Description:
We've encountered some cases where our Envoy Proxy got default xDS configurations from Envoy Gateway with
direct_response": {
status": 500
}
We didn't see anything on the Envoy Gateway pods logs indicating the reason for updating the xDS with default "direct_response 500" configuration.
We wanted to raise the log level to maybe see an higher level log indicating the reason but then, after restarting the Envoy Gateway pods, the problem resolved.
It would be good to have some indication why Envoy Gateway decided to put the default configurations on the xDS.
Default configuration example:
{
"configs": [
{
"dynamic_listeners": [
{
"active_state": {
"listener": {
"filter_chains": [
{
"filters": [
{
"name": "envoy.filters.network.http_connection_manager",
"typed_config": {
"route_config": {
"virtual_hosts": [
{
"routes": [
{
"direct_response": {
"status": 500
}
}
]
}
]
},
"http_filters": [
{
"name": "envoy.filters.http.router"
}
]
}
}
]
}
]
}
}
}
]
}
]
}
Description:
We've encountered some cases where our Envoy Proxy got default xDS configurations from Envoy Gateway with
We didn't see anything on the Envoy Gateway pods logs indicating the reason for updating the xDS with default "direct_response 500" configuration.
We wanted to raise the log level to maybe see an higher level log indicating the reason but then, after restarting the Envoy Gateway pods, the problem resolved.
It would be good to have some indication why Envoy Gateway decided to put the default configurations on the xDS.
Default configuration example: