I have configured a webhook in Marketo but Marketo is behaving weird and intermittently giving errors.
Connect to api.***.com:443 [api.***.com/192.***.***.**] failed: connect timed out.
At times Marketo captures the response successfully and at times it throws the above error. The time when Marketo throws error I checked in postman & insomnia and its working fine.
Did anyone faced the same issue?
Request Type:
|
http_post
|
Step ID:
|
17
|
Source:
|
Marketo Flow Action
|
exceptionMessage:
|
Connect to api.***.com:443 [api.***.com/192.***.***.**] failed: connect timed out
|
Response:
|
|
Lead ID:
|
111111
|
fault:
|
host: hostname: api.***.com
port: 443 address: cause: |
Error Type:
|
Failed to connect to host or proxy.
|
Well, you’re not actually calling from the same source IP and with the same frequency. I’m inclined to believe Marketo is telling the truth: the remote service is down when it says it’s down.
Are you passing a large number of people through the webhook in a given period? Do you know the rate limits of the remote service?
The rate limit I have set is 50 calls per min. I guess this shouldn't be a problem.
I have added some logs in marketo and identified that on a daily basis at 3AM PDT time this issue re-occur. And it last for few mins to few hours.
Really, I'm just re-iterating what @SanfordWhiteman has already said. I think you need to look at the destination and see why that's blocking calls at that time.
It feels unlikely to be Marketo.
Cheers
Jo
50 calls per minute is a tiny fraction of what Marketo can send. Marketo will open 20 concurrent connections, allowing thousands of calls per minute.
It’s very hard (basically impossible) to perceive of how Marketo could time out (meaning the connection is attempted) without this being the fault of the remote server’s rate limits. The fact of it happening at a particular time sounds like the remote service does some kind of garbage collection at that time.