In #19749 an extra check was added before writing each blob to ensure that we would not be
overriding an existing blob. Due to S3's weak consistency model, this check was best effort. To
make matters worse, however, this resulted in a HEAD request to be done before every PUT, in
particular also when PUTTING a new object. The approach taken in #19749 worsened our
consistency guarantees for follow-up snapshot actions, as it made it less likely for new files that
had been written to be available for reads.
This commit therefore removes this extra check. Due to the weak consistency model, this check
was a best effort thing anyway, and there's currently no way to prevent accidental overrides on S3.
This commit moves the repository-s3 fixture test added in #29296 in a
new `repository-s3/qa/amazon-s3` project. This new project allows the
REST integration tests to be executed using the real S3 service when
all the required environment variables are provided. When no env var
is provided, then the tests are executed using the fixture added
in #29296.
The REST tests located at the `repository-s3`plugin project now only
verify that the plugin is correctly loaded.
The REST tests have been adapted to allow a bucket name and a base
path to be specified as env vars. This way it is possible to run the tests
with different base paths (could be anything, like a CI job name or a
branch name) without multiplicating buckets.
Related to #29349
This commit adds the S3BlobStoreRepositoryTests class that extends the
base testing class for S3. It also removes some usage of socket servers
that emulate socket connections in unit tests. It was added to trigger
security exceptions, but this won't be needed anymore since #29296
is merged.
Today when you input a byte size setting that is out of bounds for the
setting, you get an error message that indicates the maximum value of
the setting. The problem is that because we use ByteSize#toString, we
end up with a representation of the value that does not really tell you
what the bound is. For example, if the bound is 2^31 - 1 bytes, the
output would be 1.9gb which does not really tell you want the limit as
there are many byte size values that we format to the same 1.9gb with
ByteSize#toString. We have a method ByteSize#getStringRep that uses the
input units to the value as the output units for the string
representation, so we end up with no loss if we use this to report the
bound. This commit does this.
This commit adds a new fixture that emulates a S3 service in order to
improve the existing integration tests. This is very similar to what has
been made for Google Cloud Storage in #28788, and such tests would
have helped a lot to catch bugs like #22534.
The AmazonS3Fixture is brittle and only implements the very necessary
stuff for the S3 repository to work, but at least it works and can be
adapted for specific tests needs.
* Fixes ByteSizeValue to serialise correctly
This fix makes a few fixes to ByteSizeValue to make it possible to perform round-trip serialisation:
* Changes wire serialisation to use Zlong methods instead of VLong methods. This is needed because the value `-1` is accepted but previously if `-1` is supplied it cannot be serialised using the wire protocol.
* Limits the supplied size to be no more than Long.MAX_VALUE when converted to bytes. Previously values greater than Long.MAX_VALUE bytes were accepted but would be silently interpreted as Long.MAX_VALUE bytes rather than erroring so the user had no idea the value was not being used the way they had intended. I consider this a bug and so fine to include this bug fix in a minor version but I am open to other points of view.
* Adds a `getStringRep()` method that can be used when serialising the value to JSON. This will print the bytes value if the size is positive, `”0”` if the size is `0` and `”-1”` if the size is `-1`.
* Adds logic to detect fractional values when parsing from a String and emits a deprecation warning in this case.
* Modifies hashCode and equals methods to work with long values rather than doubles so they don’t run into precision problems when dealing with large values. Previous to this change the equals method would not detect small differences in the values (e.g. 1-1000 bytes ranges) if the actual values where very large (e.g. PBs). This was due to the values being in the order of 10^18 but doubles only maintaining a precision of ~10^15.
Closes#27568
* Fix bytes settings default value to not use fractional values
* Fixes test
* Addresses review comments
* Modifies parsing to preserve unit
This should be bwc since in the case that the input is fractional it reverts back to the old method of parsing it to the bytes value.
* Addresses more review comments
* Fixes tests
* Temporarily changes version check to 7.0.0
This will be changed to 6.2 when the fix has been backported
This pull request changes the S3BlobContainer.blobExists() method implementation
to make it use the AmazonS3.doesObjectExist() method instead of
AmazonS3.getObjectMetadata(). The AmazonS3 implementation takes care of
catching any thrown AmazonS3Exception and compares its response code with 404,
returning false (object does not exist) or lets the exception be propagated.
