Uploaded image for project: 'Marathon'
  1. Marathon
  2. MARATHON-8110

TaskGroupBuilder must use only given matched resources

    Details

      Description

      TaskGroupBuilder uses matched resources only for host ports, whereas it must be doing the same for all kinds of resources, including cpu, mem and so on. Otherwise every once in a while Marathon is unable to launch a pod with PVs with the following error:

      Feb 24 16:48:11 ip-10-0-6-241.us-west-2.compute.internal mesos-master[5082]: I0224 16:48:11.690165  5119 master.cpp:7975] Sending status update TASK_ERROR for
       task persistent-volume-pod-bbfcbaffaafa4391bf0da0122fcf8a9f.instance-7f4bff69-1982-11e8-92ee-d62eb7f2d9e5.container1.1 of framework 3badfb23-8a81-4007-9eae-a
      fc2f903250f-0001 'Total resources cpus(allocated: slave_public):0.3; mem(allocated: slave_public):288; disk(allocated: slave_public):10; ports(allocated: slav
      e_public)(reservations: [(DYNAMIC,slave_public,dcos_marathon,{marathon_framework_id: 3badfb23-8a81-4007-9eae-afc2f903250f-0001, marathon_task_id: persistent-v
      olume-pod-bbfcbaffaafa4391bf0da0122fcf8a9f.instance-7f4bff69-1982-11e8-92ee-d62eb7f2d9e5})]):[21251-21251, 21250-21250]; disk(allocated: slave_public)(reserva
      tions: [(DYNAMIC,slave_public,dcos_marathon,{marathon_framework_id: 3badfb23-8a81-4007-9eae-afc2f903250f-0001, marathon_task_id: persistent-volume-pod-bbfcbaf
      faafa4391bf0da0122fcf8a9f.instance-7f4bff69-1982-11e8-92ee-d62eb7f2d9e5})])[persistent-volume-pod-bbfcbaffaafa4391bf0da0122fcf8a9f#pst#7f4bff68-1982-11e8-92ee
      -d62eb7f2d9e5:pst]:10 required by task group and its executor are more than available disk(allocated: slave_public)(reservations: [(DYNAMIC,slave_public,dcos_
      marathon,{marathon_framework_id: 3badfb23-8a81-4007-9eae-afc2f903250f-0001, marathon_task_id: persistent-volume-pod-bbfcbaffaafa4391bf0da0122fcf8a9f.instance-
      7f4bff69-1982-11e8-92ee-d62eb7f2d9e5})])[persistent-volume-pod-bbfcbaffaafa4391bf0da0122fcf8a9f#pst#7f4bff68-1982-11e8-92ee-d62eb7f2d9e5:pst]:10; ports(alloca
      ted: slave_public)(reservations: [(DYNAMIC,slave_public,dcos_marathon,{marathon_framework_id: 3badfb23-8a81-4007-9eae-afc2f903250f-0001, marathon_task_id: per
      sistent-volume-pod-bbfcbaffaafa4391bf0da0122fcf8a9f.instance-7f4bff69-1982-11e8-92ee-d62eb7f2d9e5})]):[21250-21251]; disk(allocated: slave_public)(reservation
      s: [(DYNAMIC,slave_public,dcos_marathon,{marathon_framework_id: 3badfb23-8a81-4007-9eae-afc2f903250f-0001, marathon_task_id: persistent-volume-pod-bbfcbaffaaf
      a4391bf0da0122fcf8a9f.instance-7f4bff69-1982-11e8-92ee-d62eb7f2d9e5})])
      Feb 24 16:48:11 ip-10-0-6-241.us-west-2.compute.internal mesos-master[5082]: :10; mem(allocated: slave_public)(reservations: [(DYNAMIC,slave_public,dcos_marat
      hon,{marathon_framework_id: 3badfb23-8a81-4007-9eae-afc2f903250f-0001, marathon_task_id: persistent-volume-pod-bbfcbaffaafa4391bf0da0122fcf8a9f.instance-7f4bf
      f69-1982-11e8-92ee-d62eb7f2d9e5})]):288; cpus(allocated: slave_public)(reservations: [(DYNAMIC,slave_public,dcos_marathon,{marathon_framework_id: 3badfb23-8a8
      1-4007-9eae-afc2f903250f-0001, marathon_task_id: persistent-volume-pod-bbfcbaffaafa4391bf0da0122fcf8a9f.instance-7f4bff69-1982-11e8-92ee-d62eb7f2d9e5})]):0.3'
