM, mailto:yogendra.shar...@wipro.com>> wrote: Hi Team We have enabled mutlifactor priority (PrirityWeightAge & PriorityJobSize) but PrirityWeightAge is not working, our jobs are always scheduled based on Jobsize, when we see sprio -l it always show job age = 0. Below are the priorities which we have configured and we are using slurm 2.6.2, For testing we are submitting sleep jobs i minutes. Please help me enabling Age factor also in our configuration. [root@hpca ~]# sprio -l JOBID USER PRIORITY AGE FAIRSHAREJOBSIZE PARTITIONQOS NICE 102 root968 0 0469500 0 0 103 root968 0 0469500 0 0 104 root968 0 0469500 0 0 Below is the part of my slurm.conf # TIMERS #BatchStartTimeout=10 #CompleteWait=0 #EpilogMsgTime=2000 #GetEnvTimeout=2 #HealthCheckInterval=0 #HealthCheckProgram= InactiveLimit=0 KillWait=10 #MessageTimeout=10 #ResvOverRun=0 MinJobAge=2 #OverTimeLimit=5 SlurmctldTimeout=120 SlurmdTimeout=300 #UnkillableStepTimeout=60 #VSizeFactor=0 Waittime=0 # # # SCHEDULING #DefMemPerCPU=0 FastSchedule=1 #MaxMemPerCPU=0 #SchedulerRootFilter=1 SchedulerTimeSlice=30 PreemptMode=cancel PreemptType=preempt/partition_prio #PreemptType=preempt/qos SchedulerType=sched/backfill SchedulerPort=7321 #SelectType=select/linear SelectType=select/cons_res SelectTypeParameters=CR_Core_Memory #PreemptMode=GANG # # # JOB PRIORITY PriorityType=priority/multifactor #PriorityDecayHalfLife=PriorityMaxAge=10 #PriorityUsageResetPeriod= PriorityWeightAge=10 #PriorityWeightFairshare=1000 PriorityWeightJobSize=1000 PriorityWeightPartition=1000 # -- Regards, Yogendra #PriorityCalcPeriod= PriorityFavorSmall=Yes Please do not print this email unless it is absolutely necessary. The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately and destroy all copies of this message and any attachments. WARNING: Computer viruses can be transmitted via email. The recipient should check this email and any attachments for the presence of viruses. The company accepts no liability for any damage caused by any virus transmitted by this email. www.wipro.com<http://www.wipro.com> -- -- Carles Fenoy Please do not print this email unless it is absolutely necessary. The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately and destroy all copies of this message and any attachments. WARNING: Computer viruses can be transmitted via email. The recipient should check this email and any attachments for the presence of viruses. The company accepts no liability for any damage caused by any virus transmitted by this email. www.wipro.com # slurm.conf file generated by configurator.html. # Put this file on all nodes of your cluster. # See the slurm.conf man page for more information. # ControlMachine=hpca #ControlAddr= #BackupController= #BackupAddr= # AuthType=auth/munge CacheGroups=0 #CheckpointType=checkpoint/none #CryptoType=crypto/openssl #DisableRootJobs=NO #EnforcePartLimits=NO #Epilog=/opt/epilog #EpilogSlurmctld=logout #FirstJobId=1 #MaxJobId=99 #GresTypes= #GroupUpdateForce=0
[slurm-dev] Re: Slurm configuration problem --Age factor not working @ all
You should check that the parameters of the configuration file are being loaded. For example: scontrol show conf | grep Age | grep Priority PriorityMaxAge = 02:00:00 PriorityWeightAge = 1 and check that the values are the same as in the file. Regards. Juan Pancorbo Armada juan.panco...@lrz.de<mailto:juan.panco...@lrz.de> http//www.lrz.de Leibniz-Rechenzentrum Abteilung: Hochleistungssysteme Boltzmannstrasse 1, 85748 Garching Telefon: +49 (0) 89 35831-8735 Fax: +49 (0) 89 35831-8535 Von: yogendra.shar...@wipro.com [mailto:yogendra.shar...@wipro.com] Gesendet: Dienstag, 12. November 2013 07:25 An: slurm-dev Cc: mini...