Question
2
Replies
91
Views
Posted: January 22, 2020
Last activity: January 23, 2020
Solved
System Operations - Requestor Pools - Information on timeout value
Summary: During investigation of Service Time out issue, we are looking for the number of service timeouts occured with respect to the service and node. The Requestor Pools landing page provide details about Timeout count. Unfortunately there is no information saying that timeout count is between the timeframe of x to y.
Query:
- How the timeout value is displayed in the "Requestor pool details" section (i.e. How it is calculated).
- Timeout means, PEGA Service responded with fault message as "Connection could not be established"?
- If timeout count is "15". Does it means that, from JVM start time the service package experienced 15 timeouts?
Navigation to Requestor Pool details:
Designer Studio -> System -> Operations -> Requestor Pools -> Filter the Service package -> Click "View Details"
Platform Version: PEGA 7.3.1
FYI: I have attached the screenshot.
Hi Nemi,
You are right that there is no section of info confirming that the timeout count is between the timeframe of x to y.
As an alternative, to get the timestamps you can scan the node's PegaRULES log for the below messages over a period of time until the last restart. (shutdown clears requestor pools)
"Timed out borrowing service requestor from requestor pool for service package:"
"Timeout waiting for idle object"
Ans:- Max Wait (Configured) defines the timeout value in requestor pool details. You can open the corresponding Service Package pooling tab (Maximum wait (in seconds)) provides the value for the timeout.
Ans:- This time out not Response timeout which is configured on REST services or Connectors. Based on the value specified, Pega (as a system) waits for a 1 pool thread to assign for an incoming service request when the maximum active requestors is reached. If this timeout interval passes before any requestor returns to the pool, the incoming request will be failed before invoking the service itself.
Ans:- Yes there were 15 such failed requests from JVM start time.
Also if someone manually clears the requestor pools, all the value counters will be reset.
Go through this Pooling tab properties for complete understanding - https://community.pega.com/sites/default/files/help_v731/procomhelpmain.htm#data-/data-admin-/data-admin-servicepackage/pooling.htm
Let me know if that helps!