https://pulsar.apache.org/ logo
Join Slack
Powered by
# kubernetes
  • g

    Gergely Fábián

    04/04/2024, 8:31 PM
    What would be the best way to configure Pulsar running on k8s with georeplication between multiple clusters (N>=3) in a mesh topology? What level of automation is available for this?
  • f

    Fatih

    04/18/2024, 10:42 AM
    Hello, i got some log errors in pods pulsar-proxy :
    Copy code
    08:18:38.703 [pulsar-proxy-io-2-1] WARN org.apache.pulsar.proxy.server.ProxyConnection - [/10.42.4.0:59042] Got exception NativeIoException : readAddress(..) failed: Connection reset by peer null
    2024-04-18 10:18:38.703 +02:00	pulsar-proxy-1
    io.netty.channel.unix.Errors$NativeIoException: readAddress(..) failed: Connection reset by peer
    2024-04-18 10:18:37.869 +02:00	pulsar-proxy-3
    io.netty.channel.unix.Errors$NativeIoException: readAddress(..) failed: Connection reset by peer
    2024-04-18 10:18:37.869 +02:00	pulsar-proxy-3
    08:18:37.869 [pulsar-proxy-io-2-1] WARN org.apache.pulsar.proxy.server.ProxyConnection - [/10.42.4.0:1895] Got exception NativeIoException : readAddress(..) failed: Connection reset by peer null
    2024-04-18 10:18:37.569 +02:00	pulsar-proxy-3
    08:18:37.569 [pulsar-proxy-io-2-1] WARN org.apache.pulsar.proxy.server.ProxyConnection - [/10.42.4.0:1443] Got exception NativeIoException : readAddress(..) failed: Connection reset by peer null
    2024-04-18 10:18:37.569 +02:00	pulsar-proxy-3
    io.netty.channel.unix.Errors$NativeIoException: readAddress(..) failed: Connection reset by peer
    2024-04-18 10:18:37.429 +02:00	pulsar-proxy-2
    io.netty.channel.unix.Errors$NativeIoException: readAddress(..) failed: Connection reset by peer
    2024-04-18 10:18:37.429 +02:00	pulsar-proxy-2
    08:18:37.429 [pulsar-proxy-io-2-1] WARN org.apache.pulsar.proxy.server.ProxyConnection - [/10.42.4.0:21857] Got exception NativeIoException : readAddress(..) failed: Connection reset by peer null
    2024-04-18 10:18:25.800 +02:00	pulsar-proxy-2
    io.netty.channel.unix.Errors$NativeIoException: readAddress(..) failed: Connection reset by peer
    2024-04-18 10:18:25.800 +02:00	pulsar-proxy-2
    08:18:25.079 [pulsar-proxy-io-2-1] WARN org.apache.pulsar.proxy.server.ProxyConnection - [/10.42.12.0:44329] Got exception NativeIoException : readAddress(..) failed: Connection reset by peer null
    2024-04-18 10:18:25.662 +02:00	pulsar-proxy-0
    io.netty.channel.unix.Errors$NativeIoException: readAddress(..) failed: Connection reset by peer
    2024-04-18 10:18:25.662 +02:00	pulsar-proxy-0
    08:18:25.662 [pulsar-proxy-io-2-1] WARN org.apache.pulsar.proxy.server.ProxyConnection - [/10.42.12.0:31666] Got exception NativeIoException : readAddress(..) failed: Connection reset by peer null
    l
    • 2
    • 6
  • l

    Lari Hotari

    04/20/2024, 2:02 PM
    https://apache-pulsar.slack.com/archives/C5Z4T36F7/p1713621716090769
    🎉 2
  • s

    Slackbot

    08/17/2021, 7:04 AM
    This message was deleted.
    l
    j
    +3
    • 6
    • 31
  • l

    Lari Hotari

    05/18/2024, 11:36 AM
    Apache Pulsar Helm Chart release 3.4.1-candidate-1 is available for testing and validation: https://lists.apache.org/thread/4stgyrtsnj9jhrvw0b1t8bqr4knj8bfx
  • l

    Lari Hotari

    06/04/2024, 4:54 PM
    FYI https://apache-pulsar.slack.com/archives/C5Z4T36F7/p1717520028345449
    🎉 3
  • e

