Elasticsearch: add fail_on_detecting_es_version_retry_exceed support

[ipiticas]

Q	A
Bug fix?	no
New feature?	yes
API breaks?	no
Deprecations?	no
Related tickets
License	Apache 2.0

What's in this PR?

Add fail_on_detecting_es_version_retry_exceed support for elasticsearch output.

Why?

This field is necessary to prevent Fluend from failing at startup when one or many Elasticsearch hosts are unreachable.

Additional context

Related:

• Fluentd goes down if ES output host doesn't exists #491
• Crash when one of many elasticsearch cluster cannot be connected when fluentd start. uken/fluent-plugin-elasticsearch#475

Checklist

• Code meets the Developer Guide
• User guide and development docs updated (if needed)
• Related Helm chart(s) updated (if needed)

[CLAassistant]

All committers have signed the CLA.

[ahma]

LGTM Thanks @hyperion-hyp

[siliconbrain]

@hyperion-hyp to fix the failing build, you should run make generate and commit the changes

[ipiticas]

@hyperion-hyp to fix the failing build, you should run make generate and commit the changes
... still failing; what did I miss?

[siliconbrain]

@hyperion-hyp seems like you've moved the field from before fail_on_detecting_es_version_retry_exceed to after it sometime after you last ran generate and controller-gen now mirrors this change in the manifest files when the CI build checks for differences in generated stuff. Run

make manifests        # regenerates manifest files
make generate         # regenerates vfs

and commit any changes

[ipiticas]

@siliconbrain, thanks for your guidance, i've made some progress, but still no luck.

[siliconbrain]

@hyperion-hyp you should check the build details 😉 it says that your newly added field (which is present by default) is missing from a config generation test's expected output:

⋮
--- FAIL: TestAwsElasticsearch (0.00s)
⋮
         exception_backup true
        -fail_on_detecting_es_version_retry_exceed true
         fail_on_putting_template_retry_exceed true
⋮

[ipiticas]

done!
tnx @siliconbrain

[siliconbrain]

@hyperion-hyp I hate to be the bearer of bad news...again 😅 but it seems that you have a merge conflict in the generated vfs code because another PR has been merged to master which also changed some custom resources. To resolve this, please rebase your branch onto master OR merge master into your branch, then run

make generate        # regenerate any DeepCopy implementations
make manifests       # regenerates manifest files
make generate         # regenerates vfs

and commit and push (or force push, in case of a rebase) your changes.

[aslafy-z]

LGTM

[ipiticas]

@siliconbrain, A rainbow follows every bad news :), thanks a lot!

[siliconbrain]

👍