-
Bug
-
Resolution: Done
-
High
-
Casablanca Release, Casablanca Maintenance Release
-
These results were found in the OOF project in the ONAP lab.
GET queries to multicloud infra_workload API respond with Bad Gateway errors.
It seems like it can take a very long time for multicloud to put together a response - such that the query times out.
When SO uses this multicloud API to create a stack, it is polling for CFEATE_COMPLETE.
the BUILDING statuses take about 10 seconds to respond, but once the stack is completed in the cloud, the final get query is taking about 80 seconds - which results in the bad gateway response. SO interprets this as a failure to create the stack.
this was reproducible via curl too:
$ no_proxy=10.12.6.115 curl --connect-timeout 30 -m 180 http://10.12.6.115:30280/api/multicloud/v1/CloudOwner/ONAP-POD-01-Rail-06/infra_workload/04ccd355-166a-4126-883b-2a208f7716b8 <html>
<head><title>502 Bad Gateway</title></head>
<body bgcolor="white">
<center><h1>502 Bad Gateway</h1></center>
<hr><center>openresty</center>
</body>
</html>
In this example, (by watching the multicloud log file activity), the response took 67 seconds to create - which is after the curl command gave up. Similar to what happens inside SO.
- mentioned in
-
Page Loading...