    Enrique

    06/11/2024, 7:02 PM
    Hi, listened to the instructions pinned in #C5Z4T36F7 and submitted my question here: https://github.com/apache/pulsar/discussions/22893 but incase anyone here has an insight - my pulsar manager ui - originally set up with init script and helm install - seems to have restarted/lost its config. I started getting incorrect user/pass and by creating a new user via curl command and manually creating the environment with k8s
    <cluster>.svc.cluser.local
    urls, I was able to begin seeing my topics again. Is there a reason the manager ui didn't recover with previous user + config?
    l
    • 2
    • 1
  • d

    Daniil

    06/12/2024, 7:36 AM
    Hey guys, could somebody give me a hint how to disable writing to a journal with helm charts? I don't see how I can use custom configs there during deployment
  • s

    sagar singh

    06/12/2024, 11:41 AM
    @here I am facing an issue while setting up Apache Pulsar on the IPV6 cluster, ERROR broker is not coming up, shows socket closed when connecting to zookeeper. Only Broker and Proxy are not able to start with IPV6, rest components are working with -Djava.net.preferIPv4Stack=false. Can anyone help me out on this?
    l
    • 2
    • 1
  • l

    Lari Hotari

    08/01/2024, 7:58 PM
    https://apache-pulsar.slack.com/archives/C5ZSVEN4E/p1722542304929519
  • l

    Lari Hotari

    08/06/2024, 12:24 PM
    https://apache-pulsar.slack.com/archives/C5Z4T36F7/p1722947023420459
  • k

    Khemendra

    08/22/2024, 5:54 PM
    Small Suggestion : for a production setup , running pulsar pods on AWS EKS clustter , how many pods, of broker , zookeeper , bookkeeper and proxy is best suited, like what is the configuration, if we have 3 worker nodes (3 ec2 instances) ?
    m
    • 2
    • 1
  • k

    Khemendra

    08/31/2024, 12:01 PM
    Can someone please advise on the closest compatible Kubernetes version less than 1.30 for Apache Pulsar 2.9.4?
    d
    l
    • 3
    • 3
  • l

    László Ludman

    09/02/2024, 12:02 PM
    Hi everyone, I wanted to reach out for some help. I've set up an Apache Pulsar Cluster in a GKE environment using the official Apache Pulsar Helm Chart. The cluster works perfectly until there's an automated node upgrade, after which the brokers don't recover. Unfortunately, I'm still very new to Pulsar, so I'd like to ask for your assistance in identifying the problem. How can I ensure that the Pulsar cluster can survive a node upgrade?
    ✅ 1
    l
    • 2
    • 47
  • l

    László Ludman

    10/21/2024, 9:28 AM
    Hi all, Can i configure the brokers to make the system topics persistent? Now i'm using KAAP with StreamNative function mesh and StreamNative Pulsar Resources Operator. And after i restarting the brokers or there is a gke node upgrade i got this error:
    Caused by: javax.ws.rs.ClientErrorException: HTTP 412 {"reason":"Can't find owner for topic <non-persistent://pulsar/system/loadbalancer-broker-load-data>"}
    After i manually scale down the brokers and scale back the issue is gone. But i don't want to do that 🙂 🙂 Thanks in advance for your help.
  • l

    Lari Hotari

    10/29/2024, 10:45 PM
    Pulsar 4.0.0 support in Apache Pulsar Helm chart is in voting phase: https://apache-pulsar.slack.com/archives/C5ZSVEN4E/p1730241804314059
  • l

    Lari Hotari

    10/31/2024, 11:27 AM
    Apache Pulsar Helm chart version 3.7.0 with support for Pulsar 4.0.0 has been released: https://apache-pulsar.slack.com/archives/C5Z4T36F7/p1730374036144389
  • a

    Ante Malenica

    11/20/2024, 12:25 PM
    Hi all, I am trying to setup pulsar manager with tls and jwt enabled. Using helm chart 3.7.0. Installation goes well and when i login it recognizes my environment but it seems it lacks permissions to create/delete tenants or namespaces, also cant see details of any namespace except topic names. Does anyone knows what can be problem. Also tried installation without jwt and it was working as expected
    a
    • 2
    • 5
  • l

    Lari Hotari

    11/22/2024, 2:25 PM
    FYI https://apache-pulsar.slack.com/archives/C5Z4T36F7/p1732285537246569
  • l

