Kibana: Upgrade Assistant - Scripting settings - from 5.6 to 6.x

[LukeMathWalker]

Elasticsearch version: 5.6

We are trying to upgrade our cluster from ES 5.6 to ES 6.x (currently hosted on ElasticCloud) and we are using the Upgrade Assistant in the Management section of Kibana.
We have managed to solve all issues highlighted by the Cluster checkup, except for one:

Details: [node[instance-0000000022] used these script-security settings:[script.inline, script.stored, script.file, script.engine.painless.inline, script.engine.expression.inline, script.engine.mustache.inline]]

Looking at the documentation I find the following piece concerning updates for scripting settings:

All of the existing scripting security settings have been removed. Instead they are replaced with script.allowed_types and script.allowed_contexts.
The setting script.max_compilations_per_minute has been replaced with script.max_compilation_rate, which consists of a rate and a time value, for example 75/5m

but it is not clear how to address this compatibility issue. Has anyone been through the same process facing our current problem?

[epixa]

@LukeMathWalker Those messages are being returned from Elasticsearch itself, so the remediation is ultimately to be done there rather than in Kibana. I'm not exactly sure what the best solution is for your use case, but if you reach out to the elasticsearch team on the forums, I'm sure someone there can help you out.

I don't recommend looking for this sort of help on github, as we use github for tracking confirmed bugs and enhancement requests. You should get a much more rapid response on the forums.

[bgeVam]

Hi @LukeMathWalker ,

there is an error in the documentation, it is script.max_compilations_rate, e.g.:

PUT http://localhost:9200/_cluster/settings
{
  "transient": {
    "script.max_compilations_rate": "150/1m"
  }
}

[mckingho]

For those who are using ElasticCloud and blocked to upgrade in console, this critical error is not the reason you are not able to upgrade.
Please check in Upgrade Assistant in Kibana's Management page to see if there is any cluster settings error, then resolve cluster critical issue and check your deployment is upgradable again.
For me, I spent much time trying to change the nodes' script settings. But it turns out I have a cluster critical issue and need to set indices.store.throttle.max_bytes_per_sec to null. It is not needed to change node setting for 5.6 to 6.x upgrade.