Today, certain bootstrap properties are set and read via system
properties. This action-at-distance way of managing these properties is
rather confusing, and completely unnecessary. But another problem exists
with setting these as system properties. Namely, these system properties
are interpreted as Elasticsearch settings, not all of which are
registered. This leads to Elasticsearch failing to startup if any of
these special properties are set. Instead, these properties should be
kept as local as possible, and passed around as method parameters where
needed. This eliminates the action-at-distance way of handling these
properties, and eliminates the need to register these non-setting
properties. This commit does exactly that.
Additionally, today we use the "-D" command line flag to set the
properties, but this is confusing because "-D" is a special flag to the
JVM for setting system properties. This creates confusion because some
"-D" properties should be passed via arguments to the JVM (so via
ES_JAVA_OPTS), and some should be passed as arguments to
Elasticsearch. This commit changes the "-D" flag for Elasticsearch
settings to "-E".
This commit adds fields bytes_recovered and files_recovered to the cat
recovery API. These fields, respectively, indicate the total number of
bytes and files recovered. Additionally, for consistency, some totals
fields and translog recovery fields have been renamed.
Closes#17064
Enables the touching of all memory pages used by the JVM heap spaces
during initialization of the HotSpot VM, which commits all memory pages
at initialization time. By default, pages are committed only as they are
needed.
The ingest stats include the following statistics:
* `ingest.total.count`- The total number of document ingested during the lifetime of this node
* `ingest.total.time_in_millis` - The total time spent on ingest preprocessing documents during the lifetime of this node
* `ingest.total.current` - The total number of documents currently being ingested.
* `ingest.total.failed` - The total number ingest preprocessing operations failed during the lifetime of this node
Also these stats are returned on a per pipeline basis.
This commit updates the documentation for GeoPointField by removing all references to the coerce and doc_values parameters. DocValues are enabled in lucene GeoPointField by default (required for boundary filtering). The QueryBuilders are updated to automatically normalize points (ignoring the coerce parameter) for any index created onOrAfter version 2.2.
Internally the put pipeline API uses this information in node info API to validate if all specified processors in a pipeline exist on all nodes in the cluster.
Closes#16964
Squashed commit of the following:
commit a23f9d2d29220991aa498214530753d7a5a148c6
Merge: eec9c4e 0b0a251
Author: Robert Muir <rmuir@apache.org>
Date: Mon Mar 7 04:12:02 2016 -0500
Merge branch 'master' into lucene6
commit eec9c4e5cd11e9c3e0b426f04894bb2a6dae4f21
Merge: bc67205 675d940
Author: Robert Muir <rmuir@apache.org>
Date: Fri Mar 4 13:45:00 2016 -0500
Merge branch 'master' into lucene6
commit bc67205bdfe1526eae277ab7856fc050ecbdb7b2
Author: Robert Muir <rmuir@apache.org>
Date: Fri Mar 4 09:56:31 2016 -0500
fix test bug
commit a60723b007ff12d97b1810cef473bd7b553a0327
Author: Simon Willnauer <simonw@apache.org>
Date: Fri Mar 4 15:35:35 2016 +0100
Fix SimpleValidateQueryIT to put braces around boosted terms
commit ae3a49d7ba7ced448d2a5262e5d8ec98671a9090
Author: Simon Willnauer <simonw@apache.org>
Date: Fri Mar 4 15:27:25 2016 +0100
fix multimatchquery
commit ae23fdb88a8f6d3fb7ba60fd1aaf3fd72d899aa5
Author: Simon Willnauer <simonw@apache.org>
Date: Fri Mar 4 15:20:49 2016 +0100
Rewrite DecayFunctionScoreIT to be independent of the similarity used
This test relied a lot on the term scoring and compared scores
that are dependent on the similarity. This commit changes the base query
to be a predictable constant score query.
commit 366c2d518c35d31251033f1b6f6a93f6e2ae327d
Author: Simon Willnauer <simonw@apache.org>
Date: Fri Mar 4 14:06:14 2016 +0100
Fix scoring in tests due to changes to idf calculation.
Lucene 6 uses a different default similarity as well as a different
way to calculate IDF. In contrast to older version lucene 6 uses docCount per field
to calculate the IDF not the # of docs in the index to overcome the sparse field
cases.
commit dac99fd64ac2fa71b8d8d106fe68825e574c49f8
Author: Robert Muir <rmuir@apache.org>
Date: Fri Mar 4 08:21:57 2016 -0500
don't hardcoded expected termquery score
commit 6e9f340ba49ab10eed512df86d52a121aa775b0f
Author: Robert Muir <rmuir@apache.org>
Date: Fri Mar 4 08:04:45 2016 -0500
suppress deprecation warning until migrated to points
commit 3ac8908424b3fdad44a90a4f7bdb3eff7efd077d
Author: Robert Muir <rmuir@apache.org>
Date: Fri Mar 4 07:21:43 2016 -0500
Remove invalid test: all commits have IDs, and its illegal to do this.
