aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorStaffan Olsson <staffan@repos.se>2017-11-07 12:57:01 +0100
committerStaffan Olsson <staffan@repos.se>2017-11-07 12:58:53 +0100
commitb462dac8cfd33ccac4aec84fcf5f915cbfc135a3 (patch)
tree1d7a0dda5d5f42edb532f301e46f0764ec7c9658
parente0824f64aa7feb629f3ef127c0ba85393c0e1d8c (diff)
downloadkubernetes-kafka-kafka-readiness-tcp.tar.gz
kubernetes-kafka-kafka-readiness-tcp.tar.bz2
kubernetes-kafka-kafka-readiness-tcp.zip
Actually, for readiness not liveness, let's go offline fast ...kafka-readiness-tcp
and trust alarms on Under-replicated Partitions to let us know when something is really wrong. Do clients actually care about Readiness? The bootstrap service (https://github.com/Yolean/kubernetes-kafka/pull/52) will definitely care, which is good. The `broker` service, that the StatefulSet manifest depends on (https://github.com/Yolean/kubernetes-kafka/blob/v2.1.0/50kafka.yml#L7) for naming, is without `publishNotReadyAddresses`. Clients will bootstrap, get the individual DNS names of brokers, resolve those addresses and connect directly to pods.
-rw-r--r--50kafka.yml2
1 files changed, 1 insertions, 1 deletions
diff --git a/50kafka.yml b/50kafka.yml
index cabbfad..50d0228 100644
--- a/50kafka.yml
+++ b/50kafka.yml
@@ -51,7 +51,7 @@ spec:
readinessProbe:
tcpSocket:
port: 9092
- timeoutSeconds: 9
+ timeoutSeconds: 1
volumeMounts:
- name: config
mountPath: /etc/kafka