The AWS SDK has a transitive dependency on Jackson Databind. While the
AWS SDK was recently upgraded, the Jackson Databind dependency was not
pulled along with it to the version that the AWS SDK depends on. This
commit upgrades the dependencies for discovery-ec2 and repository-s3
plugins to match versions on the AWS SDK transitive dependencies.
Relates #27361
Now the blob size information is available before writing anything,
the repository implementation can know upfront what will be the
more suitable API to upload the blob to S3.
This commit removes the DefaultS3OutputStream and S3OutputStream
classes and moves the implementation of the upload logic directly in the
S3BlobContainer.
related #26993closes#26969
With Gradle 4.1 and newer JDK versions, we can finally invoke Gradle directly using a JDK9 JAVA_HOME without requiring a JDK8 to "bootstrap" the build. As the thirdPartyAudit task runs within the JVM that Gradle runs in, it needs to be adapted now to be JDK9 aware.
This commit also changes the `JavaCompile` tasks to only fork if necessary (i.e. when Gradle's JVM and JAVA_HOME's JVM differ).
Moved SocketAccess.doPrivileged up the stack to DefaultS3OutputStream in repository-S3 plugin to avoid SecurityException by Streams.copy(). A plugin is only allowed to use its own jars when performing privileged operations. The S3 client might open a new Socket on close(). #25192
This commit renames all rest test files to use the .yml extension
instead of .yaml. This way the extension used within all of
elasticsearch for yaml is consistent.
Specifying s3 access and secret keys inside repository settings are not
secure. However, until there is a way to dynamically update secure
settings, this is the only way to dynamically add repositories with
credentials that are not known at node startup time. This commit adds
back `access_key` and `secret_key` s3 repository settings, but protects
it with a required system property `allow_insecure_settings`.
Most of these settings should always be pulled from the repository
settings. A couple were leftover that should be moved to client
settings. The path style access setting should be removed altogether.
This commit adds deprecations for all of these existing settings, as
well as adding new client specific settings for max retries and
throttling.
relates #24143
This change simplifies how the rest test runner finds test files and
removes all leniency. Previously multiple prefixes and suffixes would
be tried, and tests could exist inside or outside of the classpath,
although outside of the classpath never quite worked. Now only classpath
tests are supported, and only one resource prefix is supported,
`/rest-api-spec/tests`.
closes#20240
The S3 repostiory has many levels of settings it looks at to create a
repository, and these settings were read at repository creation time.
This meant secure settings like access and secret keys had to be
available after node construction. This change makes setting loading for
every except repository level settings eager, so that secure settings
can be stashed, and the keystore can once again be closed after
bootstrapping the node is complete.
This commit removes passing the repository metadata object through to
s3 client creation. It is not needed, and in fact in tests was confusing
because you could create the metadata but have it contain different
settings than were passed in as repository settings.
This commit removes the "legacy" feature of secure settings, which setup
a parallel setting that was a fallback in the insecure
elasticsearch.yml. This was previously used to allow the new secure
setting name to be that of the old setting name, but is now not in use
due to other refactorings. It is much cleaner to just have all secure
settings use new setting names. If in the future we want to reuse the
previous setting name, once support for the insecure settings have been
removed, we can then rename the secure setting. This also adds a test
for the behavior.
Currently, both the Amazon S3 client provides a retry mechanism, and the
S3 blob store also attempts retries for failed read/write requests.
Both retry mechanisms are controlled by the
`repositories.s3.max_retries` setting. However, the S3 blob store retry
mechanism is unnecessary because the Amazon S3 client provided by the
Amazon SDK already handles retries (with exponential backoff) based on
the provided max retry configuration setting (defaults to 3) as long as
the request is retryable. Hence, this commit removes the unneeded retry
logic in the S3 blob store and the S3OutputStream.
Closes#22845
This commit puts all the classes in the repository-s3 plugin into a
single package. In addition to simplifying the plugin, it will make it
easier to test as things that should be package private will not be
difficult to use inside tests alone.
This commit renames the random ASCII helper methods in ESTestCase. This
is because this method ultimately uses the random ASCII methods from
randomized runner, but these methods actually only produce random
strings generated from [a-zA-Z].
Relates #23886
This commit adds a convenience method for simultaneously asserting
settings deprecations and other warnings and fixes some tests where
setting deprecations and general warnings were present.
