Details
-
Bug
-
Resolution: Done
-
Major
-
None
-
Logging 5.5, Logging 5.6, Logging 5.7
-
3
-
False
-
-
False
-
OBSDOCS (Aug 21-Sep 11) #241, OBSDOCS (Sep 11 - Oct 2) #242, OBSDOCS (Oct 2 - Oct 23) #243, OBSDOCS (Oct 23 - Nov 13) #244, OBSDOCS (Nov 13 - Dev 4) #245, OBSDOCS (Dec 4 - Dev 25) #246, OBSDOCS (Jan 1 - Jan 22) #247, OBSDOCS (Jan 22 - Feb 12) #248, OBSDOCS (Feb 12 - Mar 4) #249, OBSDOCS (Mar 4 - Mar 25) #250
-
Important
Description
URL
https://docs.openshift.com/container-platform/4.11/logging/cluster-logging-loki.html
Description of problem:
Before https://issues.redhat.com/browse/LOG-4008 (Logging 5.6.5) and https://issues.redhat.com/browse/LOG-3721 (Logging 5.7.0) if it was used a no private IP for pods (clusterNetwork), then, it was failing to start the Loki pods.
The fix enables to be able to use this no Private IPs for pods, but it's needed to modify the Logging as defined in article https://access.redhat.com/solutions/6998362 where it's needed to add:
```
apiVersion: loki.grafana.com/v1
kind: LokiStack
metadata:
name: logging-loki
namespace: openshift-logging
spec:
...
hashRing: <--- this and the next lines
type: memberlist
memberlist:
instanceAddrType: podIP
```
or with:
```
$ oc patch LokiStack logging-loki -n openshift-logging --type=merge -p '{"spec": {"hashRing":{"memberlist":{"instanceAddrType":"podIP","type": "memberlist"}}}}'
```
This should be documentated for when installing Loki and how the Loki definition should be in case that using no private IPs