@gmail.com Betreff: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all Hi All, Plz help me with below issue. Is this a bug in slurm-2.6.2 because of which age factor is not contributing to job priority even little. We have set up everything fine (we believe). -- Regards, Yogendra From: Yogendra Kumar Sharma (WI01 - GIS - ENU) Sent: Monday, November 11, 2013 4:35 PM To: slurm-dev Cc: Pankaj Sharma (WI01 - GIS); mini...@gmail.com<mailto:mini...@gmail.com> Subject: RE: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all Dear Carles, Please find attachement of slurm.conf. PriorityMaxAge=10 is already set. Pls let me know if I am missing any other thing. Regards, Yogendra From: Carlos Fenoy [mini...@gmail.com] Sent: Monday, November 11, 2013 3:52 PM To: slurm-dev Cc: Pankaj Sharma (WI01 - GIS) Subject: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all Dear Yogendra, It seems you are missing the PriorityMaxAge parameter. Set it and the Age parameter should start working. Regards, Carles Fenoy On Mon, Nov 11, 2013 at 11:04 AM, mailto:yogendra.shar...@wipro.com>> wrote: Hi Team We have enabled mutlifactor priority (PrirityWeightAge & PriorityJobSize) but PrirityWeightAge is not working, our jobs are always scheduled based on Jobsize, when we see sprio -l it always show job age = 0. Below are the priorities which we have configured and we are using slurm 2.6.2, For testing we are submitting sleep jobs i minutes. Please help me enabling Age factor also in our configuration. [root@hpca ~]# sprio -l JOBID USER PRIORITY AGE FAIRSHAREJOBSIZE PARTITIONQOS NICE 102 root968 0 0469500 0 0 103 root968 0 0469500 0 0 104 root968 0 0469500 0 0 Below is the part of my slurm.conf # TIMERS #BatchStartTimeout=10 #CompleteWait=0 #EpilogMsgTime=2000 #GetEnvTimeout=2 #HealthCheckInterval=0 #HealthCheckProgram= InactiveLimit=0 KillWait=10 #MessageTimeout=10 #ResvOverRun=0 MinJobAge=2 #OverTimeLimit=5 SlurmctldTimeout=120 SlurmdTimeout=300 #UnkillableStepTimeout=60 #VSizeFactor=0 Waittime=0 # # # SCHEDULING #DefMemPerCPU=0 FastSchedule=1 #MaxMemPerCPU=0 #SchedulerRootFilter=1 SchedulerTimeSlice=30 PreemptMode=cancel PreemptType=preempt/partition_prio #PreemptType=preempt/qos SchedulerType=sched/backfill SchedulerPort=7321 #SelectType=select/linear SelectType=select/cons_res SelectTypeParameters=CR_Core_Memory #PreemptMode=GANG # # # JOB PRIORITY PriorityType=priority/multifactor #PriorityDecayHalfLife=PriorityMaxAge=10 #PriorityUsageResetPeriod= PriorityWeightAge=10 #PriorityWeightFairshare=1000 PriorityWeightJobSize=1000 PriorityWeightPartition=1000 # -- Regards, Yogendra #PriorityCalcPeriod= PriorityFavorSmall=Yes Please do not print this email unless it is absolutely necessary. The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately and destroy all copies of this message and any attachments. WARNING: Computer viruses can be transmitted via email. The recipient should check this email and any attachments for the presence of viruses. The company accepts no liability for any damage caused by any virus transmitted by this email. www.wipro.com<http://www.wipro.com> -- -- Carles Fenoy Please do not print this email unless it is absolutely necessary. The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you should not disseminate, distribute or copy this e-mail. Please notify the sender immedi
[slurm-dev] Re: Slurm configuration problem --Age factor not working @ all
