Some time ago we have created few apps and integrated it within a client portals. On this app we have created a webhook which is triggering and sending payload to url when a contact is deleted. It then sends the info to our application (which is secured with SSL provided from Comodo).
In those portals our clients sometimes deletes many contacts (up to few thousands).
The problem is that we observed in the logs that not all requests are returning the OK 200 status.
We have observed like 4-10 % of success rate depending on the portal.
The issue is always the same: Concurrent wait timed out
After reading the forum I have found only threads saying that the SSL was causing the issue and this is not the case, as we receive the request from webhook.
I have made tests from my local machine to the same server with half a milion of requests send in about 10 minutes and it handled it properly (queued all of them) and success rate was 100%. That surprised me and I have went to webhook logging to see how it is possible. I was surprised, because in the documentation of webhook api it says that contacts would be grouped in the batches to improve the performance. That is understandable for me from the perspective, althrough i have checked nearly 1k contacts and all of them were send with different batch Id (haven’t found any that is duplicated). Also our logging mechanism didn’t register a batch which size is bigger than 1.
Regarding to it I have a few questions:
Is it possible to download the logging information from webhook in webhook logging, as we need to retrieve the ids of the contacts that were deleted in hubspot? I haven’t found any option for this particular scenario neither in the documentation or in this forum. I would like to point out that only in one day (today 3.10.2017) over 15k requests via this API which failed (I have stopped counting after page 155, at this is only today). All of them are having the Retry attemp: 9 value.
In the failed requests from webhook I have seen that for all of them have following fields empty;
Attempt start timestamp
Attempt finish timestamp
Is this expected behaviour?
what is the concurrency limit? As I haven’t found it (unless this is specified in the hubspot webhook api documentation and said as 1 second?)
If you need any further details, please specify.