Disable bootstrap checks for single-node discovery #23598
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This commit adds a single node discovery type. With this discovery type, a node will elect itself as master and never form a cluster with another node.
While there are use-cases where a single-node is in production, there are also use-cases for starting a single-node that binds transport to an external interface where the node is not in production (for example, for testing the transport client against a node started in a Docker container). It's tricky to balance the desire to always enforce the bootstrap checks when a node might be in production with the need for the community to perform testing in situations that would trip the bootstrap checks. This commit enables some flexibility for these users. By setting the discovery type to "single-node", we disable the bootstrap checks independently of how transport is bound. While this sounds like a hole in the bootstrap checks, the bootstrap checks can already be avoided in the single-node use-case by binding only HTTP but not transport. For users that are genuinely in production on a single-node use-case with transport bound to an external use-case, they can set the system property "es.enable.bootstrap.checks" to force running the bootstrap checks. It would be a mistake for them not to do this.
jasontedor
added
:Core/Infra/Core
* single-node-discovery: Fix NPE getting discovery type
Member
Author
* single-node-discovery: Log the discovery type Fix variable name
* single-node-discovery: Fix license headers
* single-node-discovery: Return empty discovery stats Avoid test race when starting initial join Revert inadvertent change
* single-node-discovery: Add test single nodes do not discover
We need to negate this condition as other discovery types follow the zen logic too, only single-node needs to be excluded.
* single-node-discovery: Change discovery condition
* single-node-discovery: Wait for pinging to finish
* single-node-discovery: Remove repeat
* master: Clear the interrupt flag before joining Migrate to max line length of 100 Docs: Corrected path to elasticsearch-plugin (elastic#23622) Docs: Add comma to reverse nested agg snippet Fix third-party audit task for Gradle 3.4 (elastic#23612) Adapter action future should restore interrupts Update scripting.asciidoc Unmark reindex as experimental CompletionSuggestionContext#toQuery() should also consider text if prefix/regex missing (elastic#23451) Docs: Specify that byte units use powers of 1024 (elastic#23574) Remove Settings.settingsBuilder (elastic#23575) Change params._source to params['_source'] in example. Fix example in documentation for Painless using _source. (elastic#21322) Remove extra line from license header Fix num docs to be positive in bucket deferring collector test Mapping: Fix NPE with scaled floats stats when field is not indexed (elastic#23528)
* single-node-discovery: Clear the interrupt flag before joining Migrate to max line length of 100 Docs: Corrected path to elasticsearch-plugin (elastic#23622) Docs: Add comma to reverse nested agg snippet Fix third-party audit task for Gradle 3.4 (elastic#23612) Adapter action future should restore interrupts Update scripting.asciidoc Unmark reindex as experimental CompletionSuggestionContext#toQuery() should also consider text if prefix/regex missing (elastic#23451) Docs: Specify that byte units use powers of 1024 (elastic#23574) Remove Settings.settingsBuilder (elastic#23575) Change params._source to params['_source'] in example. Fix example in documentation for Painless using _source. (elastic#21322) Remove extra line from license header Fix num docs to be positive in bucket deferring collector test Mapping: Fix NPE with scaled floats stats when field is not indexed (elastic#23528)
* master: Eclipse: move print margin to 100 columns Add support for fragment_length in the unified highlighter (elastic#23431)
* single-node-discovery: Eclipse: move print margin to 100 columns Add support for fragment_length in the unified highlighter (elastic#23431)
* master: (146 commits) Introduce single-node discovery Await termination after shutting down executors Fix initialization issue in ElasticsearchException Fix bulk queue size in thread pool docs [DOCS] Remove line about eager loading global ordinals GceDiscoverTests - remove intitial_state_timeout SpecificMasterNodesIT shouldn't use autoMinMasterNodes testRestorePersistentSettings doesn't to mess with discovery settings testDifferentRolesMaintainPathOnRestart shouldn't use auto managing of min master nodes CONSOLEify the "using scripts" documentation Adds tests for cardinality and filter aggregations Add Backoff policy to azure repository Revert "Adds tests for cardinality and filter aggregations (elastic#23826)" Adds tests for cardinality and filter aggregations (elastic#23826) mute testDifferentRolesMaintainPathOnRestart Replace custom sort field with SortedSetSortField and SortedNumericSortField when possible (elastic#23827) Prevent nodes from joining if newer indices exist in the cluster (elastic#23843) Synchronized CollapseTopFieldDocs with lucenes relatives (elastic#23854) CONSOLEify analysis docs Adapted search_shards rest test to work with Perl ...
Member
Author
|
@bleskes This is ready for you now. |
bleskes
approved these changes
Apr 4, 2017
jasontedor
added a commit
that referenced
this pull request
Apr 4, 2017
While there are use-cases where a single-node is in production, there are also use-cases for starting a single-node that binds transport to an external interface where the node is not in production (for example, for testing the transport client against a node started in a Docker container). It's tricky to balance the desire to always enforce the bootstrap checks when a node might be in production with the need for the community to perform testing in situations that would trip the bootstrap checks. This commit enables some flexibility for these users. By setting the discovery type to "single-node", we disable the bootstrap checks independently of how transport is bound. While this sounds like a hole in the bootstrap checks, the bootstrap checks can already be avoided in the single-node use-case by binding only HTTP but not transport. For users that are genuinely in production on a single-node use-case with transport bound to an external use-case, they can set the system property "es.enable.bootstrap.checks" to force running the bootstrap checks. It would be a mistake for them not to do this. Relates #23598
Member
Author
|
Thanks @bleskes. |
This was referenced Apr 4, 2017
dliappis
added a commit
to dliappis/elasticsearch
that referenced
this pull request
Aug 18, 2017
For the single node, dev example, the `discovery.type=single-node`[1] is a perfect fit and makes the example shorter and more self explanatory. [1] elastic#23595 [2] elastic#23598
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
While there are use-cases where a single-node is in production, there are also use-cases for starting a single-node that binds transport to an external interface where the node is not in production (for example, for testing the transport client against a node started in a Docker container). It's tricky to balance the desire to always enforce the bootstrap checks when a node might be in production with the need for the community to perform testing in situations that would trip the bootstrap checks. This commit enables some flexibility for these users. By setting the discovery type to "single-node", we disable the bootstrap checks independently of how transport is bound. While this sounds like a hole in the bootstrap checks, the bootstrap checks can already be avoided in the single-node use-case by binding only HTTP but not transport. For users that are genuinely in production on a single-node use-case with transport bound to an external use-case, they can set the system property "es.enable.bootstrap.checks" to force running the bootstrap checks. It would be a mistake for them not to do this.
Relates #23585, relates #23595