There are two ways to confirm your PriorityWeightAge setting has been read in: $ scontrol show conf | grep PriorityWeightAge or $ sprio -w Once you confirm that the system has recognized the values you set in your slurm.conf file, if you still see the problem, I suggest you turn up debugging and set the following in your slurm.conf file: DebugFlags=Priority SlurmctldDebug=debug5 and see what your slurmctld.log tells you. Don From: yogendra.shar...@wipro.com [mailto:yogendra.shar...@wipro.com] Sent: Monday, November 11, 2013 10:25 PM To: slurm-dev Cc: mini...@gmail.com Subject: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all Hi All, Plz help me with below issue. Is this a bug in slurm-2.6.2 because of which age factor is not contributing to job priority even little. We have set up everything fine (we believe). -- Regards, Yogendra From: Yogendra Kumar Sharma (WI01 - GIS - ENU) Sent: Monday, November 11, 2013 4:35 PM To: slurm-dev Cc: Pankaj Sharma (WI01 - GIS); mini...@gmail.com Subject: RE: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all Dear Carles, Please find attachement of slurm.conf. PriorityMaxAge=10 is already set. Pls let me know if I am missing any other thing. Regards, Yogendra From: Carlos Fenoy [mini...@gmail.com] Sent: Monday, November 11, 2013 3:52 PM To: slurm-dev Cc: Pankaj Sharma (WI01 - GIS) Subject: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all Dear Yogendra, It seems you are missing the PriorityMaxAge parameter. Set it and the Age parameter should start working. Regards, Carles Fenoy On Mon, Nov 11, 2013 at 11:04 AM, mailto:yogendra.shar...@wipro.com>> wrote: Hi Team We have enabled mutlifactor priority (PrirityWeightAge & PriorityJobSize) but PrirityWeightAge is not working, our jobs are always scheduled based on Jobsize, when we see sprio -l it always show job age = 0. Below are the priorities which we have configured and we are using slurm 2.6.2, For testing we are submitting sleep jobs i minutes. Please help me enabling Age factor also in our configuration. [root@hpca ~]# sprio -l JOBID USER PRIORITY AGE FAIRSHAREJOBSIZE PARTITIONQOS NICE 102 root968 0 0469500 0 0 103 root968 0 0469500 0 0 104 root968 0 0469500 0 0 Below is the part of my slurm.conf # TIMERS #BatchStartTimeout=10 #CompleteWait=0 #EpilogMsgTime=2000 #GetEnvTimeout=2 #HealthCheckInterval=0 #HealthCheckProgram= InactiveLimit=0 KillWait=10 #MessageTimeout=10 #ResvOverRun=0 MinJobAge=2 #OverTimeLimit=5 SlurmctldTimeout=120 SlurmdTimeout=300 #UnkillableStepTimeout=60 #VSizeFactor=0 Waittime=0 # # # SCHEDULING #DefMemPerCPU=0 FastSchedule=1 #MaxMemPerCPU=0 #SchedulerRootFilter=1 SchedulerTimeSlice=30 PreemptMode=cancel PreemptType=preempt/partition_prio #PreemptType=preempt/qos SchedulerType=sched/backfill SchedulerPort=7321 #SelectType=select/linear SelectType=select/cons_res SelectTypeParameters=CR_Core_Memory #PreemptMode=GANG # # # JOB PRIORITY PriorityType=priority/multifactor #PriorityDecayHalfLife=PriorityMaxAge=10 #PriorityUsageResetPeriod= PriorityWeightAge=10 #PriorityWeightFairshare=1000 PriorityWeightJobSize=1000 PriorityWeightPartition=1000 # -- Regards, Yogendra #PriorityCalcPeriod= PriorityFavorSmall=Yes Please do not print this email unless it is absolutely necessary. The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately and destroy all copies of this message and any attachments. WARNING: Computer viruses can be transmitted via email. The recipient should check this email and any attachments for the presence of viruses. The company accepts no liability for any damage caused by any virus transmitted by this email. www.wipro.com<http://www.wipro.com> Error! Filename not specified. -- -- Carles Fenoy Please do not print this email unless it is absolutely necessary. The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately and destroy all copies of this message and any attachments. WARNING: Compute
[slurm-dev] Re: Slurm configuration problem --Age factor not working @ all
