We need an end to end procedure that covers:

• Minor upgrades (e.g. 3.0 -> 3.1)
• Patch upgrades (e.g. 3.0.0 -> 3.0.1)

updated draft content:

Existing doc page (covers only operator upgrade - via inplace/canary): https://docs.redhat.com/en/documentation/red_hat_openshift_service_mesh/3.0/html/updating/ossm-updating-openshift-service-mesh#ossm-updating-openshift-service-mesh.

Drafted complete documentation (covers full upgrade path with explanations): https://github.com/openshift-service-mesh/sail-operator/tree/main/docs/ossm/versioning-and-upgrades

There will be some overlap between the above two documents that may need to be aligned. The first doc page is more polished, but is missing a lot of content about updating the operator and version channels. The second page includes the full content, but hte operator upgrade procedure may be less polished than the published material in the first link. 

Old description:

This exists already, but only covers updating the Istio control plane (which is only part of the procedure): https://docs.openshift.com/service-mesh/latest/update/ossm-updating-openshift-service-mesh.html

From end to end, this page should include:

• Understanding versioning (content for this and the next two sections is drafted here)
• Understanding Service Mesh & Istio versions (We only support the latest Istio release)
• Operator Updates & Channels (Note: No update procedure for the operator if they selected 'automatic', but we still need to explain it)
• About Istio control plane update strategies (not to be confused with operator update strategies)
• Procedure: Updating the Istio control plane with InPlace strategy (existing content, upstream content)
• Procedure: Updating the Istio control plane with RevisionBased strategy (existing content, upstream content)
• Procedure: Updating the Istio control plane with RevisionBased strategy using IstioRevisionTag (upstream content)
• Procedure: Updating the Istio CNI daemonset (content drafted here)
• Procedure: Updating the data plane proxies (if not already covered) (Covered by the update strategies explanation)

Note: Existing content should be reviewed against the upstream content as well, as it may have changed. For example, I am not sure why we document "attaching workloads to the control plane" for InPlace upgrades, as that should likely not be necessary.

Anything else I'm missing?