Transient http gateway failure

I’m developing/maintaining an interface to the Infusionsoft API that has been working well for the last 3 years. On February 19 at approximately 19:17 the infusionsoft api returned …

XML-RPC: xmlrpcmsg::parseResponseHeaders: HTTP error, got response: HTTP/1.1 502 Bad Gateway

Processing has proceeded without error since. I’m assuming this was a transient error? Has anyone else seen a 502 error? Recently or in the past?

Thanks in advance


1 Like

That is usually a time out error which can be due to especially high server load, updates, or simply a short lived process on the server running up bandwidth (or a number of less likely things like DDoS attacks and such). If you’re not seeing it since then it would be something like this. I might run exception handling to retry in 30-60 seconds when you receive the server response again.

I’ve seen all sorts of errors: 502 Bad Gateway, 500 General Error, timeouts, etc. At one time there was a rough patch so bad that the API would be unavailable for several hours every day. I wrote to Support and sent them logs of API endpoints, timestamps, payloads and what not, but they were unable to resolve it. They said their dev team found nothing and asked me send the logs again for their ops team. At this point, frankly, I gave up.

Even now I get 1-2 hours of unavailability now and then. :man_shrugging:

I am curious Ankush, what country is your web server being hosted in? Is it a shared server?

Your intermittent issues seems to be a server or connection issue.
If the API connection was that bad, we would get far more developers mentioning this.

Hey Pav,

It’s a VPS hosted on LiquidWeb infrastructure in the US.

Your intermittent issues seems to be a server or connection issue.

That’s a very valid point! I’ll try to gather more data and match the outages against other API connections and server availability in general.