Motivated by slow snapshot deletes reported in e.g. #39656 and the fact that these likely are a contributing factor to repositories accumulating stale files over time when deletes fail to finish in time and are interrupted before they can complete.
* Makes snapshot deletion async and parallelizes some steps of the delete process that can be safely run concurrently via the snapshot thread poll
* I did not take the biggest potential speedup step here and parallelize the shard file deletion because that's probably better handled by moving to bulk deletes where possible (and can still be parallelized via the snapshot pool where it isn't). Also, I wanted to keep the size of the PR manageable.
* See https://github.com/elastic/elasticsearch/pull/39656#issuecomment-470492106
* Also, as a side effect this gives the `SnapshotResiliencyTests` a little more coverage for master failover scenarios (since parallel access to a blob store repository during deletes is now possible since a delete isn't a single task anymore).
* By adding a `ThreadPool` reference to the repository this also lays the groundwork to parallelizing shard snapshot uploads to improve the situation reported in #39657
* Add support for setting and keystore settings
* system properties and env var config
* use testclusters for repository-s3
* Some cleanup of the build.gradle file for plugin-s3
* add runner {} to rest integ test task
Blob store compression was not enabled for some of the files in
snapshots due to constructor accessing sub-class fields. Fixed to
instead accept compress field as constructor param. Also fixed chunk
size validation to work.
Deprecated repositories.fs.compress setting as well to be able to unify
in a future commit.
We have had various reports of problems caused by the maxRetryTimeout
setting in the low-level REST client. Such setting was initially added
in the attempts to not have requests go through retries if the request
already took longer than the provided timeout.
The implementation was problematic though as such timeout would also
expire in the first request attempt (see #31834), would leave the
request executing after expiration causing memory leaks (see #33342),
and would not take into account the http client internal queuing (see #25951).
Given all these issues, it seems that this custom timeout mechanism
gives little benefits while causing a lot of harm. We should rather rely
on connect and socket timeout exposed by the underlying http client
and accept that a request can overall take longer than the configured
timeout, which is the case even with a single retry anyways.
This commit removes the `maxRetryTimeout` setting and all of its usages.
Stop passing `Settings` to `AbstractComponent`'s ctor. This allows us to
stop passing around `Settings` in a *ton* of places. While this change
touches many files, it touches them all in fairly small, mechanical
ways, doing a few things per file:
1. Drop the `super(settings);` line on everything that extends
`AbstractComponent`.
2. Drop the `settings` argument to the ctor if it is no longer used.
3. If the file doesn't use `logger` then drop `extends
AbstractComponent` from it.
4. Clean up all compilation failure caused by the `settings` removal
and drop any now unused `settings` isntances and method arguments.
I've intentionally *not* removed the `settings` argument from a few
files:
1. TransportAction
2. AbstractLifecycleComponent
3. BaseRestHandler
These files don't *need* `settings` either, but this change is large
enough as is.
Relates to #34488
The contains syntax was added in #30874 but the skips were not properly
put in place.
The java runner has the feature so the tests will run as part of the
build, but language clients will be able to support it at their own
pace.
* Detect and prevent configuration that triggers a Gradle bug
As we found in #31862, this can lead to a lot of wasted time as it's not
immediatly obvius what's going on.
Givent how many projects we have it's getting increasingly easier to run
into gradle/gradle#847.
Adds a new parameter to the BlobContainer#write*Blob methods to specify whether the existing file
should be overridden or not. For some metadata files in the repository, we actually want to replace
the current file. This is currently implemented through an explicit blob delete and then a fresh write.
In case of using a cloud provider (S3, GCS, Azure), this results in 2 API requests instead of just 1.
This change will therefore allow us to achieve the same functionality using less API requests.
This commit removes some tests in the repository-s3 plugin that
have not been executed for 2+ years but have been maintained
for nothing. Most of the tests in AbstractAwsTestCase were
obsolete or superseded by fixture based integration tests.
This pull request merges the AzureStorageService interface and
the AzureStorageServiceImpl classes into one single
AzureStorageService class. It also removes some tests in the
repository-azure plugin that have not been executed for 2+ years.
The current AzureStorageServiceImpl always checks if the Azure container
exists before reading or writing an object to the Azure container. This commit
removes this behavior, reducing the number of overhall requests executed
for all snapshots operations.
* remove left-over comment
* make sure of the property for plugins
* skip installing modules if these exist in the distribution
* Log the distrbution being ran
* Don't allow running with integ-tests-zip passed externally
* top level x-pack/qa can't run with oss distro
* Add support for matching objects in lists
Makes it possible to have a key that points to a list and assert that a
certain object is present in the list. All keys have to be present and
values have to match. The objects in the source list may have additional
fields.
example:
```
match: { 'nodes.$master.plugins': { name: ingest-attachment } }
```
* Update plugin and module tests to work with other distributions
Some of the tests expected that the integration tests will always be ran
with the `integ-test-zip` distribution so that there will be no other
plugins loaded.
