-
Bug
-
Resolution: Done
-
Major
-
Logging 5.6.0
-
False
-
None
-
False
-
NEW
-
VERIFIED
-
Before this change prometheus exporter could not bind to ipv6 address. After this change prometheus exporter can bind to both ipv4 and ipv6 address
-
-
-
Log Collection - Sprint 229, Log Collection - Sprint 230
Description of problem:
No data point on Openshift Logging colllection dashboard on the IPv6 single stack. The port 24231 is binded to IPv4 address. It should bind to ipv6 address.
[sinks.prometheus_output] type = "prometheus_exporter" inputs = ["add_nodename_to_metric"] address = "0.0.0.0:24231" default_namespace = "collector"
By the way, collector.openshift-logging.svc:2112 works well
Version-Release number of selected component (if applicable):
Logging 5.6.0
How reproducible:
Always
Steps to Reproduce:
1) Launch a single stack cluster
2) Deploy cluster logging for forward logs to default via vector.
3) check the metrics on collector
oc exec <collector_pod> – curl -k https://collector.openshift-logging.svc:24231/metrics
oc exec <collector_pod> – curl -6 -g -k https://[::]:24231/metrics
Actual results:
curl: (7) Failed to connect to collector.openshift-logging.svc port 24231: Connection refused
command terminated with exit code 7
Additional info:
Works well on both IPv4 and IPv6 single stack cluster when I use address = "[::]:24231