When Elasticsearch is installed with a custom service_name, puppet will fail because the custom name is not used for the before relationship metaparameter.
class { 'elasticsearch':
service_name => 'elasticsearch_default',
}
results in:
Could not retrieve catalog from remote server: Error 500 on SERVER: Server Error: Could not find resource 'Service[elasticsearch]' in parameter 'before'
<!-- The following list of checkboxes are the prerequisites for getting your contribution accepted. Please check them as they are completed. -->
Pull request acceptance prerequisites:
Many tests uses $version or $elasticsearch::version (jvm.options, ssl parameters) but installation with package_url can't set this parameter.
no issues
Elasticsearch automatically adds the node name but that will be removed in 7.0. This adds the node
name explicitly to the log4j config so it does not get lost when ES 7.0 is release.
Startup warning:
Some logging configurations have %marker but don't have %nodename. We will automatically
add %nodename to the pattern to ease the migration for users who customize log4j2.properties
but will stop this behavior in 7.0. You should manually replace%node_name
with
[%node_name]%marker
in these locations: ....
<!-- The following list of checkboxes are the prerequisites for getting your contribution accepted. Please check them as they are completed. -->
Pull request acceptance prerequisites:
In https://github.com/elastic/puppet-elasticsearch/pull/1036 you fixed the autoload issue for some provider types. While testing today I found that there were still some providers that can fail to autoload.
How to reproduce:
Result:
Error: Could not retrieve catalog from remote server: Error 500 on SERVER: Server Error: Evaluation Error: Error while evaluating a Virtual Query, Could not autoload puppet/type/elasticsearch_user: Could not autoload puppet/provider/elasticsearch_user/users: no such file to load -- puppet/provider/elastic_user_command (file: /etc/puppetlabs/code/modules/elasticsearch/manifests/init.pp, line: 554, column: 3) on node agent1.vagrant.test
Warning: Not using cache on failed catalog
Error: Could not retrieve catalog; skipping run
Applying the same fix as in https://github.com/elastic/puppet-elasticsearch/pull/1036 to these types/providers fixed the issue.
Maybe it would make sense to change the imports for all the types/providers?
Pull request acceptance prerequisites:
This fixes #1197 and effectively replaces #1041 which was originally opened to support #928. I believe this will also fix #1093.
The jvm.options file is considered a config by the RPM (would expect the same to be true of a DEB) so when migrating from a multi-instance to a single instance management and not changing versions, its possible the file jvm.options
would not exist in the new configuration directory. This change ensures the file exists and if not creates an empty file.
<!--
Thank you for contributing to this project!
-->
<!--
Replace this comment with a description of your pull request.
-->
<!--
Replace this comment with the list of issues or n/a.
Use format:
Fixes #123
Fixes #124
-->
<!-- The following list of checkboxes are the prerequisites for getting your contribution accepted. Please check them as they are completed. -->
Pull request acceptance prerequisites:
Add support for composable index templates
Add support for managing SLM policies
Work is based on #1163 and makes #1084 #1102 obsolete
<!-- The following list of checkboxes are the prerequisites for getting your contribution accepted. Please check them as they are completed. -->
Pull request acceptance prerequisites:
<!-- The following list of checkboxes are the prerequisites for getting your contribution accepted. Please check them as they are completed. -->
I think this fixes things but I have run out of time now to check the rest.
Pull request acceptance prerequisites:
Fixed jvm.options.erb to add required jvm options for 7.x
Fixed jvm.options.erb to change jvm options required only for openjdk 8
Added ESPATHDIR instance environment var
Fixed log4j2.properties.erb to fix Deprecation WARNINGS
Keeped Backward Compatibility but not tested yet
<!-- The following list of checkboxes are the prerequisites for getting your contribution accepted. Please check them as they are completed. -->
Pull request acceptance prerequisites:
master
fails on Debian 8
and OpenSuse 42
]This PR adds support to manage s3 repository.
Example:
elasticsearch::snapshot_repository { 's3_repository':
repository_type => 's3',
bucket => $backup_aws_s3_bucket,
region => $backup_aws_s3_region
}
<!-- The following list of checkboxes are the prerequisites for getting your contribution accepted. Please check them as they are completed. -->
Pull request acceptance prerequisites:
Some small cleanups and updates to the module testing matrix.
6.8.6
. 2.x
, as no longer supported. This fixes #1053
Pull request acceptance prerequisites:
Employer EnableIT
If file resource is defined globally, its fails to start the elasticsearch process. so getting the mode set in first place is a right way to do it.
File { owner => root, group => root, mode => '0444' }
Reset potentially inherited defaults (since the mode is not always specified in below items
For example "${elasticsearch::homedir}/lib" was stripped of 'other' permissions, so my Elasticsearch was unable to start
<!-- The following list of checkboxes are the prerequisites for getting your contribution accepted. Please check them as they are completed. -->
Pull request acceptance prerequisites:
- [x] Signed the CLA (if not already signed)
- [x] Rebased/up-to-date with base branch
- [ ] Tests pass
- [ ] Updated CHANGELOG.md with patch notes (if necessary)
- [ ] Updated CONTRIBUTORS (if attribution is requested)
…it.d scripts
Hello,
Here is a quick improvement to be able to use auto-restart feature provided by systemd service files.
Best regards, Adam.
<!-- The following list of checkboxes are the prerequisites for getting your contribution accepted. Please check them as they are completed. -->
Pull request acceptance prerequisites:
Add attribut 'mark' on resource package
Fixes #1133
With this PR it's now possible to set the type of the private key. If, for example, an EC private key is used, it can set with value ec
, which will be passed to java_ks
. The new setting is optional, when not set the current default (rsa) is used. See https://github.com/puppetlabs/puppetlabs-java_ks/blob/main/REFERENCE.md#private_key_type
Pull request acceptance prerequisites:
Add support for managing ILM policies
Work is based on #1146 and makes #1103 obsolete
Fixes #1038
Issue elastic/puppet-elasticsearch#971 - File realm files MUST NOT trigger Elasticsearch service restart
This is a naive implementation and another way would be to handle everything in the elasticsearch_role
, elasticsearch_user
, elasticsearch_role_mapping
providers, letting them create the file if it doesn't exist instead of just parsing them.
Pull request acceptance prerequisites:
This change makes managing the secrets keystore optional. Since the code in that is still setup for instance based deployments which are no longer supported it causes failures when utilizing this module. By making the secrets keystore management optional it buys time to fix that code while still allowing users to take advantage of the other managed aspects.
This PR is based off of #1198 as we needed both changes in our environment.
modulesync 6.0.0
modulesync 5.4.0