cockroachdb

cockroachdb

19.2.5 CockroachDB is a scalable, survivable, strongly-consistent SQL database.

CockroachDB is a scalable, survivable, strongly-consistent SQL database.

CockroachDB Helm Chart

Documentation

Below is a brief overview of operating the CockroachDB Helm Chart and some specific implementation details. For additional information on deploying CockroachDB, please see:

https://www.cockroachlabs.com/docs/stable/orchestrate-cockroachdb-with-kubernetes.html

Note that the documentation requires Helm 3.0 or higher.

Prerequisites Details

StatefulSet Details

StatefulSet Caveats

Chart Details

This chart will do the following:

  • Set up a dynamically scalable CockroachDB cluster using a Kubernetes StatefulSet.

Installing the Chart

To install the chart with the release name my-release:

helm install my-release <helm-repo>/cockroachdb

Note that for a production cluster, you will likely want to override the following parameters in values.yaml with your own values.

  • statefulset.resources.requests.memory and statefulset.resources.limits.memory allocate memory resources to CockroachDB pods in your cluster.
  • conf.cache and conf.max-sql-memory are memory limits that we recommend setting to 1/4 of the above resource allocation. When running CockroachDB, you must set these limits explicitly to avoid running out of memory.
  • storage.persistentVolume.size defaults to 100Gi of disk space per pod, which you may increase or decrease for your use case.
  • storage.persistentVolume.storageClass uses the default storage class for your environment. We strongly recommend that you specify a storage class which uses an SSD.
  • tls.enabled must be set to yes/true to deploy in secure mode.

For more information on overriding the values.yaml parameters, please see:

https://www.cockroachlabs.com/docs/stable/orchestrate-cockroachdb-with-kubernetes.html#step-2-start-cockroachdb

If you are running in secure mode (with configuration parameter tls.enabled set to yes/true) and tls.certs.provided set to no/false), then you will have to manually approve the cluster's security certificates as the pods are created. You can see the pending CSRs (certificate signing requests) by running kubectl get csr, and approve them by running kubectl certificate approve <csr-name>. You'll have to approve one certificate for each CockroachDB node (e.g., default.node.my-release-cockroachdb-0 and one client certificate for the job that initializes the cluster (e.g., default.node.root).

When tls.certs.provided is set to yes/true, this chart will use certificates created outside of Kubernetes. You may want to use this if you want to use a different certificate authority from the one being used by Kubernetes or if your Kubernetes cluster doesn't fully support certificate-signing requests. To use this, first set up your certificates and load them into your Kubernetes cluster as Secrets using the commands below:

mkdir certs
mkdir my-safe-directory
cockroach cert create-ca --certs-dir=certs --ca-key=my-safe-directory/ca.key
cockroach cert create-client root --certs-dir=certs --ca-key=my-safe-directory/ca.key
kubectl create secret generic cockroachdb-root --from-file=certs
cockroach cert create-node --certs-dir=certs --ca-key=my-safe-directory/ca.key localhost 127.0.0.1 eerie-horse-cockroachdb-public eerie-horse-cockroachdb-public.default eerie-horse-cockroachdb-public.default.svc.cluster.local *.eerie-horse-cockroachdb *.eerie-horse-cockroachdb.default *.eerie-horse-cockroachdb.default.svc.cluster.local
kubectl create secret generic cockroachdb-node --from-file=certs

Set tls.certs.tlsSecret to yes/true if you make use of cert-manager in your cluster.

cert-manager stores generated certificates in dedicated TLS secrets. Thus, they are always named:

  • ca.crt
  • tls.crt
  • tls.key

On the other hand, CockroachDB also demands dedicated certificate filenames:

  • ca.crt
  • node.crt
  • node.key
  • client.root.crt
  • client.root.key

By activating tls.certs.tlsSecret we benefit from projected secrets and convert the TLS secret filenames to their according CockroachDB filenames.