With this change, we check for the presence of the plugin without
assuming exclusivity.
* Allow modules to run on other distros as well
To match the behavior of tets.distributions
* Add and use a new `contains` assertion
Replaces the previus changes that caused `match` to do a partial match.
* Implement PR review comments
Adds the ability to reread and decrypt the local node keystore.
Commonly, the contents of the keystore, backing the `SecureSettings`,
are not retrievable except during node initialization. This changes that
by adding a new API which broadcasts a password to every node. The
password is used to decrypt the local keystore and use it to populate
a `Settings` object that is passes to all the plugins implementing the
`ReloadablePlugin` interface. The plugin is then responsible to do
whatever "reload" means in his case. When the `reload`handler returns,
the keystore is closed and its contents are no longer retrievable.
Password is never stored persistently on any node.
Plugins that have been moded in this commit are: `repository-azure`,
`repository-s3`, `repository-gcs` and `discovery-ec2`.
Many fixtures have similar code for writing the pid & ports files or
for handling HTTP requests. This commit adds an AbstractHttpFixture
class in the test framework that can be extended for specific testing purposes.
There's no need for an extra blobExists() call when writing a blob to the Azure service. Azure
provides an option (with stronger consistency guarantees) on the upload method that guarantees
that the blob that's uploaded does not already exist. This saves one network roundtrip.
Relates to #19749
This commit removes some log traces in AzureStorageServiceImpl and also
fixes the AzureStorageServiceTests so that is uses the real
implementation to create Azure clients.
Similarly to what has been done in for the repository-s3 plugin, this
pull request moves the fixture test into a dedicated
repository-azure/qa/microsoft-azure-storage project.
It also exposes some environment variables which allows to execute the
integration tests against the real Azure Storage service. When the
environment variables are not defined, the integration tests are
executed using the fixture added in #29347.
Closes#29349
* Move Streams.copy into elasticsearch-core and make a multi-release jar
This moves the method `Streams.copy(InputStream in, OutputStream out)` into the
`elasticsearch-core` project (inside the `o.e.core.internal.io` package). It
also makes this class into a multi-release class where the Java 9 equivalent
uses `InputStream#transferTo`.
This is a followup from
https://github.com/elastic/elasticsearch/pull/29300#discussion_r178147495
This commit adds a new fixture that emulates an
Azure Storage 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 for Amazon S3 in #29296, and it
would have helped a lot to catch bugs like #22534.
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.
As we have factored Elasticsearch into smaller libraries, we have ended
up in a situation that some of the dependencies of Elasticsearch are not
available to code that depends on these smaller libraries but not server
Elasticsearch. This is a good thing, this was one of the goals of
separating Elasticsearch into smaller libraries, to shed some of the
dependencies from other components of the system. However, this now
means that simple utility methods from Lucene that we rely on are no
longer available everywhere. This commit copies IOUtils (with some small
formatting changes for our codebase) into the fold so that other
components of the system can rely on these methods where they no longer
depend on Lucene.
This is related to #28662. It wraps the azure repository inputstream in
an inputstream that ensures `read` calls have socket permissions. This
is because the azure inputstream internally makes service calls.
We use affix settings to group settings / values under a certain namespace.
In some cases like login information for instance a setting is only valid if
one or more other settings are present. For instance `x.test.user` is only valid
if there is an `x.test.passwd` present and vice versa. This change allows to specify
such a dependency to prevent settings updates that leave settings in an inconsistent
state.
Only tests should use the single argument Environment constructor. To
enforce this the single arg Environment constructor has been replaced with
a test framework factory method.
Production code (beyond initial Bootstrap) should always use the same
Environment object that Node.getEnvironment() returns. This Environment
is also available via dependency injection.
Today we return a `String[]` that requires copying values for every
access. Yet, we already store the setting as a list so we can also directly
return the unmodifiable list directly. This makes list / array access in settings
a much cheaper operation especially if lists are large.
While working on #26751, I found that we are passing the container name on every single method although we don't need it as it is stored within the blobstore object already.
This commit simplifies a bit that part of the code.
It also removes `repositoryName` from AzureBlobStore which was not used anymore.
Also we move some properties in AzureBlobContainer to `private` members.
Even though you annotate the Test class with `@ThirdParty` the static
code is initialized.
In that case it fails with:
```
==> Test Info: seed=529C3C6977F695FC; jvms=3; suites=6
Suite: org.elasticsearch.repositories.azure.AzureSnapshotRestoreTests
ERROR 0.00s J2 | AzureSnapshotRestoreTests (suite) <<< FAILURES!