The warning header used by Elasticsearch for delivering deprecation
warnings has a specific format (RFC 7234, section 5.5). The format
specifies that the warning header should be of the form
warn-code warn-agent warn-text [warn-date]
Here, the warn-code is a three-digit code which communicates various
meanings. The warn-agent is a string used to identify the source of the
warning (either a host:port combination, or some other identifier). The
warn-text is quoted string which conveys the semantic meaning of the
warning. The warn-date is an optional quoted date that can be in a few
different formats.
This commit corrects the warning header within Elasticsearch to follow
this specification. We use the warn-code 299 which means a
"miscellaneous persistent warning." For the warn-agent, we use the
version of Elasticsearch that produced the warning. The warn-text is
unchanged from what we deliver today, but is wrapped in quotes as
specified (this is important as a problem that exists today is that
multiple warnings can not be split by comma to obtain the individual
warnings as the warnings might themselves contain commas). For the
warn-date, we use the RFC 1123 format.
Relates #23275
This is fallout from #23297. That commit wrapped
`InstanceProfileCredentialsProvider` to ensure that the `getCredentials`
and `refresh` methods had privileged access. However, it looks like
there was a test ensuring that `buildCredentials` returned the correct
clazz type. This commit adjusts that test to check that the correct
wrapper is returned.
This commit fixes an issue that was missed in #22534.
`AWSCredentialsProvider.getCredentials()` appears to potentially open a
socket connect. This operation needed to be wrapped in `doPrivileged()`.
This should fix issue #23271.
We have a bunch of interfaces that have only a single implementation
for 6 years now. These interfaces are pretty useless from a SW development
perspective and only add unnecessary abstractions. They also require
lots of casting in many places where we expect that there is only one
concrete implementation. This change removes the interfaces, makes
all of the classes final and removes the duplicate `foo` `getFoo` accessors
in favor of `getFoo` from these classes.
Secure settings from the elasticsearch keystore were not yet validated.
This changed improves support in Settings so that secure settings more
seamlessly blend in with normal settings, allowing the existing settings
validation to work. Note that the setting names are still not validated
(yet) when using the elasticsearc-keystore tool.
This is related to #22116. Core no longer needs `SocketPermission`
`connect`.
This permission is relegated to these modules/plugins:
- transport-netty4 module
- reindex module
- repository-url module
- discovery-azure-classic plugin
- discovery-ec2 plugin
- discovery-gce plugin
- repository-azure plugin
- repository-gcs plugin
- repository-hdfs plugin
- repository-s3 plugin
And for tests:
- mocksocket jar
- rest client
- httpcore-nio jar
- httpasyncclient jar
This commit upgrades the checkstyle configuration from version 5.9 to
version 7.5, the latest version as of today. The main enhancement
obtained via this upgrade is better detection of redundant modifiers.
Relates #22960
This change removes the ability to set region for s3 repositories.
Endpoint should be used instead if a custom s3 location needs to be
used.
closes#22758
* S3 repository: Add named configurations
This change implements named configurations for s3 repository as
proposed in #22520. The access/secret key secure settings which were
added in #22479 are reverted, and the only secure settings are those
with the new named configs. All other previously used settings for the
connection are deprecated.
closes#22520
This commit replaces specialized functional interfaces in various
plugins with generic options. Instead of creating `StorageRunnable`
interfaces in every plugin we can just use `Runnable` or `CheckedRunnable`.
This commit adds a SpecialPermission constant and uses that constant
opposed to introducing new instances everywhere.
Additionally, this commit introduces a single static method to check that
the current code has permission. This avoids all the duplicated access
blocks that exist currently.
* S3 repository: Deprecate specifying credentials through env vars and sys props
This is a follow up to #22479, where storing credentials secure way was
added.
This is related to #22116. Certain plugins (discovery-azure-classic,
discovery-ec2, discovery-gce, repository-azure, repository-gcs, and
repository-s3) open socket connections. As SocketPermissions are
transitioned out of core, these plugins will require connect
permission. This pull request wraps operations that require these
permissions in doPrivileged blocks.
* Settings: Make s3 repository sensitive settings use secure settings
This change converts repository-s3 to use the new secure settings. In
order to support the multiple ways we allow aws creds to be configured,
it also moves the main methods for the keystore wrapper into a
SecureSettings interface, in order to allow settings prefixing to work.
