We have an app that is utilizing Batch call outs and when making these calls the initiation is taking an extremely long time. This behavior was not previously reported and this app has been available for over a year. This was first reported by our client on 8/22/12. Has something changed that would cause this behavior. Here is a scenario:
We investigated the complete process at our end and below are the findings :
Scenario & Observations:: - I selected 2 contacts on a Company - 2:33:00 --> Batch was initiated (Contact recs --> Pending Status) - For 1m 44s, this Batch did not even start, this is time lost in the SFDC queue - 2:34:44 --> Contact Append Batch Started & Completed (ended 2:34:50) * 2 Serial Jobs ran and completed in 4s & 6s (i.e. moving Contacts from Pending to In-Progress & Completed) - 2:34:48 --> Contact recs were updated with the "In-Progress" status by the Poller job. - 2:34:50 --> Contact Append Batch had already Completed, no more work left to do here. - 2:34:50 --> An instance of Contact Search Job is submitted. (* The only function here is to present the status of the 2 contacts created, by polling the state of the 2 serial Jobs.). - For 2m 56s, this Batch did not even start, this again is time lost in the SFDC queue - 2:37:47 --> This Job is Completed. (in the log, this job lasted only 1s, and was waiting in queue for rest of the time.)
Overall, in this test, we learned that 4m.40s were time lost in waiting.
Can you please confirm if something has recently changed? Can anything be done on your side to improve this?
This was not happening previously and seems to have been introduced in the last 4-6 months.
We have an app that is utilizing Batch call outs and when making these calls the initiation is taking an extremely long time. This behavior was not previously reported and this app has been available for over a year. This was first reported by our client on 8/22/12. Has something changed that would cause this behavior. Here is a scenario:
We investigated the complete process at our end and below are the findings :
Scenario & Observations:: - I selected 2 contacts on a Company - 2:33:00 --> Batch was initiated (Contact recs --> Pending Status) - For 1m 44s, this Batch did not even start, this is time lost in the SFDC queue - 2:34:44 --> Contact Append Batch Started & Completed (ended 2:34:50) * 2 Serial Jobs ran and completed in 4s & 6s (i.e. moving Contacts from Pending to In-Progress & Completed) - 2:34:48 --> Contact recs were updated with the "In-Progress" status by the Poller job. - 2:34:50 --> Contact Append Batch had already Completed, no more work left to do here. - 2:34:50 --> An instance of Contact Search Job is submitted. (* The only function here is to present the status of the 2 contacts created, by polling the state of the 2 serial Jobs.). - For 2m 56s, this Batch did not even start, this again is time lost in the SFDC queue - 2:37:47 --> This Job is Completed. (in the log, this job lasted only 1s, and was waiting in queue for rest of the time.)
Overall, in this test, we learned that 4m.40s were time lost in waiting.
Can you please confirm if something has recently changed? Can anything be done on your side to improve this?
This was not happening previously and seems to have been introduced in the last 4-6 months.
