Running Kafka Cluster with multiple Brokers on local minikube - apache-kafka

For testing purposes I try to create a Kafka Cluster on my local minikube. The Cluster must be reachable from outside of the Kubernetes.
When I produce/consume from inside the pods there is no problem, everything works just fine.
When I produce from my local machine with
bin/kafka-console-producer.sh --topic mytopic --broker-list 192.168.99.100:32767
where 192.168.99.100 is my minikube-ip and 32767 is the node port of the kafka service.
I get the following Error Message:
>testmessage
>[2018-04-30 11:55:04,604] ERROR Error when sending message to topic ams_stream with key: null, value: 11 bytes with error: (org.apache.kafka.clients.producer.internals.ErrorLoggingCallback)
org.apache.kafka.common.errors.TimeoutException: Expiring 1 record(s) for ams_stream-0: 1506 ms has passed since batch creation plus linger time
When I consume from my local machine I get the following warnings:
[2018-04-30 10:22:30,680] WARN Connection to node 2 could not be established. Broker may not be available. (org.apache.kafka.clients.NetworkClient)
[2018-04-30 10:23:46,057] WARN Connection to node 8 could not be established. Broker may not be available. (org.apache.kafka.clients.NetworkClient)
[2018-04-30 10:25:01,542] WARN Connection to node 2 could not be established. Broker may not be available. (org.apache.kafka.clients.NetworkClient)
[2018-04-30 10:26:17,008] WARN Connection to node 5 could not be established. Broker may not be available. (org.apache.kafka.clients.NetworkClient)
The Broker IDs are right, so it looks like I can at least reach the brokers
Edit:
I think that the Problem may be, that the service is routing me "randomly" to any of my brokers, but he needs to route me to the leader of the topic.
Could this be the Problem? Does anybody know a way around this Problem?
Additional Information:
I'm using the wurstmeister/kafka and digitalwonderland/zookeeper images
I started using the DellEMC Tutorial (and the linked one from defuze.org)
This did not work out for me so I made some changes in the kafka-service.yml (1) and the kafka-cluster.yml (2)
kafka-service.yml
added a fixed NodePort
removed id from the selector
kafka-cluster.yml
added replicas to the specification
removed id from the label
changed the broker id to be generated by the last number from the IP
replaced deprecated values advertised_host_name / advertised_port with
listeners ( pod-ip:9092 ) for communication inside the k8s
advertised_listeners ( minikube-ip:node-port ) for communication with applications outside the kubernetes
1 - kafka-service.yml:
---
apiVersion: v1
kind: Service
metadata:
name: kafka-service
labels:
name: kafka
spec:
type: NodePort
ports:
- port: 9092
nodePort: 32767
targetPort: 9092
protocol: TCP
selector:
app: kafka
type: LoadBalancer
2 - kafka-cluster.yml:
---
kind: Deployment
apiVersion: extensions/v1beta1
metadata:
name: kafka-b
spec:
replicas: 3
template:
metadata:
labels:
app: kafka
spec:
containers:
- name: kafka
image: wurstmeister/kafka
ports:
- containerPort: 9092
env:
- name: HOSTNAME_COMMAND
value: "ifconfig |grep 'addr:172' |cut -d':' -f 2 |cut -d ' ' -f 1"
- name: KAFKA_ZOOKEEPER_CONNECT
value: zk1:2181
- name: BROKER_ID_COMMAND
value: "ifconfig |grep 'inet addr:172' | cut -d'.' -f '4' | cut -d' ' -f '1'"
- name: KAFKA_ADVERTISED_LISTENERS
value: "INTERNAL://192.168.99.100:32767"
- name: KAFKA_LISTENERS
value: "INTERNAL://_{HOSTNAME_COMMAND}:9092"
- name: KAFKA_LISTENER_SECURITY_PROTOCOL_MAP
value: "INTERNAL:PLAINTEXT"
- name: KAFKA_INTER_BROKER_LISTENER_NAME
value: "INTERNAL"
- name: KAFKA_CREATE_TOPICS
value: mytopic:1:3

Related

How to make a redis cluster in k8s environment using nodeport service type?

I have tried to make a redis cluster in k8s environment using "NodePort" type of service.