commit c12976288124ad1a26467e7e848fb810548e7eab
Author: Robert Muir <rmuir@apache.org>
Date: Fri Mar 4 07:06:14 2016 -0500
don't test with unsupported back compat
commit 18bbfe76128570bc70883bf91ff4c44c82d27817
Author: Robert Muir <rmuir@apache.org>
Date: Fri Mar 4 07:02:18 2016 -0500
remove now invalid lucene 4 backcompat test
commit 7e730e572886f0ef2d3faba712e4256216ff01ec
Author: Robert Muir <rmuir@apache.org>
Date: Fri Mar 4 06:58:52 2016 -0500
remove now invalid lucene 4 backwards test
commit 244d2ab6868ba5ac9e0bcde3c2833743751a25ec
Author: Robert Muir <rmuir@apache.org>
Date: Fri Mar 4 06:47:23 2016 -0500
use 6.0 codec
commit 5f64d4a431a6fdaa1234adca23f154c2a1de8284
Author: Robert Muir <rmuir@apache.org>
Date: Fri Mar 4 06:43:08 2016 -0500
compile, javadocs, forbidden-apis, etc
commit 1f273cd62a7fe9ca8f8944acbbfc5cbdd3d81ccb
Merge: cd33921 29e3443
Author: Simon Willnauer <simonw@apache.org>
Date: Fri Mar 4 10:45:29 2016 +0100
Merge branch 'master' into lucene6
commit cd33921ac742ef9fb351012eff35f3c7dbda7264
Author: Robert Muir <rmuir@apache.org>
Date: Thu Mar 3 23:58:37 2016 -0500
fix hunspell dictionary loading
commit c7fdbd837b01f7defe9cb1c24e2ec65604b0dc96
Merge: 4d4190f d8948ba
Author: Robert Muir <rmuir@apache.org>
Date: Thu Mar 3 23:41:53 2016 -0500
Merge branch 'master' into lucene6
commit 4d4190fd82601aaafac6b8254ccb3edf218faa34
Author: Robert Muir <rmuir@apache.org>
Date: Thu Mar 3 23:39:14 2016 -0500
remove nocommit
commit 77ca69e288b1a41aa9595c921ed166c272a00ea8
Author: Robert Muir <rmuir@apache.org>
Date: Thu Mar 3 23:38:24 2016 -0500
clean up numericutils vs legacynumericutils
commit a466d696fbaad04b647ffbc0857a9439b583d0bf
Author: Robert Muir <rmuir@apache.org>
Date: Thu Mar 3 23:32:43 2016 -0500
upgrade spatial4j
commit 5412c747a8cfe638bacedbc8233163cb75cc3dc5
Author: Robert Muir <rmuir@apache.org>
Date: Thu Mar 3 23:19:28 2016 -0500
move to 6.0.0-snapshot-8eada27
commit b32bfe924626b87e540692375ece09e7c2edb189
Author: Adrien Grand <jpountz@gmail.com>
Date: Thu Mar 3 11:30:09 2016 +0100
Fix some test compile errors.
commit 6ccde35e9840b03c68d1a2cd47c7923a06edf64a
Author: Adrien Grand <jpountz@gmail.com>
Date: Thu Mar 3 11:25:51 2016 +0100
Current Lucene version is 6.0.0.
commit f62e1015d931b4cc04c778298a8fa1ba65e97ad9
Author: Adrien Grand <jpountz@gmail.com>
Date: Thu Mar 3 11:20:48 2016 +0100
Fix compile errors in NGramTokenFilterFactory.
commit 6837c6eabf96075f743649da9b9b52dd39611c58
Author: Adrien Grand <jpountz@gmail.com>
Date: Thu Mar 3 10:50:59 2016 +0100
Fix the edge ngram tokenizer/filter.
commit ccd7f070de5efcdfbeb34b9555c65c4990bf1ba6
Author: Adrien Grand <jpountz@gmail.com>
Date: Thu Mar 3 10:42:44 2016 +0100
The missing value is now accessible through a getter.
commit bd3b77f9b28e5b05daa3d49683a9922a6baf2963
Author: Adrien Grand <jpountz@gmail.com>
Date: Thu Mar 3 10:41:51 2016 +0100
Remove IndexCacheableQuery.
commit 05f3091c347aeae80eeb16349ac51d2b53cf86f7
Author: Adrien Grand <jpountz@gmail.com>
Date: Thu Mar 3 10:39:43 2016 +0100
Fix compilation of function_score queries.
commit 81cda79a2431ac78f56b0cc5a5765387f662d801
Author: Adrien Grand <jpountz@gmail.com>
Date: Thu Mar 3 10:35:02 2016 +0100
Fix compile errors in BlendedTermQuery.
