Checking host factory logs

Check the host factory logs to monitor cloud bursting in your cluster through scale-out and scale-in resource requests.

About this task

Monitoring cloud requests and hosts is easiest from the cluster management console. Especially for scale-out requests, you can easily filter your requestor instance, both from the Cloud Requests page (see Viewing cloud request instances) and the Cloud Hosts page (see Viewing cloud hosts). If you want detailed messages, you can always check the host factory logs.

• To monitor your cluster for scale-out requests to include cloud hosts:
  1. Look for SCALE-OUT messages in the hostfactory.hostname.log file at:
     • Installation_top\hostfactory\log\hostfactory.hostname.log on Windows.
     • $EGO_TOP/hostfactory/log/hostfactory.hostname.log on Linux®.
  2. Log in to your cloud service provider's console and look for newly provisioned hosts.
  3. Check whether the cluster now includes cloud hosts. For example:

     egosh resource list
     NAME                      status        mem    swp    tmp   ut    it    pg   r1m  r15s  r15m  ls
     ip-192-0-2-199.ec2.int*    ok           698M     0M    22G   2%     6   0.0   0.0   0.0   0.1   0
     ip-192-0-2-28.ec2.inte*    ok           698M     0M    22G   2%     6   0.0   0.1   0.3   0.1   0
     ip-192-0-2-49.ec2.inte*    ok           697M     0M    22G   2%     6   0.0   0.1   0.0   0.2   0
     ip-192-0-2-35.ec2.inte*    ok           698M     0M    22G   2%     6   0.0   0.0   0.0   0.1   0
     ip-192-0-2-151.ec2.int*    ok           693M     0M    22G   2%     6   0.0   0.2   0.0   0.1   0
     ip-192-0-2-77.ec2.inte*    ok           699M     0M    22G   4%     5   0.0   0.2   0.1   0.3   0
     ip-192-0-2-112.ec2.int*    ok           695M     0M    22G   2%     6   0.0   0.0   0.1   0.1   0
     ip-192-0-2-105.ec2.int*    ok           693M     0M    22G   3%     6   0.0   0.0   0.0   0.2   0
     doesym.softlayer.com       ok          9599M  2037M    10G   7%     0   0.0   0.1   0.3   0.1   1

  4. Check whether cloud hosts are running workload. For example, check details for the symping 7.3.2 application:

     soamview resource symping 7.3.2
     Application: symping7.3.2
     HOST                  SLOTS           RESGROUP            STATE       SLOTDEFNAME
     ip-192-0-2-105.ec2.*  1.00            ComputeHosts        OCCUPIED    -
     ip-192-0-2-112.ec2.*  1.00            ComputeHosts        OCCUPIED    -
     ip-192-0-2-151.ec2.*  1.00            ComputeHosts        OCCUPIED    -
     ip-192-0-2-199.ec2.*  1.00            ComputeHosts        OCCUPIED    -
     ip-192-0-2-28.ec2.i*  1.00            ComputeHosts        OCCUPIED    -
     ip-192-0-2-35.ec2.i*  1.00            ComputeHosts        OCCUPIED    -
     ip-192-0-2-49.ec2.i*  1.00            ComputeHosts        OCCUPIED    -
     ip-192-0-2-77.ec2.i*  1.00            ComputeHosts        OCCUPIED    -
     doesym.softlayer*     4.00            ComputeHosts        OCCUPIED    -

• To monitor your cluster for scale-in requests to release cloud hosts:
  1. Look for SCALE-IN messages in the hostfactory.hostname.log file at:
     • $EGO_TOP/hostfactory/log/hostfactory.hostname.log on Linux.
     • Installation_top\hostfactory\log\hostfactory.hostname.log on Windows.
  2. Check your cloud service provider's console to see whether cloud hosts are released.