[root@hpca ~]# scontrol show config | grep PriorityWeightAge PriorityWeightAge = 1000 [root@hpca ~]# sprio -w JOBID PRIORITYAGEJOBSIZE Weights 1000 1000 [root@hpca ~]# sprio -l JOBID USER PRIORITYAGE FAIRSHAREJOBSIZE PARTITION QOS NICE 168 root468 0 0469 0 0 0 [root@hpca ~]# From: Pancorbo, Juan [mailto:juan.pancorbo.arm...@lrz.de] Sent: Tuesday, November 12, 2013 9:21 PM To: slurm-dev Subject: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all You should check that the parameters of the configuration file are being loaded. For example: scontrol show conf | grep Age | grep Priority PriorityMaxAge = 02:00:00 PriorityWeightAge = 1 and check that the values are the same as in the file. Regards. Juan Pancorbo Armada juan.panco...@lrz.de<mailto:juan.panco...@lrz.de> http//www.lrz.de Leibniz-Rechenzentrum Abteilung: Hochleistungssysteme Boltzmannstrasse 1, 85748 Garching Telefon: +49 (0) 89 35831-8735 Fax: +49 (0) 89 35831-8535 Von: yogendra.shar...@wipro.com<mailto:yogendra.shar...@wipro.com> [mailto:yogendra.shar...@wipro.com] Gesendet: Dienstag, 12. November 2013 07:25 An: slurm-dev Cc: mini...@gmail.com<mailto:mini...@gmail.com> Betreff: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all Hi All, Plz help me with below issue. Is this a bug in slurm-2.6.2 because of which age factor is not contributing to job priority even little. We have set up everything fine (we believe). -- Regards, Yogendra From: Yogendra Kumar Sharma (WI01 - GIS - ENU) Sent: Monday, November 11, 2013 4:35 PM To: slurm-dev Cc: Pankaj Sharma (WI01 - GIS); mini...@gmail.com<mailto:mini...@gmail.com> Subject: RE: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all Dear Carles, Please find attachement of slurm.conf. PriorityMaxAge=10 is already set. Pls let me know if I am missing any other thing. Regards, Yogendra From: Carlos Fenoy [mini...@gmail.com] Sent: Monday, November 11, 2013 3:52 PM To: slurm-dev Cc: Pankaj Sharma (WI01 - GIS) Subject: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all Dear Yogendra, It seems you are missing the PriorityMaxAge parameter. Set it and the Age parameter should start working. Regards, Carles Fenoy On Mon, Nov 11, 2013 at 11:04 AM, mailto:yogendra.shar...@wipro.com>> wrote: Hi Team We have enabled mutlifactor priority (PrirityWeightAge & PriorityJobSize) but PrirityWeightAge is not working, our jobs are always scheduled based on Jobsize, when we see sprio -l it always show job age = 0. Below are the priorities which we have configured and we are using slurm 2.6.2, For testing we are submitting sleep jobs i minutes. Please help me enabling Age factor also in our configuration. [root@hpca ~]# sprio -l JOBID USER PRIORITY AGE FAIRSHAREJOBSIZE PARTITIONQOS NICE 102 root968 0 0469500 0 0 103 root968 0 0469500 0 0 104 root968 0 0469500 0 0 Below is the part of my slurm.conf # TIMERS #BatchStartTimeout=10 #CompleteWait=0 #EpilogMsgTime=2000 #GetEnvTimeout=2 #HealthCheckInterval=0 #HealthCheckProgram= InactiveLimit=0 KillWait=10 #MessageTimeout=10 #ResvOverRun=0 MinJobAge=2 #OverTimeLimit=5 SlurmctldTimeout=120 SlurmdTimeout=300 #UnkillableStepTimeout=60 #VSizeFactor=0 Waittime=0 # # # SCHEDULING #DefMemPerCPU=0 FastSchedule=1 #MaxMemPerCPU=0 #SchedulerRootFilter=1 SchedulerTimeSlice=30 PreemptMode=cancel PreemptType=preempt/partition_prio #PreemptType=preempt/qos SchedulerType=sched/backfill SchedulerPort=7321 #SelectType=select/linear SelectType=select/cons_res SelectTypeParameters=CR_Core_Memory #PreemptMode=GANG # # # JOB PRIORITY PriorityType=priority/multifactor #PriorityDecayHalfLife=PriorityMaxAge=10 #PriorityUsageResetPeriod= PriorityWeightAge=10 #PriorityWeightFairshare=1000 PriorityWeightJobSize=1000 PriorityWeightPartition=1000 # -- Regards, Yogendra #PriorityCalcPeriod= PriorityFavorSmall=Yes Please do not print this email unless it is absolutely necessary. The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately and destroy all copies of this message and any attachments. WARNING: Computer viruses
