kibana/docs/setup/upgrade.asciidoc
Rudolf Meijering d2a5ebd1fc
[docs] Resolving failed Kibana upgrade migrations (#80999)
* Resolving failed Kibana upgrade migrations

* Move warning against rolling upgrades into upgrade-standard and call out stopping all instances in specific upgrade steps

* Add preventing migration failures section

* Add incompatible xpack.tasks.index: .tasks setting to preventing migration failures

* Fix link

Co-authored-by: Kibana Machine <42973632+kibanamachine@users.noreply.github.com>
2020-10-27 17:31:36 +01:00

68 lines
3 KiB
Plaintext

[[upgrade]]
== Upgrade {kib}
Depending on the {kib} version you're upgrading from, the upgrade process to 7.0
varies.
[float]
[[upgrade-before-you-begin]]
=== Before you begin
WARNING: {kib} automatically runs upgrade migrations when required. To roll back to an earlier version in case of an upgrade failure, you **must** have a backup snapshot available. Use <<snapshot-repositories, Snapshot and Restore>> to back up {kib} data by targeting the `.kibana*` indices. For more information see <<upgrade-migrations, upgrade migrations>>.
Before you upgrade {kib}:
* Consult the <<breaking-changes,breaking changes>>.
* Back up your data with <<snapshot-repositories, Snapshot and Restore>>. To roll back to an earlier version, you **must** have a snapshot of the `.kibana*` indices.
* Although not a requirement for rollbacks, we recommend taking a snapshot of all {kib} indices created by the plugins you use such as the `.reporting*` indices created by the reporting plugin.
* Before you upgrade production servers, test the upgrades in a dev environment.
* See <<preventing-migration-failures, preventing migration failures>> for common reasons upgrades fail and how to prevent these.
* If you are using custom plugins, check that a compatible version is
available.
* Shut down all {kib} instances. Running more than one {kib} version against
the same Elasticseach index is unsupported. Upgrading while older {kib}
instances are running can cause data loss or upgrade failures.
To identify the changes you need to make to upgrade, and to enable you to
perform an Elasticsearch rolling upgrade with no downtime, you must upgrade to
6.7 before you upgrade to 7.0.
For a comprehensive overview of the upgrade process, refer to
*{stack-ref}/upgrading-elastic-stack.html[Upgrading the Elastic Stack]*.
[float]
[[upgrade-5x-earlier]]
=== Upgrade from 5.x or earlier
{es} can read indices created in the previous major version. Before you upgrade
to 7.0.0, you must reindex or delete any indices created in 5.x or earlier.
For more information, refer to
{stack-ref}/upgrading-elastic-stack.html#oss-stack-upgrade[Upgrading the Elastic Stack].
When your reindex is complete, follow the <<upgrade-standard, Standard upgrade>>
instructions.
[float]
[[upgrade-6x]]
=== Upgrade from 6.x
The recommended path is to upgrade to 6.8 before upgrading to 7.0. This makes it
easier to identify the required changes, and enables you to use the Upgrade
Assistant to prepare for your upgrade to 7.0.
TIP: The ability to import {kib} 6.x saved searches, visualizations, and
dashboards is supported.
[float]
[[upgrade-67]]
=== Upgrade from 6.8
To help you prepare for your upgrade to 7.0, 6.8 includes an https://www.elastic.co/guide/en/kibana/6.8/upgrade-assistant.html[Upgrade Assistant]
To access the assistant, go to *Management > 7.0 Upgrade Assistant*.
After you have addressed any issues that were identified by the Upgrade
Assistant, <<upgrade-standard,upgrade to 7.0>>.
include::upgrade/upgrade-standard.asciidoc[]
include::upgrade/upgrade-migrations.asciidoc[]