More specifically, I want to compose a redis cluster across two different k8s cluster.
When I used LoadBalancer(External IP) for service type, cluster was made successfully.
The problem is NodePort.
After I command redis-cli --cluster create, it stucks on "Waiting for the cluster to join"
Below is the logs of cluster create command.
I deployed 4 leader pods and 4 slave pods with individual nodeport service.
root#redis-leader00-5fc546c4bd-28x8w:/data# redis-cli -a mypassword --cluster create --cluster-replicas 1 \
> 192.168.9.194:30030 192.168.9.199:30031 192.168.9.194:30032 192.168.9.199:30033 \
> 192.168.9.199:30030 192.168.9.194:30031 192.168.9.199:30032 192.168.9.194:30033
Warning: Using a password with '-a' or '-u' option on the command line interface may not be safe.
>>> Performing hash slots allocation on 8 nodes...
Master[0] -> Slots 0 - 4095
Master[1] -> Slots 4096 - 8191
Master[2] -> Slots 8192 - 12287
Master[3] -> Slots 12288 - 16383
Adding replica 192.168.9.199:30030 to 192.168.9.194:30030
Adding replica 192.168.9.194:30033 to 192.168.9.199:30031
Adding replica 192.168.9.199:30032 to 192.168.9.194:30032
Adding replica 192.168.9.194:30031 to 192.168.9.199:30033
M: 94bf3c6760e6b3a91c408eda497822b4961e8d82 192.168.9.194:30030
slots:[0-4095] (4096 slots) master
M: 31f4a9604b15109316f91956aa4a32b0c6952a4d 192.168.9.199:30031
slots:[4096-8191] (4096 slots) master
M: 0738d1e1a677352fc3b0b3600a67d837b795fa8a 192.168.9.194:30032
slots:[8192-12287] (4096 slots) master
M: 7dd7edbfab6952273460778d1f140b0716118042 192.168.9.199:30033
slots:[12288-16383] (4096 slots) master
S: 17e044681319d0a05bd92deeb4ead31c0cd468e2 192.168.9.199:30030
replicates 94bf3c6760e6b3a91c408eda497822b4961e8d82
S: 9c9e47ec566ac781e8e3dcb51398a27d1da71004 192.168.9.194:30031
replicates 7dd7edbfab6952273460778d1f140b0716118042
S: b8f7028b56f96565a91fdb442c94fbedcee088c2 192.168.9.199:30032
replicates 0738d1e1a677352fc3b0b3600a67d837b795fa8a
S: e4c9ffdf67e2b2ef9f840618110738358bde52d5 192.168.9.194:30033
replicates 31f4a9604b15109316f91956aa4a32b0c6952a4d
Can I set the above configuration? (type 'yes' to accept): yes
>>> Nodes configuration updated
>>> Assign a different config epoch to each node
>>> Sending CLUSTER MEET messages to join the cluster
Waiting for the cluster to join
...........................................
The weird point is that the other redis containers received the signal but there is no progress.
Below is the logs of other redis containers.
.... // Some logs for initializing redis
1:M 20 Jan 2022 06:09:12.055 * Ready to accept connections
1:M 20 Jan 2022 06:13:41.263 # configEpoch set to 5 via CLUSTER SET-CONFIG-EPOCH
I thought the communication was successful but gossip port didn't work properly.
So, I modified redis.conf and set cluster-announce-bus-port but it also didn't work.
How can I compose a redis cluster using nodeport type of service?
Please refer the one of .yaml files
apiVersion: v1
kind: ConfigMap
metadata:
name: redis-cluster-leader00
namespace: redis
labels:
app: redis-cluster
leader: "00"
data:
fix-ip.sh: |
#!/bin/sh
CLUSTER_CONFIG="/data/nodes.conf"
if [ -f ${CLUSTER_CONFIG} ]; then
if [ -z "${HOST_IP}" ]; then
echo "Unable to determine Pod IP address!"