> Throwable #1: java.lang.IllegalStateException: to run integration tests, you need to set -Dtests.thirdparty=true and -Dtests.azure.account=azure-account -Dtests.azure.key=azure-key
> at org.elasticsearch.cloud.azure.AzureTestUtils.generateMockSecureSettings(AzureTestUtils.java:37)
> at org.elasticsearch.repositories.azure.AzureSnapshotRestoreTests.generateMockSettings(AzureSnapshotRestoreTests.java:81)
> at org.elasticsearch.repositories.azure.AzureSnapshotRestoreTests.<clinit>(AzureSnapshotRestoreTests.java:84)
> at java.lang.Class.forName0(Native Method)
> at java.lang.Class.forName(Class.java:348)
Completed [1/6] on J2 in 2.21s, 0 tests, 1 error <<< FAILURES!
```
Closes#26812.
(cherry picked from commit eb6d714 for master branch)
* Use Azure upload method instead of our own implementation
We are not following the Azure documentation about uploading blobs to Azure storage. https://docs.microsoft.com/en-us/azure/storage/blobs/storage-java-how-to-use-blob-storage#upload-a-blob-into-a-container
Instead we are using our own implementation which might cause some troubles and rarely some blobs can be not immediately commited just after we close the stream. Using the standard implementation provided by Azure team should allow us to benefit from all the magic Azure SDK team already wrote.
And well... Let's just read the doc!
* Adapt integration tests to secure settings
That was a missing part in #23405.
* Simplify all the integration tests and *extends ESBlobStoreRepositoryIntegTestCase tests
* removes IT `testForbiddenContainerName()` as it is useless. The plugin does not create anymore the container but expects that the user has created it before registering the repository
* merges 2 IT classes so all IT tests are ran from one single class
* We don't remove/create anymore the container between each single test but only for the test suite
While working on #26751 and doing some manual integration testing I found that this #22858 removed an important line of our code:
`AzureRepository` overrides default `initializeSnapshot` method which creates metadata files and do other stuff.
But with PR #22858, I wrote:
```java
@Override
public void initializeSnapshot(SnapshotId snapshotId, List<IndexId> indices, MetaData clusterMetadata) {
if (blobStore.doesContainerExist(blobStore.container()) == false) {
throw new IllegalArgumentException("The bucket [" + blobStore.container() + "] does not exist. Please create it before " +
" creating an azure snapshot repository backed by it.");
}
}
```
instead of
```java
@Override
public void initializeSnapshot(SnapshotId snapshotId, List<IndexId> indices, MetaData clusterMetadata) {
if (blobStore.doesContainerExist(blobStore.container()) == false) {
throw new IllegalArgumentException("The bucket [" + blobStore.container() + "] does not exist. Please create it before " +
" creating an azure snapshot repository backed by it.");
}
super.initializeSnapshot(snapshotId, indices, clusterMetadata);
}
```
As we never call `super.initializeSnapshot(...)` files are not created and we can't restore what we saved.
Closes#26777.
You can define a proxy using the following settings:
```yml
azure.client.default.proxy.host: proxy.host
azure.client.default.proxy.port: 8888
azure.client.default.proxy.type: http
```
Supported values for `proxy.type` are `direct`, `http` or `socks`. Defaults to `direct` (no proxy).
Closes#23506
BTW I changed a test `testGetSelectedClientBackoffPolicyNbRetries` as it was using an old setting name `cloud.azure.storage.azure.max_retries` instead of `azure.client.azure1.max_retries`.
Follow up for #23405.
We remove azure deprecated settings in 7.0:
* The legacy azure settings which where starting with `cloud.azure.storage.` prefix have been removed.
This includes `account`, `key`, `default` and `timeout`.
You need to use settings which are starting with `azure.client.` prefix instead.
* Global timeout setting `cloud.azure.storage.timeout` has been removed.
You must set it per azure client instead. Like `azure.client.default.timeout: 10s` for example.
We should have the same behavior for Azure repositories as we have for S3 (see #22762).
Instead of:
```yml
cloud:
azure:
storage:
my_account1:
account: your_azure_storage_account1
key: your_azure_storage_key1
default: true
my_account2:
account: your_azure_storage_account2
key: your_azure_storage_key2
```
Support something like:
```
azure.client:
default:
account: your_azure_storage_account1
key: your_azure_storage_key1
my_account2:
account: your_azure_storage_account2
key: your_azure_storage_key2
```
Then instead of:
```
PUT _snapshot/my_backup3
{
"type": "azure",
"settings": {
"account": "my_account2"
}
}
```
Use:
```
PUT _snapshot/my_backup3
{
"type": "azure",
"settings": {
"config": "my_account2"
}
}
```
If someone uses:
```
PUT _snapshot/my_backup3
{
"type": "azure"
}
```
It will use the `default` azure repository settings.
And mark as deprecated old settings.
Closes#22763.
Alpar Torok
Armin Braun
Henning Andersen
Luca Cavanna
Gordon Brown
Jernej Klancic
Yannick Welsch
Nik Everett
Pratik Sanglikar
Vladimir Dolzhenko
Tanguy Leroux
Albert Zaharovits
Lee Hinman
Jason Tedor
Tim Brooks
Simon Willnauer
David Roberts
David Pilato
Hendrik Muhs
kel
etiennecarriere