Hello,

I have a long running HTTP task which fires a request through HaProxy to a backend pool. I have keepalived turned on for this request and a 3 hour timeout configured. I am seeing in HaProxys logs evidence of VCron giving up on the connection. Per HaProxy the disconnect code means:

CD The client unexpectedly aborted during data transfer. This can be
caused by a browser crash, by an intermediate equipment between the
client and haproxy which decided to actively break the connection,
by network routing issues between the client and haproxy, or by a
keep-alive session between the server and the client terminated first
by the client.

I am seeing this every 15 minutes. It seems the connection is getting severed just not sure where. I've dug around and didn't' see any settings in VCron which could be causing this. Any insight is greatly appreciated.

Have you done some tests against other urls? Like google.com

---

Not sure how you would recommend testing a long lasting https request against google. Most all of our VisualCron usage is HTTP related with hundreds of jobs spanning multiple datacenters. Most of our jobs don't need to run very long but we have the occasional request that goes on for an hour or more. Generally speaking our jobs call an internal URL kicking off a process and on completion of the job the HTTP request should finish. Because this happens predictably at the 15 minute mark it makes me think there is some setting which we could tweak which would fix this.

After exactly how long time does it disconnect?

---

We see the connection severed every 15 minutes and it looks like VCron makes another request although we have those features turned off in VCron.

Can you provide a sample url where we can reproduce this?

---

They are all internal URLS so not likely.

Not sure how we can reproduce this. We have not heard of this issue before.

---

Not entirely sure either. I'll let you know if I have any ideas. This could be an issue with HaProxy and not an issue with the VCron.