[DOC-1120] Upgrade K8s deployment #29837
Conversation
ddhodge
requested review from
aishwarya24,
amannijhawan,
ddorian,
hari90 and
vipul-yb
✅ Deploy Preview for infallible-bardeen-164bc9 ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
|
|
||
| ## Example deployment | ||
|
|
||
| The universe is deployed across three distinct zones in us-west1, specifically us-west1-a, us-west1-b, and us-west1-c, with one Master and one TServer provisioned in each zone using the following Helm command. |
There was a problem hiding this comment.
| The universe is deployed across three distinct zones in us-west1, specifically us-west1-a, us-west1-b, and us-west1-c, with one Master and one TServer provisioned in each zone using the following Helm command. | |
| The universe is deployed across three distinct zones in us-west1, specifically us-west1-a, us-west1-b, and us-west1-c, with one Master and one TServer provisioned in each zone using the following Helm commands. |
| --- | ||
|
|
||
| The following steps assume a multi-zone Kubernetes cluster, deployed as described in [Deploy on Google Kubernetes Engine](../../deploy/kubernetes/multi-zone/gke/helm-chart/), but you can adapt these steps to upgrade other types of deployments. | ||
|
|
There was a problem hiding this comment.
So can all the steps on this page be used for both GKE-specific or general Kubernetes? Or just specific sections?
There was a problem hiding this comment.
@amannijhawan some questions can you TAL
| --wait | ||
| ``` | ||
|
|
||
| To prevent software upgrades on a TServer during a Master upgrade, configure the partition.tserver setting to equal the number of TServers in that specific zone. |
There was a problem hiding this comment.
Unclear what partition.tserver=1 is actually doing in the example. Why would you require to provide this setting if TServers should not be upgraded?
| --wait | ||
| ``` | ||
|
|
||
| After upgrading all the YB-Master and YB-TServer processes, monitor the cluster to ensure it is healthy. Make sure workloads are running as expected and there are no errors in the logs. |
There was a problem hiding this comment.
Is there a command to check if the masters and tservers are running fine after the upgrades?
|
|
||
| ```sh | ||
| helm upgrade yb-demo-us-west1-a yugabytedb/yugabyte \ | ||
| --version 2024.2.3 \ |
There was a problem hiding this comment.
the upgrade version in the example is 2024.2.3. Line 154: states upgrading from 2024.2.3.3-b4 to 2025.1.1.2-b3.
|
|
||
| To do this: | ||
|
|
||
| 1. Copy the new software tar package into a TServer pod temporarily. This allows the pg_upgrade binary in the new software to access the data directory. |
There was a problem hiding this comment.
where to get the package from... how and which pod to copy (kubectl cp)?
| --wait | ||
| ``` | ||
|
|
||
| 1. If auth is enabled on the cluster, do the following: |
There was a problem hiding this comment.
When is authentication required vs optional ?
Co-authored-by: Aishwarya Chakravarthy <[email protected]>
3 participants
@netlify /stable/manage/upgrade-deployment/