exit 1
fi
echo "Updating my IP to ${HOST_IP} in ${CLUSTER_CONFIG}"
sed -i.bak -e "/myself/ s/[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}/${HOST_IP}/" ${CLUSTER_CONFIG}
fi
exec "$#"
redis.conf: |+
bind 0.0.0.0
cluster-enabled yes
cluster-require-full-coverage no
cluster-node-timeout 15000
cluster-config-file /data/nodes.conf
cluster-migration-barrier 1
appendonly no
save ""
protected-mode no
requirepass "mypassword"
masterauth "mypassword"
cluster-announce-ip 192.168.9.194
cluster-announce-port 30030
cluster-announce-bus-port 31030
---
apiVersion: v1
kind: Service
metadata:
name: redis-leader00
namespace: redis
labels:
app: redis
role: leader
tier: backend
leader: "00"
spec:
ports:
- port: 6379
targetPort: 6379
nodePort: 30030
name: client
- port: 16379
targetPort: 16379
nodePort: 31030
name: gossip
selector:
app: redis
role: leader
tier: backend
leader: "00"
type: NodePort
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: redis-leader00
namespace: redis
labels:
app: redis
role: leader
tier: backend
leader: "00"
spec:
replicas: 1
selector:
matchLabels:
app: redis
leader: "00"
template:
metadata:
labels:
app: redis
role: leader
tier: backend
leader: "00"
spec:
containers:
- name: leader
image: docker.io/redis:6.0.5
resources:
requests:
cpu: 100m
memory: 100Mi
ports:
- containerPort: 6379
name: client
- containerPort: 16379
name: gossip
volumeMounts:
- name: conf
mountPath: /conf
readOnly: false
args: ["--requirepass", "mypassword"]
command: ["/conf/fix-ip.sh", "redis-server", "/conf/redis.conf"]
env:
- name: HOST_IP
valueFrom:
fieldRef:
fieldPath: status.hostIP
volumes:
- name: conf
configMap:
name: redis-cluster-leader00
defaultMode: 0755
Also see nodes.conf file in container after I commanded redis-cli create
root#redis-leader01-87ccb466-bsnq4:/data# cat nodes.conf
31f4a9604b15109316f91956aa4a32b0c6952a4d 192.168.9.199:30031#31031 myself,master - 0 0 2 connected 4096-8191
vars currentEpoch 2 lastVoteEpoch 0
i am not sure actual process you are following to create the cluster of Redis however i would suggest checking out the helm chart to deploy the Redis cluster on K8s.
Using helm chart it's easy to manage and deploy the Redis cluster on K8s.
https://github.com/bitnami/charts/tree/master/bitnami/redis
To deploy chart you just have to run command :
helm install my-release bitnami/redis
On Node Port side once helm chart is deployed you can edit the service type of else you can also update the first helm chart and after that apply those changes to K8s.
This will create the Node port on K8s for Redis service.
You can find service template here : https://github.com/bitnami/charts/tree/master/bitnami/redis/templates
There was someone else who had already found the problem. This link is the issue he made. https://github.com/redis/redis/issues/6432
The cause of problem is that CLUSTER MEET is not use cluster-announce-bus-port. Actually, there is even no variable for storing cluster-announce-bus-port in clusterManagerNode struct.
There is already PR for this problem. https://github.com/redis/redis/pull/6442
It's not merged or closed yet so this is the ongoing problem...

Kafka on Kubernetes in minikube not working

I am trying to setup Kafka on Minikube, a very basic setup. I can't validate if Kafka and Zookeeper have been setup correctly because kafkacat fails.