      Feb 24 16:48:11 ip-10-0-6-241.us-west-2.compute.internal mesos-master[5082]: I0224 16:48:11.690263  5119 master.cpp:7975] Sending status update TASK_ERROR for task persistent-volume-pod-bbfcbaffaafa4391bf0da0122fcf8a9f.instance-7f4bff69-1982-11e8-92ee-d62eb7f2d9e5.container2.1 of framework 3badfb23-8a81-4007-9eae-afc2f903250f-0001 'Total resources cpus(allocated: slave_public):0.3; mem(allocated: slave_public):288; disk(allocated: slave_public):10; ports(allocated: slave_public)(reservations: [(DYNAMIC,slave_public,dcos_marathon,{marathon_framework_id: 3badfb23-8a81-4007-9eae-afc2f903250f-0001, marathon_task_id: persistent-volume-pod-bbfcbaffaafa4391bf0da0122fcf8a9f.instance-7f4bff69-1982-11e8-92ee-d62eb7f2d9e5})]):[21251-21251, 21250-21250]; disk(allocated: slave_public)(reservations: [(DYNAMIC,slave_public,dcos_marathon,{marathon_framework_id: 3badfb23-8a81-4007-9eae-afc2f903250f-0001, marathon_task_id: persistent-volume-pod-bbfcbaffaafa4391bf0da0122fcf8a9f.instance-7f4bff69-1982-11e8-92ee-d62eb7f2d9e5})])[persistent-volume-pod-bbfcbaffaafa4391bf0da0122fcf8a9f#pst#7f4bff68-1982-11e8-92ee-d62eb7f2d9e5:pst]:10 required by task group and its executor are more than available disk(allocated: slave_public)(reservations: [(DYNAMIC,slave_public,dcos_marathon,{marathon_framework_id: 3badfb23-8a81-4007-9eae-afc2f903250f-0001, marathon_task_id: persistent-volume-pod-bbfcbaffaafa4391bf0da0122fcf8a9f.instance-7f4bff69-1982-11e8-92ee-d62eb7f2d9e5})])[persistent-volume-pod-bbfcbaffaafa4391bf0da0122fcf8a9f#pst#7f4bff68-1982-11e8-92ee-d62eb7f2d9e5:pst]:10; ports(allocated: slave_public)(reservations: [(DYNAMIC,slave_public,dcos_marathon,{marathon_framework_id: 3badfb23-8a81-4007-9eae-afc2f903250f-0001, marathon_task_id: persistent-volume-pod-bbfcbaffaafa4391bf0da0122fcf8a9f.instance-7f4bff69-1982-11e8-92ee-d62eb7f2d9e5})]):[21250-21251]; disk(allocated: slave_public)(reservations: [(DYNAMIC,slave_public,dcos_marathon,{marathon_framework_id: 3badfb23-8a81-4007-9eae-afc2f903250f-0001, marathon_task_id: persistent-volume-pod-bbfcbaffaafa4391bf0da0122fcf8a9f.instance-7f4bff69-1982-11e8-92ee-d62eb7f2d9e5})])
      Feb 24 16:48:11 ip-10-0-6-241.us-west-2.compute.internal mesos-master[5082]: :10; mem(allocated: slave_public)(reservations: [(DYNAMIC,slave_public,dcos_marathon,{marathon_framework_id: 3badfb23-8a81-4007-9eae-afc2f903250f-0001, marathon_task_id: persistent-volume-pod-bbfcbaffaafa4391bf0da0122fcf8a9f.instance-7f4bff69-1982-11e8-92ee-d62eb7f2d9e5})]):288; cpus(allocated: slave_public)(reservations: [(DYNAMIC,slave_public,dcos_marathon,{marathon_framework_id: 3badfb23-8a81-4007-9eae-afc2f903250f-0001, marathon_task_id: persistent-volume-pod-bbfcbaffaafa4391bf0da0122fcf8a9f.instance-7f4bff69-1982-11e8-92ee-d62eb7f2d9e5})]):0.3'
      

        Attachments

          Activity

            People

            • Assignee:
              ivanchernetsky Ivan Chernetsky
              Reporter:
              ivanchernetsky Ivan Chernetsky
              Team:
              Orchestration Team
              Watchers:
              Ivan Chernetsky
            • Watchers:
              1 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved: