aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
...
* | | Merge pull request #81 from Yolean/kafka-readiness-tcpsolsson2017-11-091-5/+3
|\ \ \ | | | | | | | | Use Kubernetes native TCP readiness probe for kafka
| * | | Actually, for readiness not liveness, let's go offline fast ...kafka-readiness-tcpStaffan Olsson2017-11-071-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 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.
| * | | Timeout was 1s with the old /bin/sh exec, now it's interval -1 sStaffan Olsson2017-11-071-1/+1
| | | |
| * | | More lightweight kafka probe, essentially the same thingStaffan Olsson2017-10-161-5/+3
| | | |
* | | | Merge pull request #85 from Yolean/storage-classes-1.8solsson2017-11-097-9/+37
|\ \ \ \ | | | | | | | | | | Use k8s 1.8 storage class syntax and set Retain by default
| * | | | I don't know if Retain makes any difference in minikube,storage-classes-1.8Staffan Olsson2017-11-092-0/+2
| | | | | | | | | | | | | | | | | | | | but it's a sensible recommendation for storage classes.
| * | | | The configure/ folder is meant as a collection of examples,Staffan Olsson2017-11-093-1/+1
| | | | | | | | | | | | | | | | | | | | for the choices you'll have to make for any kafka cluster.
| * | | | Uses the new storage class field, requires 1.8Staffan Olsson2017-11-095-12/+4
| | | | |
| * | | | Suggests a single configuration/, as a collection of examplesStaffan Olsson2017-11-093-1/+1
| | | | |
| * | | | Adds storage class for kafka brokersStaffan Olsson2017-11-093-0/+14
| | | | |
| * | | | Adds storage class for Zookeeper, with sample manifestsStaffan Olsson2017-11-094-1/+21
|/ / / / | | | | | | | | | | | | This reverts commit efb1019fd9881f316e0ee355cbafc26e3d146be2.
* | | | Merge pull request #83 from Yolean/yahoo-kafka-managersolsson2017-11-092-0/+38
|\ \ \ \ | |_|/ / |/| | | Adds yahoo/kafka-manger
| * | | Adds Yahoo/kafka-manger, all defaults, and you need to "Add cluster"yahoo-kafka-managerStaffan Olsson2017-10-212-0/+38
| |/ /
* | | Merge pull request #52 from Yolean/addon-bootstrap-servicesolsson2017-11-071-0/+11
|\ \ \ | | | | | | | | Addon: bootstrap.kafka service with current brokers as endpoint
| * | | Revert "Evaluates bootstrap; kafkacat ERRs on missing brokers ..."addon-bootstrap-serviceStaffan Olsson2017-11-071-2/+1
| | | | | | | | | | | | | | | | This reverts commit fabb292ba83821cfa7d5f5a0c8cf54a1bc33d454.
| * | | Evaluates bootstrap; kafkacat ERRs on missing brokers ...Staffan Olsson2017-08-091-1/+2
| | | | | | | | | | | | | | | | so this solves the problem with maintaining the BOOTSTRAP string
| * | | Adds a service to use as "bootstrap" servers url ...Staffan Olsson2017-08-091-0/+11
| | | | | | | | | | | | | | | | | | | | i.e. to connect to any kafka broker and get a list of actual broker DNS names to talk to.
* | | | Merge pull request #78 from Yolean/outside-servicessolsson2017-11-075-1/+71
|\ \ \ \ | | | | | | | | | | Outside (i.e. cluster-external) access through per-broker Service
| * | | | Adds port names as documentation in kafka manifestoutside-servicesStaffan Olsson2017-10-251-2/+4
| | | | |
| * | | | Keeps internal listener first in protocol map, as ...Staffan Olsson2017-10-251-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | documented in https://cwiki.apache.org/confluence/display/KAFKA/KIP-103%3A+Separation+of+Internal+and+External+traffic#KIP-103:SeparationofInternalandExternaltraffic-Compatibility,Deprecation,andMigrationPlan "ZooKeeper-based consumers will use the first listener with PLAINTEXT as the security protocol, so listener ordering is important in such cases." Confirmed this using kafka-manager (#83)
| * | | | Revert "Uses pod name for broker id"; keep doing like before and like zooStaffan Olsson2017-10-151-1/+1
| | | | | | | | | | | | | | | | | | | | This reverts commit dcdac0fb36a4476f494af419527ba42f279c485b.