Here is my config:
zookeeper
kind: Deployment
apiVersion: apps/v1
metadata:
name: zookeeper-deployment-1
spec:
selector:
matchLabels:
app: zookeeper-1
template:
metadata:
labels:
app: zookeeper-1
spec:
containers:
- name: zoo1
image: digitalwonderland/zookeeper
ports:
- containerPort: 2181
env:
- name: ZOOKEEPER_ID
value: "1"
- name: ZOOKEEPER_SERVER_1
value: zoo1
---
apiVersion: v1
kind: Service
metadata:
name: zoo1
labels:
app: zookeeper-1
spec:
ports:
- name: client
port: 2181
protocol: TCP
- name: follower
port: 2888
protocol: TCP
- name: leader
port: 3888
protocol: TCP
selector:
app: zookeeper-1
kafka
kind: Deployment
apiVersion: apps/v1
metadata:
name: kafka-broker0
spec:
selector:
matchLabels:
app: kafka
template:
metadata:
labels:
app: kafka
id: "0"
spec:
containers:
- name: kafka
image: wurstmeister/kafka
ports:
- containerPort: 9092
env:
- name: KAFKA_ADVERTISED_PORT
value: "9092"
- name: KAFKA_ADVERTISED_HOST_NAME
value: kafka-service
- name: KAFKA_ZOOKEEPER_CONNECT
value: zoo1:2181
- name: KAFKA_BROKER_ID
value: "0"
---
apiVersion: v1
kind: Service
metadata:
name: kafka-service
labels:
name: kafka
spec:
ports:
- port: 9092
name: kafka-port
protocol: TCP
selector:
app: kafka
id: "0"
pods
kafka-broker0-6885746967-6vktz 1/1 Running 0 5m20s
zookeeper-deployment-1-7f5bb9785f-7pplk 1/1 Running 0 5m25s
svc
kafka-service ClusterIP 10.99.226.129 <none> 9092/TCP 6m30s
zoo1 ClusterIP 10.96.140.187 <none> 2181/TCP,2888/TCP,3888/TCP 6m35s
kafkacat logs
✗ kafkacat -L -b kafka-service:9092 -d broker
%7|1596239513.610|BRKMAIN|rdkafka#producer-1| [thrd::0/internal]: :0/internal: Enter main broker thread
%7|1596239513.610|BROKER|rdkafka#producer-1| [thrd:app]: kafka-service:9092/bootstrap: Added new broker with NodeId -1
%7|1596239513.610|BRKMAIN|rdkafka#producer-1| [thrd:kafka-service:9092/bootstrap]: kafka-service:9092/bootstrap: Enter main broker thread
%7|1596239513.610|CONNECT|rdkafka#producer-1| [thrd:app]: kafka-service:9092/bootstrap: Selected for cluster connection: bootstrap servers added (broker has 0 connection attempt(s))
%7|1596239513.610|INIT|rdkafka#producer-1| [thrd:app]: librdkafka v1.4.0 (0x10400ff) rdkafka#producer-1 initialized (builtin.features gzip,snappy,ssl,sasl,regex,lz4,sasl_gssapi,sasl_plain,sasl_scram,plugins,zstd,sasl_oauthbearer, CC CXX PKGCONFIG OSXLD LIBDL PLUGINS ZLIB SSL SASL_CYRUS ZSTD HDRHISTOGRAM LZ4_EXT SYSLOG SNAPPY SOCKEM SASL_SCRAM SASL_OAUTHBEARER CRC32C_HW, debug 0x2)
%7|1596239513.610|CONNECT|rdkafka#producer-1| [thrd:kafka-service:9092/bootstrap]: kafka-service:9092/bootstrap: Received CONNECT op
%7|1596239513.610|STATE|rdkafka#producer-1| [thrd:kafka-service:9092/bootstrap]: kafka-service:9092/bootstrap: Broker changed state INIT -> TRY_CONNECT
%7|1596239513.610|CONNECT|rdkafka#producer-1| [thrd:kafka-service:9092/bootstrap]: kafka-service:9092/bootstrap: broker in state TRY_CONNECT connecting
%7|1596239513.610|STATE|rdkafka#producer-1| [thrd:kafka-service:9092/bootstrap]: kafka-service:9092/bootstrap: Broker changed state TRY_CONNECT -> CONNECT
%7|1596239513.610|CONNECT|rdkafka#producer-1| [thrd:app]: Not selecting any broker for cluster connection: still suppressed for 49ms: application metadata request
%7|1596239513.610|CONNECT|rdkafka#producer-1| [thrd:app]: Not selecting any broker for cluster connection: still suppressed for 49ms: application metadata request
%7|1596239513.610|CONNECT|rdkafka#producer-1| [thrd:app]: Not selecting any broker for cluster connection: still suppressed for 49ms: application metadata request
%7|1596239513.610|CONNECT|rdkafka#producer-1| [thrd:app]: Not selecting any broker for cluster connection: still suppressed for 49ms: application metadata request
%7|1596239513.614|BROKERFAIL|rdkafka#producer-1| [thrd:kafka-service:9092/bootstrap]: kafka-service:9092/bootstrap: failed: err: Local: Host resolution failure: (errno: Bad address)
NODEPORT UPDATE
✗ kafkacat -L -b kafka-service:30236 -d broker
%7|1596476848.078|STATE|rdkafka#producer-1| [thrd:kafka-service:30236/bootstrap]: kafka-service:30236/bootstrap: Broker changed state CONNECT -> DOWN
%7|1596476848.078|CONNECT|rdkafka#producer-1| [thrd:app]: Not selecting any broker for cluster connection: still suppressed for 46ms: application metadata request
%7|1596476848.078|CONNECT|rdkafka#producer-1| [thrd:app]: Not selecting any broker for cluster connection: still suppressed for 46ms: application metadata request
%7|1596476849.065|CONNECT|rdkafka#producer-1| [thrd:app]: Cluster connection already in progress: application metadata request
%7|1596476849.065|CONNECT|rdkafka#producer-1| [thrd:app]: Not selecting any broker for cluster connection: still suppressed for 49ms: application metadata request
% ERROR: Failed to acquire metadata: Local: Broker transport failure
minikube ip
✗ kafkacat -L -b 192.168.64.2:30236 -d broker
|1596476908.164|BRKMAIN|rdkafka#producer-1| [thrd:192.168.64.2:30236/bootstrap]: 192.168.64.2:30236/bootstrap: Enter main broker thread
%7|1596476908.164|CONNECT|rdkafka#producer-1| [thrd:app]: 192.168.64.2:30236/bootstrap: Selected for cluster connection: bootstrap servers added (broker has 0 connection attempt(s))
%7|1596476908.164|INIT|rdkafka#producer-1| [thrd:app]: librdkafka v1.4.0 (0x10400ff) rdkafka#producer-1 initialized (builtin.features gzip,snappy,ssl,sasl,regex,lz4,sasl_gssapi,sasl_plain,sasl_scram,plugins,zstd,sasl_oauthbearer, CC CXX PKGCONFIG OSXLD LIBDL PLUGINS ZLIB SSL SASL_CYRUS ZSTD HDRHISTOGRAM LZ4_EXT SYSLOG SNAPPY SOCKEM SASL_SCRAM SASL_OAUTHBEARER CRC32C_HW, debug 0x2)
%7|1596476908.164|CONNECT|rdkafka#producer-1| [thrd:192.168.64.2:30236/bootstrap]: 192.168.64.2:30236/bootstrap: Received CONNECT op
%7|1596476908.164|STATE|rdkafka#producer-1| [thrd:192.168.64.2:30236/bootstrap]: 192.168.64.2:30236/bootstrap: Broker changed state INIT -> TRY_CONNECT
%7|1596476908.164|CONNECT|rdkafka#producer-1| [thrd:192.168.64.2:30236/bootstrap]: 192.168.64.2:30236/bootstrap: broker in state TRY_CONNECT connecting
%7|1596476908.164|STATE|rdkafka#producer-1| [thrd:192.168.64.2:30236/bootstrap]: 192.168.64.2:30236/bootstrap: Broker changed state TRY_CONNECT -> CONNECT
%7|1596476908.164|CONNECT|rdkafka#producer-1| [thrd:app]: Not selecting any broker for cluster connection: still suppressed for 49ms: application metadata request
✗ kafkacat -L -b localhost:30236 -d broker
%7|1596477098.454|CONNECT|rdkafka#producer-1| [thrd:localhost:30236/bootstrap]: localhost:30236/bootstrap: Received CONNECT op
%7|1596477098.454|INIT|rdkafka#producer-1| [thrd:app]: librdkafka v1.4.0 (0x10400ff) rdkafka#producer-1 initialized (builtin.features gzip,snappy,ssl,sasl,regex,lz4,sasl_gssapi,sasl_plain,sasl_scram,plugins,zstd,sasl_oauthbearer, CC CXX PKGCONFIG OSXLD LIBDL PLUGINS ZLIB SSL SASL_CYRUS ZSTD HDRHISTOGRAM LZ4_EXT SYSLOG SNAPPY SOCKEM SASL_SCRAM SASL_OAUTHBEARER CRC32C_HW, debug 0x2)
%7|1596477098.454|STATE|rdkafka#producer-1| [thrd:localhost:30236/bootstrap]: localhost:30236/bootstrap: Broker changed state INIT -> TRY_CONNECT
%7|1596477098.454|CONNECT|rdkafka#producer-1| [thrd:localhost:30236/bootstrap]: localhost:30236/bootstrap: broker in state TRY_CONNECT connecting
%7|1596477098.454|CONNECT|rdkafka#producer-1| [thrd:app]: Not selecting any broker for cluster connection: still suppressed for 49ms: application metadata request