[slurm-dev] Re: Slurm configuration problem --Age factor not working @ all
[root@hpca ~]# scontrol show config | grep PriorityWeightAge PriorityWeightAge = 1000 [root@hpca ~]# sprio -w JOBID PRIORITYAGEJOBSIZE Weights 1000 1000 [root@hpca ~]# sprio -l JOBID USER PRIORITYAGE FAIRSHAREJOBSIZE PARTITION QOS NICE 168 root468 0 0469 0 0 0 [root@hpca ~]# Don’t know what’s issue . Should We need to define age_factor also somewhere which seems to be 0. -- Regards, Yogendra From: Lipari, Don [mailto:lipa...@llnl.gov] Sent: Tuesday, November 12, 2013 9:24 PM To: slurm-dev Cc: mini...@gmail.com Subject: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all There are two ways to confirm your PriorityWeightAge setting has been read in: $ scontrol show conf | grep PriorityWeightAge or $ sprio -w Once you confirm that the system has recognized the values you set in your slurm.conf file, if you still see the problem, I suggest you turn up debugging and set the following in your slurm.conf file: DebugFlags=Priority SlurmctldDebug=debug5 and see what your slurmctld.log tells you. Don From: yogendra.shar...@wipro.com<mailto:yogendra.shar...@wipro.com> [mailto:yogendra.shar...@wipro.com] Sent: Monday, November 11, 2013 10:25 PM To: slurm-dev Cc: mini...@gmail.com<mailto:mini...@gmail.com> Subject: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all Hi All, Plz help me with below issue. Is this a bug in slurm-2.6.2 because of which age factor is not contributing to job priority even little. We have set up everything fine (we believe). -- Regards, Yogendra From: Yogendra Kumar Sharma (WI01 - GIS - ENU) Sent: Monday, November 11, 2013 4:35 PM To: slurm-dev Cc: Pankaj Sharma (WI01 - GIS); mini...@gmail.com<mailto:mini...@gmail.com> Subject: RE: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all Dear Carles, Please find attachement of slurm.conf. PriorityMaxAge=10 is already set. Pls let me know if I am missing any other thing. Regards, Yogendra From: Carlos Fenoy [mini...@gmail.com] Sent: Monday, November 11, 2013 3:52 PM To: slurm-dev Cc: Pankaj Sharma (WI01 - GIS) Subject: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all Dear Yogendra, It seems you are missing the PriorityMaxAge parameter. Set it and the Age parameter should start working. Regards, Carles Fenoy On Mon, Nov 11, 2013 at 11:04 AM, mailto:yogendra.shar...@wipro.com>> wrote: Hi Team We have enabled mutlifactor priority (PrirityWeightAge & PriorityJobSize) but PrirityWeightAge is not working, our jobs are always scheduled based on Jobsize, when we see sprio -l it always show job age = 0. Below are the priorities which we have configured and we are using slurm 2.6.2, For testing we are submitting sleep jobs i minutes. Please help me enabling Age factor also in our configuration. [root@hpca ~]# sprio -l JOBID USER PRIORITY AGE FAIRSHAREJOBSIZE PARTITIONQOS NICE 102 root968 0 0469500 0 0 103 root968 0 0469500 0 0 104 root968 0 0469500 0 0 Below is the part of my slurm.conf # TIMERS #BatchStartTimeout=10 #CompleteWait=0 #EpilogMsgTime=2000 #GetEnvTimeout=2 #HealthCheckInterval=0 #HealthCheckProgram= InactiveLimit=0 KillWait=10 #MessageTimeout=10 #ResvOverRun=0 MinJobAge=2 #OverTimeLimit=5 SlurmctldTimeout=120 SlurmdTimeout=300 #UnkillableStepTimeout=60 #VSizeFactor=0 Waittime=0 # # # SCHEDULING #DefMemPerCPU=0 FastSchedule=1 #MaxMemPerCPU=0 #SchedulerRootFilter=1 SchedulerTimeSlice=30 PreemptMode=cancel PreemptType=preempt/partition_prio #PreemptType=preempt/qos SchedulerType=sched/backfill SchedulerPort=7321 #SelectType=select/linear SelectType=select/cons_res SelectTypeParameters=CR_Core_Memory #PreemptMode=GANG # # # JOB PRIORITY PriorityType=priority/multifactor #PriorityDecayHalfLife=PriorityMaxAge=10 #PriorityUsageResetPeriod= PriorityWeightAge=10 #PriorityWeightFairshare=1000 PriorityWeightJobSize=1000 PriorityWeightPartition=1000 # -- Regards, Yogendra #PriorityCalcPeriod= PriorityFavorSmall=Yes Please do not print this email unless it is absolutely necessary. The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately and destroy all copies of this message an