| * | | | Exemplifies external access using host lookup that works with minikubeStaffan Olsson2017-10-155-0/+58
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Listener ports from https://hub.docker.com/r/wurstmeister/kafka/ Config based on KIP-103
| * | | | Labels each statefulset pod with the index, to allow individual servicesStaffan Olsson2017-10-151-0/+2
| | | | |
| * | | | Uses pod name for broker id, instead of the host=pod assumptionStaffan Olsson2017-10-151-1/+1
| | | | |
| * | | | Gives init container access to pod name and namespaceStaffan Olsson2017-10-151-0/+8
| | |/ / | |/| |
* | | | Changes replication-factor for test topics to 2 ...Staffan Olsson2017-11-052-3/+26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | because 1 is useless, as even a normal reschedule of the broker pod that happens to own the replica+partition will disrupt clients. Testing with replication-factor=2 but using 3 for production topics is sensible because we can test with 3 brokers how replication performs.
* | | | Runs Kafka 1.0.0v2.1.0Staffan Olsson2017-11-016-10/+7
|/ / /
* | | kafkacat, too, based on debian:stretch-slim 20171009Staffan Olsson2017-10-151-1/+1
| | |
* | | Shares debian with kafka-initutils -> faster first startStaffan Olsson2017-10-155-7/+7
| | |
* | | initutils based on latest debian slimStaffan Olsson2017-10-151-1/+1
| | |
* | | Some ideas on how to keep this repo simpleStaffan Olsson2017-10-151-1/+7
| | |
* | | Merge pull request #74 from elm-/mastersolsson2017-10-142-2/+2
|\ \ \ | | | | | | | | - fix #65: wait for reply
| * | | - fix #65: wait for replyElmar Weber2017-10-132-2/+2
|/ / /
* | | Merge pull request #71 from kmova/patch-1solsson2017-10-061-1/+1
|\ \ \ | | | | | | | | _nit : Kubernets -> Kubernetes
| * | | _nit : Kubernets -> KubernetesKiran Mova2017-10-061-1/+1
|/ / /
* | | ./update-kafka-image.sh to 0.11.0.1Staffan Olsson2017-10-025-7/+7
| | |
* | | Merge pull request #62 from kahwai1984/patch-1solsson2017-08-161-1/+1
|\ \ \ | |/ / |/| | Update README.md
| * | Update README.mdkahwai19842017-08-151-1/+1
|/ / | | | | fix typo
* | Merge branch 'persistent-volumes-retain', the readme ...v2.0.0Staffan Olsson2017-08-081-4/+14
|\ \ | | | | | | | | | part that recommends manual intervention
| * | Still shortpersistent-volumes-retainStaffan Olsson2017-08-081-4/+6
| | |
| * | Shorter, possibly clearerStaffan Olsson2017-08-051-5/+5
| | |
| * | The default is Delete, which violates our no-surpises ambitionStaffan Olsson2017-08-051-0/+8
| |/
* | I think Schema Registry + REST Proxy is ok for production nowStaffan Olsson2017-08-081-0/+1
| |
* | Don't look like env var interpolation, when it's notStaffan Olsson2017-08-071-3/+3
| |
* | We prefer Ready:False status instead of restarted pods,Staffan Olsson2017-08-053-13/+1
|/ | | | | | at least for now, as it allows exec into the pods to investigate. We've been having frequent restarts that are not due to OOMKilled (i.e. not #49). Now failed probes will lead to unready pods, which we can monitor for using #60.
* got mergedStaffan Olsson2017-08-051-1/+0
|
* Merge pull request #41 from Yolean/multizone-rack-awarenesssolsson2017-08-053-1/+58
|\ | | | | Multizone rack awareness
| * Adds RBAC policy for kubectl to look up node's zonemultizone-rack-awarenessStaffan Olsson2017-08-051-0/+37
| |
| * I suppose most init scripts will do fine with curl+kubectl+bashStaffan Olsson2017-08-051-1/+1
| |
| * Uses an image with kubectl based on the same debian tag as kafkaStaffan Olsson2017-08-051-1/+1
| |
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-09 20:33:36 +0000