%7|1596477098.454|STATE|rdkafka#producer-1| [thrd:localhost:30236/bootstrap]: localhost:30236/bootstrap: Broker changed state TRY_CONNECT -> CONNECT
%7|1596477098.454|CONNECT|rdkafka#producer-1| [thrd:app]: Not selecting any broker for cluster connection: still suppressed for 49ms: application metadata request
%7|1596477098.454|CONNECT|rdkafka#producer-1| [thrd:app]: Not selecting any broker for cluster connection: still suppressed for 49ms: application metadata request
%7|1596477098.454|CONNECT|rdkafka#producer-1| [thrd:app]: Not selecting any broker for cluster connection: still suppressed for 49ms: application metadata request
%7|1596477098.460|CONNECT|rdkafka#producer-1| [thrd:localhost:30236/bootstrap]: localhost:30236/bootstrap: Connecting to ipv4#127.0.0.1:30236 (plaintext) with socket 9
%7|1596477098.461|BROKERFAIL|rdkafka#producer-1| [thrd:localhost:30236/bootstrap]: localhost:30236/bootstrap: failed: err: Local: Broker transport failure: (errno: Connection refused)
%7|1596477098.461|STATE|rdkafka#producer-1| [thrd:localhost:30236/bootstrap]: localhost:30236/bootstrap: Broker changed state CONNECT -> DOWN
%7|1596477098.461|STATE|rdkafka#producer-1| [thrd:localhost:30236/bootstrap]: localhost:30236/bootstrap: Broker changed state DOWN -> INIT
%7|1596477098.461|CONNECT|rdkafka#producer-1| [thrd:app]: Not selecting any broker for cluster connection: still suppressed for 43ms: application metadata request
%7|1596477098.461|CONNECT|rdkafka#producer-1| [thrd:app]: Not selecting any broker for cluster connection: still suppressed for 42ms: application metadata request
apiVersion: v1
kind: Service
metadata:
name: kafka-service
labels:
name: kafka
spec:
type: NodePort
ports:
- port: 9092
nodePort: 30236
name: kafka-port
protocol: TCP
selector:
app: kafka
id: "0"
I have never used kafkacat, but if it is a cli that is installed on your host, and not inside of another container, you can use it like this now:
✗ kafkacat -L -b localhost:30236 -d broker
localhost, or the ip of your kubernetes node.
Using a service default is of type ClusterIp, and this can be accessed only inside the kubernetes cluster

How to access kafka installed outside kubernetes cluster from a service provisioned inside a kubernetes cluster

My setup is like, I have a producer service provisioned as part of minikube cluster and it is trying to publish messages to a kafka instance running on the host machine.
I have written a kafka service and endpoints yaml as follows:
kind: Service
apiVersion: v1
metadata:
name: kafka
spec:
ports:
- name: "broker"
protocol: "TCP"
port: 9092
targetPort: 9092
nodePort: 0
---
kind: Endpoints
apiVersion: v1
metadata:
name: kafka
namespace: default
subsets:
- addresses:
- ip: 10.0.2.2
ports:
- name: "broker"
port: 9092
The ip address of the host machine from inside the minikube cluster mentioned in the endpoint is acquired from the following command:
minikube ssh "route -n | grep ^0.0.0.0 | awk '{ print \$2 }'"
The problem I am facing is that the topic is getting created when producer tries to publish message for the first time but no messages are getting written on to that topic.
Digging into the pod logs, I found that producer is trying to connect to kafka instance on localhost or something (not really sure of it).
2020-05-17T19:09:43.021Z [warn] org.apache.kafka.clients.NetworkClient [] -
[Producer clientId=/system/sharding/kafkaProducer-greetings/singleton/singleton/producer]
Connection to node 0 (omkara/127.0.1.1:9092) could not be established. Broker may not be available.
