Dual Server configuration showing different statuses for SLA with CA7 jobs

Document ID : KB000010178
Last Modified Date : 14/02/2018
Show Technical Document Details
Introduction:

With a primary and secondary server configured for iDash and connected to CA7 both servers are subscribed to JFM subscriptions coming from the iDash for CA7 interface on the mainframe.  While both servers should consistently receive the same updates from CA7 if something causes that subscription to be lost then one server may erroneously report an alarm on an SLA that the other does not see and could trigger other workflow for a customer based off of the SLA definition.  Below are some configurations that need to be set to ensure connectivity along with what to look for in the logs and common resolutions to issues.

Environment:
iDash 12.x.x on any platform with a connection to CA7 mainframe
Instructions:

Important configuration setting that is often missed: 

Each iDash server has its own idash.conf file in the respective $IDASH_HOME/etc or %IDASH_HOME%/etc folder.  Within this configuration file there is a parameter idash.jfm.subscription.target.  In a single server configuration this setting is typically not set and it is not set by the installer or any other automatic method when configuring a dual server setup.  For a dual server setup this parameter needs to be set on both primary and secondary to point to their respective hostname or IP address (Port is not required).  This parameter is passed to CA7 when we request subscriptions.

Not having this parameter set can cause the customer to show a valid JFM subscription but not receive job updates appropriately.  An example from a log where this was the case would not show incrementing JFM events received such as [ INFO] site: Total JFM Events received since startup: 0

 

Common Issue:

iDash will retry connections to CA7 up to the amount of retry times that you specify in the admin tool and if it does not succeed during those attempts it will stop retrying.  The interval between connections and amount of connection retries are configurable in the Admin Tool under the CA7 tab.  Once you select an instance on the CA7 tab, under the Event Subscription tab there are parameters for Retry Interval, Retry Count, and iDash Resubscribe Count.  When this occurs you will see messages like the below in the logs:

[ WARN] instance_id: iDash has stopped retrying for JFM subscriptions from the CA 7 instance instance_id, Name=subscription_name ca7_host

[ WARN] NSC2: To restart subscribing again, re-save the instance definition for CA 7 instance instance_name in the Admin Tool

To Resubscribe all you need to do is go into the Admin Tool and open that CA7 instance.  Backspace out a letter of the hostname and then re-enter it and re-save the connection.  Upon re-saving it will re-test the connection and also resubscribe to the JFM subscription.