If you are running in secure mode, then you will have to manually approve the cluster's security certificates as the pods are created. You can see the pending CSRs (certificate signing requests) by running kubectl get csr, and approve them by running kubectl certificate approve <csr-name>. You'll have to approve one certificate for each CockroachDB node (e.g., default.node.my-release-cockroachdb-0 and one client certificate for the job that initializes the cluster (e.g., default.node.root).

Confirm that all pods are Running successfully and init has been completed:

kubectl get pods
NAME                                READY     STATUS      RESTARTS   AGE
my-release-cockroachdb-0            1/1       Running     0          1m
my-release-cockroachdb-1            1/1       Running     0          1m
my-release-cockroachdb-2            1/1       Running     0          1m
my-release-cockroachdb-init-k6jcr   0/1       Completed   0          1m

Confirm that persistent volumes are created and claimed for each pod:

kubectl get pv
NAME                                       CAPACITY   ACCESS MODES   RECLAIM POLICY   STATUS    CLAIM                                      STORAGECLASS   REASON    AGE
pvc-64878ebf-f3f0-11e8-ab5b-42010a8e0035   100Gi      RWO            Delete           Bound     default/datadir-my-release-cockroachdb-0   standard                 51s
pvc-64945b4f-f3f0-11e8-ab5b-42010a8e0035   100Gi      RWO            Delete           Bound     default/datadir-my-release-cockroachdb-1   standard                 51s
pvc-649d920d-f3f0-11e8-ab5b-42010a8e0035   100Gi      RWO            Delete           Bound     default/datadir-my-release-cockroachdb-2   standard                 51s

Upgrading the cluster

Chart version 3.0.0 and after

Launch a temporary interactive pod and start the built-in SQL client:

kubectl run cockroachdb --rm -it \
--image=cockroachdb/cockroach \
--restart=Never \
-- sql --insecure --host=my-release-cockroachdb-public

If you are running in secure mode, you will have to provide a client certificate to the cluster in order to authenticate, so the above command will not work. See here for an example of how to set up an interactive SQL shell against a secure cluster or here for an example application connecting to a secure cluster.

Set cluster.preserve_downgrade_option, where $current_version is the CockroachDB version currently running (e.g., 19.2):

> SET CLUSTER SETTING cluster.preserve_downgrade_option = '$current_version';

Exit the shell and delete the temporary pod:

> \q

Kick off the upgrade process by changing the new Docker image, where $new_version is the CockroachDB version to which you are upgrading:

kubectl delete job my-release-cockroachdb-init
helm upgrade my-release stable/cockroachdb \
--set image.tag=$new_version \
--reuse-values

Kubernetes will carry out a safe rolling upgrade of your CockroachDB nodes one-by-one. Monitor the cluster's pods until all have been successfully restarted:

kubectl get pods
NAME                                READY     STATUS              RESTARTS   AGE
my-release-cockroachdb-0            1/1       Running             0          2m
my-release-cockroachdb-1            1/1       Running             0          3m
my-release-cockroachdb-2            1/1       Running             0          3m
my-release-cockroachdb-3            0/1       ContainerCreating   0          25s
my-release-cockroachdb-init-nwjkh   0/1       ContainerCreating   0          6s
kubectl get pods \
-o jsonpath='{range .items[*]}{.metadata.name}{"\t"}{.spec.containers[0].image}{"\n"}'
my-release-cockroachdb-0    cockroachdb/cockroach:v19.2.5
my-release-cockroachdb-1    cockroachdb/cockroach:v19.2.5
my-release-cockroachdb-2    cockroachdb/cockroach:v19.2.5
my-release-cockroachdb-3    cockroachdb/cockroach:v19.2.5

Resume normal operations. Once you are comfortable that the stability and performance of the cluster is what you'd expect post-upgrade, finalize the upgrade:

kubectl run cockroachdb --rm -it \
--image=cockroachdb/cockroach \
--restart=Never \
-- sql --insecure --host=my-release-cockroachdb-public
> RESET CLUSTER SETTING cluster.preserve_downgrade_option;
> \q

Chart versions prior to 3.0.0

Due to a change in the label format in version 3.0.0 of this chart, upgrading requires that you delete the StatefulSet. Luckily there is a way to do it without actually deleting all the resources managed by the StatefulSet. Use the workaround below to upgrade from charts versions previous to 3.0.0:

