[
https://issues.jboss.org/browse/TEIIDSB-121?page=com.atlassian.jira.plugi...
]
Steven Hawkins commented on TEIIDSB-121:
----------------------------------------
And i can't get swagger.json at all. Only CPU utilization 100%.
I'll have a look using something similar to what you have provided. More than likely
we'd need to cut down on the amount of metadata generated. A simple workaround would
probably be to split your public facing schema up into several smaller ones.
How deep are your relationships? As I recall the openapi metadata generator only
generates a single nested level, but I'll need to verify that.
P.S. in stateless micro-service architecture we do not have
persistence layer (FS).
You don't have to view this as state. It's more akin to a cache than persistence.
For example you can use an emptyDir volume:
https://kubernetes.io/docs/concepts/storage/volumes/#emptydir
cpu utilization on 100% when generate openapi doc by metadata
-------------------------------------------------------------
Key: TEIIDSB-121
URL:
https://issues.jboss.org/browse/TEIIDSB-121
Project: Teiid Spring Boot
Issue Type: Bug
Components: datasource, OData
Affects Versions: 1.2.0
Reporter: Renat Eskenin
Priority: Major
When i get request as
http://localhost:8080/odata/swagger.json or
http://localhost:8080/odata/datasource/Account/swagger.json or
http://localhost:8080/odata/datasource/swagger.json
cpu utilization 100% and i do not get any response.
We have 2 bad approach:
1. In org.teiid.olingo.web.OpenApiHandler:151 teiid create tmp file in filesystem.
Its bad because in Docker k8s systems applications can't create any files in FS.
2. In org.apache.xalan.transformer.TransformerImpl:709 we get system freeze (CPU
utilization 100% infinite)
--
This message was sent by Atlassian Jira
(v7.13.8#713008)