Please carefully read the entries for all versions between the version you are running and the version you want to upgrade to.
v0.2.x
to v0.3.x
If you are upgrading from v0.2.x
to v0.3.x
, please follow the Upgrading to Apache guide. This is a special upgrade that requires different instructions.
v_.X._
)In order to upgrade minor versions (e.g. v0.2.5
-> v0.3.0
), you must upgrade one minor version at a time (e.g. v0.2.0
-> v0.3.0
-> v0.4.0
). It is also necessary to upgrade to the latest patch version before upgrading to the next minor version. Therefore if you are running v0.2.5
and you want to upgrade to v0.3.0
, you must first upgrade to v0.2.8
before upgrading to v0.3.0
.
v_._.X
)You should be able to upgrade from a version to any patch version with the same minor and major versions. It is always encouraged to upgrade to the latest patch version of the minor and major version you are running. There is no need to upgrade one patch version at a time (e.g. v0.2.5
-> v0.2.6
-> v0.2.7
-> v0.2.8
), instead you can leap to the latest patch version (e.g. v0.2.5
-> v0.2.8
).
Installing the Solr Operator, especially via the Helm Chart, does not necessarily mean that you are installing the required CRDs for that version of the Solr Operator.
If the Solr CRDs already exist in your Kubernetes cluster, then Helm will not update them even if the CRDs have changed between the Helm chart versions. Instead, you will need to manually install the CRDs whenever upgrading your Solr Operator.
You should always upgrade your CRDs before upgrading the Operator
You can do this via the following command, replacing <version>
with the version of the Solr Operator you are installing:
# Just replace the Solr CRDs kubectl replace -f "http://solr.apache.org/operator/downloads/crds/<version>/all.yaml" # Just replace the Solr CRDs and all CRDs it might depend on (e.g. ZookeeperCluster) kubectl replace -f "http://solr.apache.org/operator/downloads/crds/<version>/all-with-dependencies.yaml"
It is strongly recommended to use kubectl create
or kubectl replace
, instead of kubectl apply
when creating/updating CRDs.
When upgrading the Solr Operator, you may need to upgrade the Zookeeper Operator at the same time. If you are using the Solr Helm chart to deploy the Zookeeper operator, then you won‘t need to do anything besides installing the CRD’s with dependencies, and upgrade the Solr Operator helm deployment.
# Just replace the Solr CRDs and all CRDs it might depend on (e.g. ZookeeperCluster) kubectl replace -f "http://solr.apache.org/operator/downloads/crds/v0.3.0/all-with-dependencies.yaml" helm upgrade solr-operator apache-solr/solr-operator --version 0.3.0
Note that the Helm chart version does not contain a v
prefix, which the downloads version does. The Helm chart version is the only part of the Solr Operator release that does not use the v
prefix.
The deprecated Solr Operator Helm chart option useZkOperator
has been removed, use zookeeper-operator.use
instead.
Note: The old option takes a string "true"
/"false"
, while the new option takes a boolean true
/false
.
The default Solr version for SolrCloud
and SolrPrometheusExporter
resources has been upgraded from 7.7.0
to 8.9
. This will not effect any existing resources, as default versions are hard-written to the resources immediately. Only new resources created after the Solr Operator is upgraded to v0.4.0
will be affected.
In previous versions of the Solr Operator, the provided Zookeeper instances could only use Persistent Storage. Now ephemeral storage is enabled, and used by default if Solr is using ephemeral storage. The ZK storage type can be explicitly set via Spec.zookeeperRef.provided.ephemeral
or Spec.zookeeperRef.provided.persistence
, however if neither is set, the Solr Operator will default to use the type of storage (persistent or ephemeral) that Solr is using.
This means that the default Zookeeper Storage type can change for users using ephemeral storage for Solr. If you require ephemeral Solr storage and persistent Zookeeper Storage, be sure to explicitly set that starting in v0.4.0
.
All deprecated CRD fields and Solr Operator options from v0.2.*
have been removed.