Get the new labels from the specs rendered by Helm:

helm template -f deploy.vals.yml stable/cockroachdb -x templates/statefulset.yaml \
| yq r - spec.template.metadata.labels
app.kubernetes.io/name: cockroachdb
app.kubernetes.io/instance: my-release
app.kubernetes.io/component: cockroachdb

Place the new labels on all pods of the StatefulSet (change my-release-cockroachdb-0 to the name of each pod):

kubectl label pods my-release-cockroachdb-0 \
app.kubernetes.io/name=cockroachdb \
app.kubernetes.io/instance=my-release \
app.kubernetes.io/component=cockroachdb

Delete the StatefulSet without deleting pods:

kubectl delete statefulset my-release-cockroachdb --cascade=false

Verify that no pod is deleted and then upgrade as normal. A new StatefulSet will be created, taking over the management of the existing pods and upgrading them if needed.

Configuration

The following table lists the configurable parameters of the CockroachDB chart and their default values.
For details see the values.yaml file.

Parameter Description Default
clusterDomain Cluster's default DNS domain cluster.local
conf.attrs CockroachDB node attributes []
conf.cache Size of CockroachDB's in-memory cache 25%
conf.cluster-name Name of CockroachDB cluster ""
conf.disable-cluster-name-verification Disable CockroachDB cluster name verification no
conf.join List of already-existing CockroachDB instances []
conf.max-disk-temp-storage Max storage capacity for temp data 0
conf.max-offset Max allowed clock offset for CockroachDB cluster 500ms
conf.max-sql-memory Max memory to use processing SQL querie 25%
conf.locality Locality attribute for this deployment ""
conf.single-node Disable CockroachDB clustering (standalone mode) no
conf.sql-audit-dir Directory for SQL audit log ""
conf.port CockroachDB primary serving port in Pods 26257
conf.http-port CockroachDB HTTP port in Pods 8080
image.repository Container image name cockroachdb/cockroach
image.tag Container image tag v19.2.5
image.pullPolicy Container pull policy IfNotPresent
image.credentials registry, user and pass credentials to pull private image {}
statefulset.replicas StatefulSet replicas number 3
statefulset.updateStrategy Update strategy for StatefulSet Pods {"type": "RollingUpdate"}
statefulset.podManagementPolicy OrderedReady/Parallel Pods creation/deletion order Parallel
statefulset.budget.maxUnavailable k8s PodDisruptionBudget parameter 1
statefulset.args Extra command-line arguments []
statefulset.env Extra env vars []
statefulset.secretMounts Additional Secrets to mount at cluster members []
statefulset.labels Additional labels of StatefulSet and its Pods {"app.kubernetes.io/component": "cockroachdb"}
statefulset.annotations Additional annotations of StatefulSet Pods {}
statefulset.nodeAffinity Node affinity rules of StatefulSet Pods {}
statefulset.podAffinity Inter-Pod affinity rules of StatefulSet Pods {}
statefulset.podAntiAffinity Anti-affinity rules of StatefulSet Pods auto
statefulset.podAntiAffinity.type Type of auto anti-affinity rules soft
statefulset.podAntiAffinity.weight Weight for soft auto anti-affinity rules 100
statefulset.nodeSelector Node labels for StatefulSet Pods assignment {}
statefulset.tolerations Node taints to tolerate by StatefulSet Pods []
statefulset.resources Resource requests and limits for StatefulSet Pods {}
service.ports.grpc.external.port CockroachDB primary serving port in Services 26257
service.ports.grpc.external.name CockroachDB primary serving port name in Services grpc
service.ports.grpc.internal.port CockroachDB inter-communication port in Services 26257
service.ports.grpc.internal.name CockroachDB inter-communication port name in Services grpc-internal
service.ports.http.port CockroachDB HTTP port in Services 8080
service.ports.http.name CockroachDB HTTP port name in Services http
service.public.type Public Service type ClusterIP
service.public.labels Additional labels of public Service {"app.kubernetes.io/component": "cockroachdb"}
service.public.annotations Additional annotations of public Service {}
service.discovery.labels Additional labels of discovery Service {"app.kubernetes.io/component": "cockroachdb"}
service.discovery.annotations Additional annotations of discovery Service {}
storage.hostPath Absolute path on host to store data ""
storage.persistentVolume.enabled Whether to use PersistentVolume to store data yes
storage.persistentVolume.size PersistentVolume size 100Gi
storage.persistentVolume.storageClass PersistentVolume class ""
storage.persistentVolume.labels Additional labels of PersistentVolumeClaim {}
storage.persistentVolume.annotations Additional annotations of PersistentVolumeClaim {}
init.labels Additional labels of init Job and its Pod {"app.kubernetes.io/component": "init"}
init.annotations Additional labels of the Pod of init Job {}
init.affinity Affinity rules of init Job Pod {}
init.nodeSelector Node labels for init Job Pod assignment {}
init.tolerations Node taints to tolerate by init Job Pod []
init.resources Resource requests and limits for the Pod of init Job {}
tls.enabled Whether to run securely using TLS certificates no
tls.serviceAccount.create Whether to create a new RBAC service account yes
tls.serviceAccount.name Name of RBAC service account to use ""
tls.certs.provided Bring your own certs scenario, i.e certificates are provided no
tls.certs.clientRootSecret If certs are provided, secret name for client root cert cockroachdb-root
tls.certs.nodeSecret If certs are provided, secret name for node cert cockroachdb-node
tls.certs.tlsSecret Own certs are stored in TLS secret no
tls.init.image.repository Image to use for requesting TLS certificates cockroachdb/cockroach-k8s-request-cert
tls.init.image.tag Image tag to use for requesting TLS certificates 0.4
tls.init.image.pullPolicy Requesting TLS certificates container pull policy IfNotPresent
tls.init.image.credentials registry, user and pass credentials to pull private image {}
networkPolicy.enabled Enable NetworkPolicy for CockroachDB's Pods no
networkPolicy.ingress.grpc Whitelist resources to access gRPC port of CockroachDB's Pods []
networkPolicy.ingress.http Whitelist resources to access gRPC port of CockroachDB's Pods []