    Lari Hotari

    12/10/2024, 5:52 PM
    FYI https://apache-pulsar.slack.com/archives/C5Z4T36F7/p1733853168684129
    🦜 1
    🎉 1
  • t

    Thomas O'Neill

    01/29/2025, 2:00 PM
    Is anyone using datadog to collect metrics and logs and can share their annotations used to enable the
    puslar
    integration?
  • s

    Simon

    02/28/2025, 12:46 PM
    Hi guys, i'm operating a pulsar cluster set up via latest helm charts in a k8s env. I want to switch the logging format to json, is there an easy way to achieve this? Didn't found anything related in the docs.
    m
    • 2
    • 6
  • l

    Lari Hotari

    03/04/2025, 9:28 PM
    FYI https://apache-pulsar.slack.com/archives/C5ZSVEN4E/p1741123685060969
  • l

    Lari Hotari

    03/06/2025, 1:14 PM
    FYI https://apache-pulsar.slack.com/archives/C5ZSVEN4E/p1741266873834719
  • l

    Lari Hotari

    03/11/2025, 12:58 AM
    4.0.0-candidate-3 is current in release voting stage. Notable improvements to monitoring.
    Enhanced Monitoring Experience: Usable Metrics and Grafana Dashboards pre-installed
    • Pre-configured and thoroughly tested dashboards now included
    • Dashboards maintained in dedicated repository: https://github.com/lhotari/pulsar-grafana-dashboards
    • Sourced from various Apache 2.0 licensed projects, then optimized and updated for Pulsar 4.0.0 compatibility
    • Resolves long-standing dashboard compatibility issues with recent Pulsar and Bookkeeper
    ◦ Many metrics names have changed since Pulsar 2.10 and previously available dashboards were broken
    • While testing, some bugs were faced in Prometheus 3.2.1 were metrics scraping failed without proper errors even with debug log level and that triggered a switch from kube-prometheus-stack to victoriametrics-k8s-stack and VictoriaMetrics.
    ◦ VictoriaMetrics is more efficient than Prometheus and it doesn't have issues with Pulsar metrics.
  • l

    Lari Hotari

    03/14/2025, 8:05 AM
    FYI https://apache-pulsar.slack.com/archives/C5Z4T36F7/p1741939461803199
  • l

    Lari Hotari

    05/28/2025, 12:49 PM
    https://apache-pulsar.slack.com/archives/C5Z4T36F7/p1748378053052519
    🎉 2
  • y

    Yabin m

    06/26/2025, 4:18 PM
    @Yabin m has left the channel
  • g

    Gary

    07/08/2025, 6:28 PM
    Hi, has anyone had any luck configuring the Pulsar Helm chart to authenticate with mTLS in Kubernetes? I've set some broker and proxy configs and clients are able to connect by providing certs, but I am having some issues creating Functions. With mTLS enabled, the function pods spin up without creation of secrets and certs. I couldn't seem to pinpoint where the secrets and certs were being generated, I assume that I'd have to modify some of the templates on the helm chart? https://github.com/apache/pulsar-helm-chart Pulsar 4.0.5 Would appreciate some direction. Thanks!
    m
    • 2
    • 1
  • g

    Gary

    07/17/2025, 3:53 AM
    Hi, has anyone setup mTLS and token auth on the broker before? I'm looking for examples or guidance on setting up mTLS for external clients and token auth for internal Pulsar components. I'm using the https://github.com/apache/pulsar-helm-chart to deploy to K8s. I've set the auth providers like below.
    Copy code
    authenticationProviders: org.apache.pulsar.broker.authentication.AuthenticationProviderToken,org.apache.pulsar.broker.authentication.AuthenticationProviderTLS
    However, I am running into an issue where the
    KubernetesSecretsTokenAuthProvider
    cannot find the token that is already mounted on the broker when the embedded function worker is spinning up a separate function pod. When I had the JWT auth enabled and set, the
    pf-secret-<id> (ca.pem and TOKEN)
    , the volume, and volume mount were automatically created as part of the creation of the function. As I understood it, the embedded function worker on the broker inherits the permission of the broker and should have access to create those secrets. The broker actually shows logs of the function pod attempting connection with both the token and the TLS certs. Both methods fail because there is no secret mounting. Thanks in advance!