Max Attempts in Queue-For-Agent
I have an activity calling queue-for-agent with max attempts "5".
Its a periodic agent with time provided 30 seconds.
But, on tracing that agent, it runs 5 times at once.
Instead the expected behaviour should be like the agent should try 5 times at interval of 30 seconds.
***Edited by Moderator: Pallavi to update platform capability tags***
Pegasystems Inc.
US
Please check the settings described in this Support Article.
https://community.pega.com/support/support-articles/all-attempts-execute-within-same-agent-run
Capgemini
IN
Thanks for answer.
In the help instructions for agent schedule if the auto-queue management is checked then on failing of Agent the agent should stop retrying until the time interval provided. Once it wakes up again it retry. But, this is not happening instead in a single run agent is retrying the same queue again and again until max attempts is reached.
As per the article provided you providing minimum age for processing would delay each of the queue going to agent.
But, this behavior is the expected behavior of the time interval provided in the schedule of periodic agent.
So, if we are providing a time interval for the running of agent why should we delay sending of queue to agent again ?
Thanks for answer.
In the help instructions for agent schedule if the auto-queue management is checked then on failing of Agent the agent should stop retrying until the time interval provided. Once it wakes up again it retry. But, this is not happening instead in a single run agent is retrying the same queue again and again until max attempts is reached.
As per the article provided you providing minimum age for processing would delay each of the queue going to agent.
But, this behavior is the expected behavior of the time interval provided in the schedule of periodic agent.
So, if we are providing a time interval for the running of agent why should we delay sending of queue to agent again ?
Can you also explain the use of max records in the above screen ?
Pegasystems Inc.
IN
max records indicates how many items the agent should process for the every wake up, so in your case it is 50 default so you have kept maxAttempts as 3, once is it failed again it will queued again to the agent so in the current iteration only the agent will process the next try.
Optional. The minimum age of a queue entry in milliseconds before the agent tries to process it. The dequeueing process ignores (skips over) any entries that are newer than this value.
For example, if users create queue entries while processing work items, it is important to ensure that a user releases all work item locks before an agent attempts to process the queue entry. To address this requirement, configure a small delay with the MinimumAgeForProcessing parameter.
so if you keep some time than agent will not process the item
Example:
max records indicates how many items the agent should process for the every wake up, so in your case it is 50 default so you have kept maxAttempts as 3, once is it failed again it will queued again to the agent so in the current iteration only the agent will process the next try.
Optional. The minimum age of a queue entry in milliseconds before the agent tries to process it. The dequeueing process ignores (skips over) any entries that are newer than this value.
For example, if users create queue entries while processing work items, it is important to ensure that a user releases all work item locks before an agent attempts to process the queue entry. To address this requirement, configure a small delay with the MinimumAgeForProcessing parameter.
so if you keep some time than agent will not process the item
Example:
Agent Delay is 30Sec
MinAge : 50 sec
so on 1st attempt the item will be failed and re queued . now the min age is 50 sec so when checks the item for executing the age is not yet reached so it will ignore the item
Next the condition will be true and process the item
Hope this will helps you
Capgemini
IN
Thanks for the help.
As the max attempt provided is 3 I would expect that the agent should try again, when it wakes up, if it fails in its first attempt.
But, this is not happening rather in the first run the agent is making 3 attempts in single run.
And, its not trying again.