Override the default parameters using the --set key=value[,key=value] argument to helm install.

Alternatively, a YAML file that specifies custom values for the parameters can be provided while installing the chart. For example:

helm install my-release -f my-values.yaml <helm-repo>/cockroachdb

Tip: You can use the default values.yaml

Deep dive

Connecting to the CockroachDB cluster

Once you've created the cluster, you can start talking to it by connecting to its -public Service. CockroachDB is PostgreSQL wire protocol compatible, so there's a wide variety of supported clients. As an example, we'll open up a SQL shell using CockroachDB's built-in shell and play around with it a bit, like this (likely needing to replace my-release-cockroachdb-public with the name of the -public Service that was created with your installed chart):

kubectl run cockroach-client --rm -it \
--image=cockroachdb/cockroach \
--restart=Never \
-- sql --insecure --host my-release-cockroachdb-public
Waiting for pod default/cockroach-client to be running, status is Pending,
pod ready: false
If you don't see a command prompt, try pressing enter.
root@my-release-cockroachdb-public:26257> SHOW DATABASES;
+--------------------+
|      Database      |
+--------------------+
| information_schema |
| pg_catalog         |
| system             |
+--------------------+
(3 rows)
root@my-release-cockroachdb-public:26257> CREATE DATABASE bank;
CREATE DATABASE
root@my-release-cockroachdb-public:26257> CREATE TABLE bank.accounts (id INT
PRIMARY KEY, balance DECIMAL);
CREATE TABLE
root@my-release-cockroachdb-public:26257> INSERT INTO bank.accounts VALUES
(1234, 10000.50);
INSERT 1
root@my-release-cockroachdb-public:26257> SELECT * FROM bank.accounts;
+------+---------+
|  id  | balance |
+------+---------+
| 1234 | 10000.5 |
+------+---------+
(1 row)
root@my-release-cockroachdb-public:26257> \q
Waiting for pod default/cockroach-client to terminate, status is Running
pod "cockroach-client" deleted

If you are running in secure mode, you will have to provide a client certificate to the cluster in order to authenticate, so the above command will not work. See here for an example of how to set up an interactive SQL shell against a secure cluster or here for an example application connecting to a secure cluster.