The SolrCollection
and SolrCollectionAlias
have been removed. Please use the Solr APIs to manage these resources instead. Discussion around the removal can be found in Issue #204.
The required version of the Zookeeper Operator to use with this version has been upgraded from v0.2.6
to v0.2.9
. If you use the Solr Operator helm chart, then by default the new version of the Zookeeper Operator will be installed as well. Refer to the helm chart documentation if you want to manage the Zookeeper Operator installation yourself.
Please refer to the Zookeeper Operator release notes before upgrading.
Do to the addition of possible sidecar/initContainers for SolrClouds, the version of CRDs used had to be upgraded to apiextensions.k8s.io/v1
.
This means that Kubernetes support is now limited to 1.16+.
If you are unable to use a newer version of Kubernetes, please install the v0.2.6
version of the Solr Operator for use with Kubernetes 1.15 and below.
The location of backup-restore volume mounts in Solr containers has changed from /var/solr/solr-backup-restore
to /var/solr/data/backup-restore
. This change was made to ensure that there were no issues using the backup API with solr 8.6+, which restricts the locations that backup data can be saved to and read from. This change should be transparent if you are merely using the SolrBackup CRD. All files permissions issues with SolrBackups should now be addressed.
The default PodManagementPolicy
for StatefulSets has been changed to Parallel
from OrderedReady
. This change will not affect existing StatefulSets, as PodManagementPolicy
cannot be updated. In order to continue using OrderedReady
on new SolrClouds, please use the following setting:SolrCloud.spec.customSolrKubeOptions.statefulSetOptions.podManagementPolicy
The SolrCloud
and SolrPrometheusExporter
services' portNames have changed to "solr-client"
and "solr-metrics"
from "ext-solr-client"
and "ext-solr-metrics"
, respectively. This is due to a bug in Kubernetes where portName
and targetPort
must match for services.
Support for etcd
/zetcd
deployments has been removed.
The section for a Zookeeper cluster Spec SolrCloud.spec.zookeeperRef.provided.zookeeper
has been DEPRECATED. The same fields (except for the deprecated persistentVolumeClaimSpec
option) are now available under SolrCloud.spec.zookeeperRef.provided
.
Data Storage options have been expanded, and moved from their old locations.
SolrCloud.spec.dataPvcSpec
has been DEPRECATED.SolrCloud.spec.dataStorage.persistent.pvcTemplate.spec=<spec>
SolrCloud.spec.backupRestoreVolume
has been DEPRECATED.SolrCloud.spec.dataStorage.backupRestoreOptions.Volume=<volume-source>
--ingressBaseDomain
has been DEPRECATED. In order to set the external baseDomain of your clouds, please begin to use SolrCloud.spec.solrAddressability.external.domainName
instead. You will also need to set SolrCloud.spec.solrAddressability.external.method
to Ingress
. The --ingressBaseDomain
argument is backwards compatible, and all existing SolrCloud objects will be auto-updated once your operator is upgraded to v0.2.6
. The argument will be removed in a future version (v0.3.0
).v0.2.6
.SolrCloud.spec.solrPodPolicy
has been DEPRECATED in favor of the SolrCloud.spec.customSolrKubeOptions.podOptions
option.
This option is backwards compatible, but will be removed in a future version (v0.3.0
).
SolrPrometheusExporter.spec.solrPodPolicy
has been DEPRECATED in favor of the SolrPrometheusExporter.spec.customKubeOptions.podOptions
option.
This option is backwards compatible, but will be removed in a future version (v0.3.0
).
ZkCluster.Status
, which used an IP, to using the service name. This will cause a rolling restart of your solrs using the provided zookeeper option, but there will be no data loss.gomod
instead of dep
SolrCloud.spec.zookeeperRef.provided.zookeeper.persistentVolumeClaimSpec
has been DEPRECATED in favor of the SolrCloud.zookeeperRef.provided.zookeeper.persistence
option.v0.3.0
).0.2.4
is required.