Following which I suspected that probably I need to modify server.properties with the following change:
listeners=PLAINTEXT://localhost:9092
This however resulted in the change in the ip address in the log:
2020-05-17T19:09:43.021Z [warn] org.apache.kafka.clients.NetworkClient [] -
[Producer clientId=/system/sharding/kafkaProducer-greetings/singleton/singleton/producer]
Connection to node 0 (omkara/127.0.0.1:9092) could not be established. Broker may not be available.
I am not sure what ip address must be mentioned here? Or what is an alternate solution? And if it is even possible to connect from inside the kubernetes cluster to the kafka instance installed outside the kubernetes cluster.
Since producer kafka client is on the very same network as the brokers, we need to configure an additional listener like so:
listeners=INTERNAL://0.0.0.0:9093,EXTERNAL://0.0.0.0:9092
listener.security.protocol.map=INTERNAL:PLAINTEXT,EXTERNAL:PLAINTEXT
advertised.listeners=INTERNAL://localhost:9093,EXTERNAL://10.0.2.2:9092
inter.broker.listener.name=INTERNAL
We can verify messages in topic like so:
kafka-console-consumer.sh --bootstrap-server INTERNAL://0.0.0.0:9093 --topic greetings --from-beginning
{"name":"Alice","message":"Namastey"}
You can find a detailed explaination on understanding & provisioning kafka listeners here.

strimzi 0.14: Kafka broker failed authentication

I tried to build a Kafka cluster using Strimzi (0.14) in a Kubernetes cluster.
I use the examples come with the strimzi, i.e. examples/kafka/kafka-persistent.yaml.
This yaml file looks like this:
apiVersion: kafka.strimzi.io/v1beta1
kind: Kafka
metadata:
name: my-cluster
spec:
kafka:
version: 2.3.0
replicas: 3
listeners:
plain: {}
tls: {}
config:
offsets.topic.replication.factor: 3
transaction.state.log.replication.factor: 3
transaction.state.log.min.isr: 2
log.message.format.version: "2.3"
storage:
type: jbod
volumes:
- id: 0
type: persistent-claim
size: 12Gi
deleteClaim: false
zookeeper:
replicas: 3
storage:
type: persistent-claim
size: 9Gi
deleteClaim: false
entityOperator:
topicOperator: {}
userOperator: {}
kubectl apply -f examples/kafka/kafka-persistent.yaml
Both zookeepers and kafka brokers were brought up.
However, I saw errors in kafka broker logs:
[SocketServer brokerId=1] Failed authentication with /10.244.5.94 (SSL handshake failed) (org.apache.kafka.common.network.Selector) [data-plane-kafka-network-thread-1-ListenerName(REPLICATION)-SSL-0]
Anyone know how to fix the problem?
One of the things which can cause this is if your cluster is using different DNS suffix for service domains (default is .cluster.local). You need to find out the right DNS suffix and use the environment variable KUBERNETES_SERVICE_DNS_DOMAIN in the Strimzi Cluster Operator deployment to override the default value.
If you exec into one of the Kafka or Zookeeper pods and do hostname -f it should show you the full hostname from which you can identifythe suffix.
(I pasted this from comments as a full answer since it helped to solve the question.)

How to pass down kubernetes service endpoints IP into KAFKA advertised listener

I would like to configure Kafka broker in Kubernetes. The docker image I am using is confluentinc/cp-kafka:latest. It requires KAFKA_ADVERTISED_LISTENERS environment variable which allows Kafka client to communicate with broker.
The problem is the difficulty to assign service endpoints IP to KAFKA_ADVERTISED_LISTENERS. If I am using localhost as this value, it is only working in local Kafka broker pod but it won't work for some Kafka client pods in kubernetes cluster to communicate with it. If I am using the service endpoint IP coming from kubectl get endpoints -l app=kafka, this is working but it is little overhead to use some audit script set this dynamic value every time.
I wonder is there a better way I can dynamically set this value in Kubernetes yaml file, so I don't need to programatically set this IP every time.
