Why would an idle z/OS Connect Enterprise Edition (zCEE) server do an excessive amount of I/O? I created a default server and started the BAQSTRT job and even though the server is idle it is generating 100K EXCPs each minute.
Answer by BillBulfin (11793) | Feb 27, 2017 at 03:51 PM
See topic Server configuration updates on demand in the z/OS Connect EE documentation to review the various default polling intervals.
The zCEE server.xml 'config' element has a MonitorInterval that defaults to 500ms. The description is:
Rate at which the server checks for configuration updates. Specify a positive integer followed by a unit of time, which can be hours (h), minutes (m), seconds (s), or milliseconds (ms). For example, specify 500 milliseconds as 500ms. You can include multiple values in a single entry. For example, 1s500ms is equivalent to 1.5 seconds.
You can review the above description which is in the topic Configuration elements in the z/OS Connect EE documentation and decide if you want to modify the MonitorInterval value or specify the updateTrigger=disabled to not monitor updates.
Other server.xml configuration elements that poll are the element zosconnect_zosConnectAPIs where the pollingRate default to 5s. The description is:
Controls how often the server polls the apis directory. Specify a positive integer followed by a unit of time, which can be hours (h), minutes (m), seconds (s), or milliseconds (ms). For example, specify 500 milliseconds as 500ms. You can include multiple values in a single entry. For example, 1s500ms is equivalent to 1.5 seconds.
It can also be disabled using updateTrigger=disabled.
The other zCEE configuration element to review is the zosconnect_zosConnectDataXform element which defaults to 2s. The description is:
Rate at which the server checks for updates to data transformation related files such as bind or schema files. Specify a positive integer followed by a unit of time, which can be hours (h), minutes (m), seconds (s), or milliseconds (ms). For example, specify 500 milliseconds as 500ms. You can include multiple values in a single entry. For example, 1s500ms is equivalent to 1.5 seconds.
Each of these default intervals will result in EXCP counts going up over time while the zCEE server is active.
The application dropins directory is also monitored by default at 500ms and can be disabled in server.xml by specifying:
<applicationMonitor dropins="dropins" dropinsEnabled="false" pollingRate="360s" updateTrigger="mbean"/>
Bill Bulfin
IBM CICS Level2 Support
Answer by Andy_Armstrong (143) | Apr 21, 2017 at 04:56 PM
Very useful - thanks @BillBulfin
High CPU usage and high EXCP count in "idle" z/OS Connect EE server 1 Answer
High CPU and I/O for ZOS connect test and Prod 1 Answer
zos connect server creation error 2 Answers
Does Bluemix offer dynamic allocation of servers based on load of a service? 2 Answers
System lane tasks work fine in my PC environment but they are very slow in my PS environment 5 Answers