* Remove a checked exception, replacing it with `ParsingException`.
* Remove all Parser classes for the yaml sections, replacing them with static methods.
* Remove `ClientYamlTestFragmentParser`. Isn't used any more.
* Remove `ClientYamlTestSuiteParseContext`, replacing it with some static utility methods.
I did not rewrite the parsers using `ObjectParser` because I don't think it is worth it right now.
We are currenlty checking that no deprecation warnings are emitted in our query tests. That can be moved to ESTestCase (disabled in ESIntegTestCase) as it allows us to easily catch where our tests use deprecated features and assert on the expected warnings.
At one point in the past when moving out the rest tests from core to
their own subproject, we had multiple test classes which evenly split up
the tests to run. However, we simplified this and went back to a single
test runner to have better reproduceability in tests. This change
removes the remnants of that multiplexing support.
Currently the default S3 buffer size is 100MB, which can be a lot for small
heaps. This pull request updates the default to be 100MB for heaps that are
greater than 2GB and 5% of the heap size otherwise.
This commit fixes an issue with the configuration for the AwsSdkMetrics
logger; the issue is that the logging configuration had used underscores
instead of periods for the settings key (the perils of lenient settings
parsing).
Relates #20313
In 2.x, the S3 repository accepted a `/` (forward slash) to start
the repositories.s3.base_path, and it used a different string splitting
method that removed the forward slash from the base path, so there
were no issues.
In 5.x, we removed this custom string splitting method in favor of
the JDK's string splitting method, which preserved the leading `/`.
The AWS SDK does not like the leading `/` in the key path after the
bucket name, and so it could not find any objects in the S3 repository.
This commit fixes the issue by removing the leading `/` if it exists
and adding a deprecation notice that leading `/` will not be supported
in the future in S3 repository's base_path.
This commit removes `ByteSizeValue`'s methods that are duplicated (ex: `mbFrac()` and `getMbFrac()`) in order to only keep the `getN` form.
It also renames `mb()` -> `getMb()`, `kb()` -> `getKB()` in order to be more coherent with the `ByteSizeUnit` method names.
Because of security permissions that we do not grant to the AWS SDK (for
use in discovery-ec2 and repository-s3 plugins), certain calls in the
AWS SDK will lead to security exceptions that are logged at the warning
level. These warnings are noise and we should suppress them. This commit
adds plugin log configurations for discovery-ec2 and repository-s3 to
ship with default Log4j 2 configurations that suppress these log
warnings.
Relates #20313
This commit modifies the call sites that allocate a parameterized
message to use a supplier so that allocations are avoided unless the log
level is fine enough to emit the corresponding log message.
conform with the requirements of the writeBlob method by
throwing a FileAlreadyExistsException if attempting to write
to a blob that already exists. This change means implementations
of BlobContainer should never overwrite blobs - to overwrite a
blob, it must first be deleted and then can be written again.
Closes#15579
While I was working on #18703, I discovered a bad behavior when people don't provide AWS key/secret as part as their `elasticsearch.yml` but rely on SysProps or env. variables...
In [`InternalAwsS3Service#getClient(...)`](d4366f8493/plugins/repository-s3/src/main/java/org/elasticsearch/cloud/aws/InternalAwsS3Service.java (L76-L141)), we have:
```java
Tuple<String, String> clientDescriptor = new Tuple<>(endpoint, account);
AmazonS3Client client = clients.get(clientDescriptor);
```
But if people don't provide credentials, `account` is `null`.
Even if it actually could work, I think that we should use the `AWSCredentialsProvider` we create later on and extract from it the `account` (AWS KEY actually) and then use it as the second value of the tuple.
Closes#19557.
This makes it obvious that these tests are for running the client yaml
suites. Now that there are other ways of running tests using the REST
client against a running cluster we can't go on calling the shared
client yaml tests "REST tests". They are rest tests, but they aren't
**the** rest tests.
This adds a header that looks like `Location: /test/test/1` to the
response for the index/create/update API. The requirement for the header
comes from https://www.w3.org/Protocols/rfc2616/rfc2616-sec10.htmlhttps://tools.ietf.org/html/rfc7231#section-7.1.2 claims that relative
URIs are OK. So we use an absolute path which should resolve to the
appropriate location.