Here is the yaml file:
---
apiVersion: v1
kind: Service
metadata:
name: kafka-broker
labels:
app: kafka
spec:
type: NodePort
ports:
- port: 9092
targetPort: 9092
selector:
app: kafka
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: kafka-broker
spec:
replicas: 1
selector:
matchLabels:
app: kafka
template:
metadata:
labels:
app: kafka
spec:
hostname: broker
containers:
- name: kafka
image: confluentinc/cp-kafka:latest
ports:
- containerPort: 9092
env:
- name: KAFKA_ADVERTISED_LISTENERS
value: "PLAINTEXT://DYNAMIC_ENDPOINT_IP:9092"
- name: KAFKA_OFFSETS_TOPIC_REPLICATION_FACTOR
value: "1"
- name: KAFKA_ZOOKEEPER_CONNECT
value: zoo1:2181
Thanks in advance.
Edit: I tried to use Server name, Service host environment variable, service source IP and Pod IP. Unfortunately, I still get the error in kafka log:
java.lang.IllegalArgumentException: Error creating broker listeners from 'PLAINTEXT://$KAFKA_BROKER_SERVICE_HOST:9092': Unable to parse PLAINTEXT://$KAFKA_BROKER_SERVICE_HOST:9092 to a broker endpoint
If I am using kubectl exec -it kafa-broker-ssfjks env, those environment variables are actually set correctly in this pod. I guess it may be related to a Kafka broker configuration issue ?
You should let your clients connect through the service, so exposing the ip or dns of the service should work. By default services are exposed as variable names in the pod. If a dns plugin is configured dns can be used. More info: https://kubernetes.io/docs/concepts/services-networking/service/#environment-variables
Use the service name (kafka-broker) instead of it's IP. Kube-dns will resolve it for you. If the kafka client was placed at same namespace, you should use just "kafka-broker", if not, you must use the qualified name "kafka-broker.YOURNAMESPACE.svc"
#Jakub got me on the right track, so for something like cp-kafka-connect my Dockerfile looks like:
FROM confluentinc/cp-kafka-connect:5.4.0
ENV CONNECT_GROUP_ID='kafkatosql'
ENV CONNECT_CONFIG_STORAGE_TOPIC="kafkatosql-config"
ENV CONNECT_OFFSET_STORAGE_TOPIC="kafkatosql-offsets"
ENV CONNECT_STATUS_STORAGE_TOPIC="kafkatosql-status"
ENV CONNECT_KEY_CONVERTER="io.confluent.connect.avro.AvroConverter"
ENV CONNECT_VALUE_CONVERTER="io.confluent.connect.avro.AvroConverter"
ENV CONNECT_LOG4J_ROOT_LOGLEVEL="ERROR"
ENV CONNECT_PLUGIN_PATH="/usr/share/java,/usr/share/confluent-hub-components"
RUN confluent-hub install --no-prompt confluentinc/kafka-connect-jdbc:5.4.0
WORKDIR /app
COPY start.sh .
CMD exec ./start.sh
and then start.sh looks like:
kafka_connect_host=localhost:8083
export CONNECT_REST_ADVERTISED_HOST_NAME=$(hostname -I)
/etc/confluent/docker/run &
wait_counter=0
echo "Waiting for Kafka Connect to start listening on kafka-connect ⏳"
while true; do
status=$(curl -s -o /dev/null -w %{http_code} http://$kafka_connect_host/connectors)
if [ $status -eq 000 ]; then
wait_counter=$((wait_counter+1))
echo "Kafka Connect listener HTTP status: $status (waiting for 200)"
if [ $wait_counter = 15 ]; then
echo 'Waited too long!';
exit 1;
else
echo "Retries: $wait_counter"
sleep 3
fi
else
break
fi
done
echo -e "\n--\n+> Creating Kafka Connect Postgresql Sink"
curl -X PUT http://$kafka_connect_host/connectors/jdbc_sink_postgresql_00/config -H "Content-Type: application/json" -d '{
"connector.class": "io.confluent.connect.jdbc.JdbcSinkConnector",
"tasks.max": 1,
"topics": "users",
"connection.url": "jdbc:'"$DB_URL"'",
"auto.create": false
}'
# ... other stuff
trap : TERM INT; sleep infinity & wait