Cluster health

Because our pod spec includes regular health checks of the CockroachDB processes, simply running kubectl get pods and looking at the STATUS column is sufficient to determine the health of each instance in the cluster.

If you want more detailed information about the cluster, the best place to look is the Admin UI.

Accessing the Admin UI

If you want to see information about how the cluster is doing, you can try pulling up the CockroachDB Admin UI by port-forwarding from your local machine to one of the pods (replacing my-release-cockroachdb-0 with the name of one of your pods:

kubectl port-forward my-release-cockroachdb-0 8080

You should then be able to access the Admin UI by visiting http://localhost:8080/ in your web browser.

Failover

If any CockroachDB member fails, it is restarted or recreated automatically by the Kubernetes infrastructure, and will re-join the cluster automatically when it comes back up. You can test this scenario by killing any of the CockroachDB pods:

kubectl delete pod my-release-cockroachdb-1
kubectl get pods -l "app.kubernetes.io/instance=my-release,app.kubernetes.io/component=cockroachdb"
NAME                      READY     STATUS        RESTARTS   AGE
my-release-cockroachdb-0  1/1       Running       0          5m
my-release-cockroachdb-2  1/1       Running       0          5m

After a while:

kubectl get pods -l "app.kubernetes.io/instance=my-release,app.kubernetes.io/component=cockroachdb"
NAME                      READY     STATUS        RESTARTS   AGE
my-release-cockroachdb-0  1/1       Running       0          5m
my-release-cockroachdb-1  1/1       Running       0          20s
my-release-cockroachdb-2  1/1       Running       0          5m

You can check the state of re-joining from the new pod's logs:

kubectl logs my-release-cockroachdb-1
[...]
I161028 19:32:09.754026 1 server/node.go:586  [n1] node connected via gossip and
verified as part of cluster {"35ecbc27-3f67-4e7d-9b8f-27c31aae17d6"}
[...]
cockroachdb-0.my-release-cockroachdb.default.svc.cluster.local:26257
build:      beta-20161027-55-gd2d3c7f @ 2016/10/28 19:27:25 (go1.7.3)
admin:      http://0.0.0.0:8080
sql:
postgresql://root@my-release-cockroachdb-1.my-release-cockroachdb.default.svc.cluster.local:26257?sslmode=disable
logs:       cockroach-data/logs
store[0]:   path=cockroach-data
status:     restarted pre-existing node
clusterID:  {35ecbc27-3f67-4e7d-9b8f-27c31aae17d6}
nodeID:     2
[...]

NetworkPolicy

To enable NetworkPolicy for CockroachDB, install a networking plugin that implements the Kubernetes NetworkPolicy spec, and set networkPolicy.enabled to yes/true.

For Kubernetes v1.5 & v1.6, you must also turn on NetworkPolicy by setting the DefaultDeny Namespace annotation. Note: this will enforce policy for all pods in the Namespace:

kubectl annotate namespace default "net.beta.kubernetes.io/network-policy={\"ingress\":{\"isolation\":\"DefaultDeny\"}}"

For more precise policy, set networkPolicy.ingress.grpc and networkPolicy.ingress.http rules. This will only allow pods that match the provided rules to connect to CockroachDB.

Scaling

Scaling should be managed via the helm upgrade command. After resizing your cluster on your cloud environment (e.g., GKE or EKS), run the following command to add a pod. This assumes you scaled from 3 to 4 nodes:

helm upgrade \
my-release \
stable/cockroachdb \
--set statefulset.replicas=4 \
--reuse-values

Note, that if you are running in secure mode (tls.enabled is yes/true) and increase the size of your cluster, you will also have to approve the CSR (certificate-signing request) of each new node (using kubectl get csr and kubectl certificate approve).