Timeline for Ghost processes / batch processes (NA34 and NA39)
Current License: CC BY-SA 3.0
31 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 4, 2017 at 16:47 | vote | accept | Eric | ||
| S Apr 4, 2017 at 14:40 | history | bounty ended | CommunityBot | ||
| S Apr 4, 2017 at 14:40 | history | notice removed | CommunityBot | ||
| Mar 30, 2017 at 20:44 | history | edited | Eric | CC BY-SA 3.0 |
added 62 characters in body; added 10 characters in body; edited body
|
| Mar 30, 2017 at 20:23 | history | edited | Eric | CC BY-SA 3.0 |
deleted 228 characters in body
|
| Mar 30, 2017 at 18:51 | answer | added | Adrian Larson♦ | timeline score: 3 | |
| Mar 30, 2017 at 16:09 | comment | added | Eric | @AdrianLarson - If I answer my own question will the bounty still be split across the uprooted non accepted answers? I would still like the bounty to be split for participation.... | |
| Mar 30, 2017 at 16:07 | answer | added | Aidan | timeline score: 4 | |
| Mar 30, 2017 at 16:06 | vote | accept | Eric | ||
| Mar 30, 2017 at 16:08 | |||||
| Mar 30, 2017 at 16:06 | answer | added | Eric | timeline score: 8 | |
| Mar 30, 2017 at 3:12 | history | edited | Eric | CC BY-SA 3.0 |
added 329 characters in body
|
| Mar 27, 2017 at 20:14 | comment | added | Eric | @DanielBallinger - Unclear - but it looks like all records (90% sure of that). And they have not reported that it is still happening, or at least our frontline has not escalated it back to me that it is | |
| Mar 27, 2017 at 20:13 | comment | added | Daniel Ballinger | @Eric Are all the records in the batch getting replayed, or just a subset of them? | |
| Mar 27, 2017 at 20:11 | comment | added | Eric | @DanielBallinger - Correct. But there is nothing in the org(s) (that I can find) causing it to happen. It cannot be a human as there are a few hundred records processed at the same time (Ghost). Its like the batch ran again, but there is no indication that it did. And if it did it would have been on old data. Its like there is a shadow org or something. But even then the timestamps do not make sense and cause more question if that was the case. To make it more perplexing, one customer has been running a version from 4 years ago without issue and they reported it happening too. | |
| Mar 27, 2017 at 20:08 | comment | added | Daniel Ballinger | @Eric It sounds like something happening outside the batch as I assume when the batch completes it updates the records to indicate they have been processed. If it was the batch running again you would see a change in the timestamps. | |
| Mar 27, 2017 at 20:00 | comment | added | Eric | @DanielBallinger - the first callout is successful so there is no error there. Also, if the batch was manually ran again, our code would not process it because the previous batch successfully created a record. It would not be processed again for another month or year depending on the config for that record. First time in a long time I have been stumped. I am usually a bloodhound when it comes to tracking down issues...... | |
| Mar 27, 2017 at 20:00 | comment | added | Eric | @DanielBallinger - there are no batch jobs being ran at anytime close to the duplicates. Some as long as several hours away. As for the external system, it varies, Authorize.net and other payment gateways. Happening on records that have been recurring charges annually for a few years without modification. I believe I have ruled out all possibilities for duplicate processes or something in the external systems. No out bound messages. It is a small number of clients who only thing in common is that nothing was running at the time. (in the org according to standard monitoring) | |
| Mar 27, 2017 at 19:56 | comment | added | Daniel Ballinger | What is handling the request on their end? Do they have some form of message queuing system that might be retrying the request? | |
| Mar 27, 2017 at 19:55 | comment | added | Daniel Ballinger | Are there any Outbound Messages involved in the Callout process where an inital NACK response or timeout could cause the request to be retried. | |
| Mar 27, 2017 at 19:54 | comment | added | Daniel Ballinger | I assume the batch is making a Callout to the external system. Do you have access to the Apex Callout EventLogFile? I know it would be a paid feature, but it might show you when the duplicate callouts are occurring. | |
| Mar 27, 2017 at 19:52 | comment | added | Daniel Ballinger | "Batch takes on average 5-10 to complete". Seconds/Minutes/Hours/Iterations? Not sure it really matters which, just trying to clarify. | |
| Mar 27, 2017 at 15:46 | history | tweeted | twitter.com/StackSalesforce/status/846387842024648704 | ||
| Mar 27, 2017 at 14:42 | answer | added | Phil Hawthorn | timeline score: 8 | |
| Mar 27, 2017 at 13:30 | history | edited | Eric | CC BY-SA 3.0 |
added 132 characters in body
|
| S Mar 27, 2017 at 13:30 | history | bounty started | Eric | ||
| S Mar 27, 2017 at 13:30 | history | notice added | Eric | Draw attention | |
| Mar 23, 2017 at 21:39 | comment | added | Eric | @sfdcfox - for this issue yes, I believe it would be the lesser of two evils...The process leave a lot to be desired IMHO......All in good fun of course. | |
| Mar 23, 2017 at 21:27 | comment | added | sfdcfox♦ | You'd rather get married again than submit a support ticket?! | |
| Mar 23, 2017 at 21:04 | history | edited | Eric | CC BY-SA 3.0 |
added 15 characters in body
|
| Mar 23, 2017 at 20:55 | history | edited | Eric | CC BY-SA 3.0 |
added 39 characters in body
|
| Mar 23, 2017 at 20:47 | history | asked | Eric | CC BY-SA 3.0 |