[slurm-dev] Re: Slurm configuration problem --Age factor not working @ all
Dear Yogendra, writes: > [root@hpca ~]# scontrol show config | grep PriorityWeightAge > > PriorityWeightAge = 1000 > > [root@hpca ~]# sprio -w > > JOBID PRIORITY AGE JOBSIZE > > Weights 1000 1000 > > [root@hpca ~]# sprio -l > > JOBID USER PRIORITY AGE FAIRSHARE JOBSIZE PARTITION > QOS NICE > > 168 root 468 0 0 469 0 > > 0 0 > > [root@hpca ~]# It is usually better to ask a specific question rather than just stating the something is "not working at all". So I can only guess that you are expecting the value in the column 'AGE' to be non-zero. If a job has just been started, then the value for 'AGE' will be zero. As the job waits, the value will increase linearly to PriorityWeightAge over a period PriorityMaxAge. Regards Loris -- Dr. Loris Bennett (Mr.) ZEDAT, Freie Universität Berlin Email loris.benn...@fu-berlin.de
[slurm-dev] Re: Slurm configuration problem --Age factor not working @ all
Dear Loris, This is what I exactly want. My job waited for 1 day but its Age remains 0 . I want some nonzero value in age column. I suspect that multifactor for age factor is not working. but sprio -w shows that its set properly.Please help me with this. -- Regards, Yogendra From: Loris Bennett [loris.benn...@fu-berlin.de] Sent: Friday, November 15, 2013 2:13 PM To: slurm-dev Subject: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all Dear Yogendra, writes: > [root@hpca ~]# scontrol show config | grep PriorityWeightAge > > PriorityWeightAge = 1000 > > [root@hpca ~]# sprio -w > > JOBID PRIORITYAGEJOBSIZE > > Weights 1000 1000 > > [root@hpca ~]# sprio -l > > JOBID USER PRIORITYAGE FAIRSHAREJOBSIZE PARTITION > QOS NICE > > 168 root468 0 0469 0 > 0 0 > > [root@hpca ~]# It is usually better to ask a specific question rather than just stating the something is "not working at all". So I can only guess that you are expecting the value in the column 'AGE' to be non-zero. If a job has just been started, then the value for 'AGE' will be zero. As the job waits, the value will increase linearly to PriorityWeightAge over a period PriorityMaxAge. Regards Loris -- Dr. Loris Bennett (Mr.) ZEDAT, Freie Universität Berlin Email loris.benn...@fu-berlin.de Please do not print this email unless it is absolutely necessary. The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately and destroy all copies of this message and any attachments. WARNING: Computer viruses can be transmitted via email. The recipient should check this email and any attachments for the presence of viruses. The company accepts no liability for any damage caused by any virus transmitted by this email. www.wipro.com
[slurm-dev] Re: Slurm configuration problem --Age factor not working @ all
Hi All, I have configured Age_factor with multifactor but it’s not working . Do I need to configure QOS also to enable it ??? -- Regards, Yogendra From: Lipari, Don [mailto:lipa...@llnl.gov] Sent: Tuesday, November 12, 2013 9:24 PM To: slurm-dev Cc: mini...@gmail.com Subject: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all There are two ways to confirm your PriorityWeightAge setting has been read in: $ scontrol show conf | grep PriorityWeightAge or $ sprio -w Once you confirm that the system has recognized the values you set in your slurm.conf file, if you still see the problem, I suggest you turn up debugging and set the following in your slurm.conf file: DebugFlags=Priority SlurmctldDebug=debug5 and see what your slurmctld.log tells you. Don From: yogendra.shar...