Closes#19079
This makes large changes to our rest test infrastructure, allowing us
to write junit tests that test a running cluster via the rest client.
It does this by splitting ESRestTestCase into two classes:
* ESRestTestCase is the superclass of all tests that use the rest client
to interact with a running cluster.
* ESClientYamlSuiteTestCase is the superclass of all tests that use the
rest client to run the yaml tests. These tests are shared across all
official clients, thus the `ClientYamlSuite` part of the name.
Follow up for #18662 and #18690.
* For consistency, we rename method parameters and use `key` and `secret` instead of `account` and `key`.
* We add some tests to check that settings are correctly applied.
* Tests revealed that some checks are bad like for #18662.
Add test and fix issue for getting the right S3 endpoint
Test when Repository, Repositories or global settings are defined
But ignore testAWSCredentialsWithSystemProviders test
Add tests for AWS Client Configuration
Fix NPE when no region is set
We used to transform region="" to region=null but it's not needed anymore and would actually cause NPE from now.
Follow up for https://github.com/elastic/elasticsearch/pull/17784#discussion_r64575845
Today we are registering repository settings when `S3RepositoryPlugin` starts:
```java
settingsModule.registerSetting(S3Repository.Repository.KEY_SETTING);
settingsModule.registerSetting(S3Repository.Repository.SECRET_SETTING);
settingsModule.registerSetting(S3Repository.Repository.BUCKET_SETTING);
settingsModule.registerSetting(S3Repository.Repository.ENDPOINT_SETTING);
settingsModule.registerSetting(S3Repository.Repository.PROTOCOL_SETTING);
settingsModule.registerSetting(S3Repository.Repository.REGION_SETTING);
settingsModule.registerSetting(S3Repository.Repository.SERVER_SIDE_ENCRYPTION_SETTING);
settingsModule.registerSetting(S3Repository.Repository.BUFFER_SIZE_SETTING);
settingsModule.registerSetting(S3Repository.Repository.MAX_RETRIES_SETTING);
settingsModule.registerSetting(S3Repository.Repository.CHUNK_SIZE_SETTING);
settingsModule.registerSetting(S3Repository.Repository.COMPRESS_SETTING);
settingsModule.registerSetting(S3Repository.Repository.STORAGE_CLASS_SETTING);
settingsModule.registerSetting(S3Repository.Repository.CANNED_ACL_SETTING);
settingsModule.registerSetting(S3Repository.Repository.BASE_PATH_SETTING);
```
We don't need to register those settings as they are repository level settings and not node level settings.
Closes#18945.
Repository plugins currently use a lot of custom classes like
RepositoryName and RepositorySettings in order to use guice to construct
repository implementations. But repositories now only really need their
settings to be constructed. Anything else they need (eg a cloud client)
can be constructed within the plugin, instead of via guice.
This change makes repository plugins use the new pull model. It removes
guice from the construction of Repository objects (no more child
injectors) and also from all repository plugins.
The api for snapshot/restore was split up between two interfaces,
Repository and IndexShardRepository. There was also complex
initialization and injection between the two. However, there is always a
one to one relationship between the two.
This change moves the IndexShardRepository api into Repository, as well
as updates the API so as not to require any services to be injected for
sublcasses.
Today throughout the codebase, catch throwable is used with reckless
abandon. This is dangerous because the throwable could be a fatal
virtual machine error resulting from an internal error in the JVM, or an
out of memory error or a stack overflow error that leaves the virtual
machine in an unstable and unpredictable state. This commit removes
catch throwable from the codebase and removes the temptation to use it
by modifying listener APIs to receive instances of Exception instead of
the top-level Throwable.
Relates #19231
The only reason for LifecycleComponent taking a generic type was so that
it could return that type on its start and stop methods. However, this
chaining has no practical necessity. Instead, start and stop can be
void, and a whole bunch of confusing generics disappear.
Tanguy Leroux
Yannick Welsch
Albert Zaharovits
Jason Tedor
kel
Nhat Nguyen
Colin Goodheart-Smithe
desmorto
joachimdraeger
Ryan Ernst
Ali Beyad
Tim Brooks
Christoph Büscher
Simon Willnauer
animageofmine
Igor Motov
Nik Everett
Nic Palmer
javanna
David Pilato
Alexander Kazakov
Adrien Grand
gfyoung