@wipro.com<mailto:yogendra.shar...@wipro.com> [mailto:yogendra.shar...@wipro.com] Sent: Monday, November 11, 2013 10:25 PM To: slurm-dev Cc: mini...@gmail.com<mailto:mini...@gmail.com> Subject: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all Hi All, Plz help me with below issue. Is this a bug in slurm-2.6.2 because of which age factor is not contributing to job priority even little. We have set up everything fine (we believe). -- Regards, Yogendra From: Yogendra Kumar Sharma (WI01 - GIS - ENU) Sent: Monday, November 11, 2013 4:35 PM To: slurm-dev Cc: Pankaj Sharma (WI01 - GIS); mini...@gmail.com<mailto:mini...@gmail.com> Subject: RE: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all Dear Carles, Please find attachement of slurm.conf. PriorityMaxAge=10 is already set. Pls let me know if I am missing any other thing. Regards, Yogendra From: Carlos Fenoy [mini...@gmail.com] Sent: Monday, November 11, 2013 3:52 PM To: slurm-dev Cc: Pankaj Sharma (WI01 - GIS) Subject: [slurm-dev] Re: Slurm configuration problem --Age factor not working @ all Dear Yogendra, It seems you are missing the PriorityMaxAge parameter. Set it and the Age parameter should start working. Regards, Carles Fenoy On Mon, Nov 11, 2013 at 11:04 AM, mailto:yogendra.shar...@wipro.com>> wrote: Hi Team We have enabled mutlifactor priority (PrirityWeightAge & PriorityJobSize) but PrirityWeightAge is not working, our jobs are always scheduled based on Jobsize, when we see sprio -l it always show job age = 0. Below are the priorities which we have configured and we are using slurm 2.6.2, For testing we are submitting sleep jobs i minutes. Please help me enabling Age factor also in our configuration. [root@hpca ~]# sprio -l JOBID USER PRIORITY AGE FAIRSHAREJOBSIZE PARTITIONQOS NICE 102 root968 0 0469500 0 0 103 root968 0 0469500 0 0 104 root968 0 0469500 0 0 Below is the part of my slurm.conf # TIMERS #BatchStartTimeout=10 #CompleteWait=0 #EpilogMsgTime=2000 #GetEnvTimeout=2 #HealthCheckInterval=0 #HealthCheckProgram= InactiveLimit=0 KillWait=10 #MessageTimeout=10 #ResvOverRun=0 MinJobAge=2 #OverTimeLimit=5 SlurmctldTimeout=120 SlurmdTimeout=300 #UnkillableStepTimeout=60 #VSizeFactor=0 Waittime=0 # # # SCHEDULING #DefMemPerCPU=0 FastSchedule=1 #MaxMemPerCPU=0 #SchedulerRootFilter=1 SchedulerTimeSlice=30 PreemptMode=cancel PreemptType=preempt/partition_prio #PreemptType=preempt/qos SchedulerType=sched/backfill SchedulerPort=7321 #SelectType=select/linear SelectType=select/cons_res SelectTypeParameters=CR_Core_Memory #PreemptMode=GANG # # # JOB PRIORITY PriorityType=priority/multifactor #PriorityDecayHalfLife=PriorityMaxAge=10 #PriorityUsageResetPeriod= PriorityWeightAge=10 #PriorityWeightFairshare=1000 PriorityWeightJobSize=1000 PriorityWeightPartition=1000 # -- Regards, Yogendra #PriorityCalcPeriod= PriorityFavorSmall=Yes Please do not print this email unless it is absolutely necessary. The information contained in this electronic message and any attachments to this message are intended for the exclusive use of the addressee(s) and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately and destroy all copies of this message and any attachments. WARNING: Computer viruses can be transmitted via email. The recipient should check this email and any attachments for the presence of viruses. The company accepts no liability for any damage caused by any virus transmitted by this email. www.wipro.com<http://www.wipro.com> Error! Filename not specified. -- -- Carles Fenoy Please do not print this email unles