commit 70994ce8dd1eca0b995870974a38e20f26f96a7b
Author: Robert Muir <rmuir@apache.org>
Date: Wed Mar 2 23:33:03 2016 -0500
add bug ID
commit 29d4f1a71f36f646b5a6060bed3db019564a279d
Author: Robert Muir <rmuir@apache.org>
Date: Wed Mar 2 21:02:32 2016 -0500
easy .store changes
commit 5e1a1e6fd665fa455e88d3a8987362fad5f44bb1
Author: Robert Muir <rmuir@apache.org>
Date: Wed Mar 2 20:47:24 2016 -0500
cleanups mostly around boosting
commit 333a669ec6c305ada5645d13ed1da0e19ec1d053
Author: Robert Muir <rmuir@apache.org>
Date: Wed Mar 2 20:27:56 2016 -0500
more simple fixes
commit bd5cd98a1e089c866b6b4a5e159400b110140ce6
Author: Robert Muir <rmuir@apache.org>
Date: Wed Mar 2 19:49:38 2016 -0500
more easy fixes and removal of ancient cruft
commit a68f419ee47da5f9c9ce5b372f01d707e902474c
Author: Robert Muir <rmuir@apache.org>
Date: Wed Mar 2 19:35:02 2016 -0500
cutover numerics
commit 4ca5dc1fa47dd5892db00899032133318fff3116
Author: Robert Muir <rmuir@apache.org>
Date: Wed Mar 2 18:34:18 2016 -0500
fix some constants
commit 88710a17817086e477c6c021ec346d0534b7fb88
Author: Robert Muir <rmuir@apache.org>
Date: Wed Mar 2 18:14:25 2016 -0500
Add spatial-extras jar as a core dependency
commit c8cd6726583e5ce3f546ed355d4eca037164a30d
Author: Robert Muir <rmuir@apache.org>
Date: Wed Mar 2 18:03:33 2016 -0500
update to lucene 6 jars
The cluster stats api now returns counts for each node role. The `master_data`, `master_only`, `data_only` and `client` fields have been removed from the response in favour of `master`, `data`, `ingest` and `coordinating_only`. The same node can have multiple roles, hence contribute to multiple roles counts. Every node is implicitly a coordinating node, so whenever a node has no explicit roles, it will be counted as coordinating only.
_cat/nodes used to return `c` for client node or `d` for data node as part of the node.role column. This commit changes it to return `m` for master eligible, `d` for data and/or `i` for ingest. A node with no explicit roles will be a coordinating only node and marked with `-`. A node can obviously have multiple roles. The master column has been adapted to return only whether a node is the current master (`*`) or not (`-`).
The field name is a required argument for all suggesters, but
it was specified via a field() setter in SuggestionBuilder so far.
This changes field name to being a mandatory constructor argument
and lets suggestion builders throw an error if field name is missing
or the empty string.
Use 'includeSegmentFileSizes' as the flag name to report disk usage.
Added test that verifies reported segment disk usage is growing accordingly after adding a document.
Documentation: Reference the new parameter as part of indices stats.
Currently we suggesting users create a Node (using NodeBuilder in 2.x) to have a client that is capable of keeping up-to-date information. This is generally a bad idea as it means elasticsearch has no control over eg max heap size or gc settings, and is also problematic for users because they must deal with dependency collisions (and in 2.x+ dependencies of elasticsearch itself).
A better alternative, and what we should document, is to run a local elasticsearch server using bin/elasticsearch, and then use the transport client to connect to that local node. This local connection is virtually free, and allows the client code to be completely isolated from the elasticsearch process. Plugins are then also easy to deal with: just install them in elasticsearch as usual.
Related to #16679
ba5be0332d removed support for degrading
to slf4j and j.u.l but didn't document this as a breaking change because
it is only breaking for folks using Elasticsearch's jar as a java client.
People do that so this counts as a breaking change.
Also, if anyone was brave enough to try and replace log4j on an installed
version of Elasticsearch that will no longer work and this documents that
as well. It doens't get a full heading and instead lives with the java
client notes. Mostly because I can't imagine it worked consistently enough
for anyone to actually do it in the first place. We just never tested it
well enough to make sure we didn't break it after it was implemented.
Currently we suggesting users create a Node (using NodeBuilder in 2.x) to have a client that is capable of keeping up-to-date information. This is generally a bad idea as it means elasticsearch has no control over eg max heap size or gc settings, and is also problematic for users because they must deal with dependency collisions (and in 2.x+ dependencies of elasticsearch itself).
A better alternative, and what we should document, is to run a local elasticsearch server using bin/elasticsearch, and then use the transport client to connect to that local node. This local connection is virtually free, and allows the client code to be completely isolated from the elasticsearch process. Plugins are then also easy to deal with: just install them in elasticsearch as usual.
Closes#15383
This commit adds a note to the migration docs regarding the reduction of
the Groovy dependencies from the groovy-all artifact to the groovy
artifact that was previously done in
180ab2493e.
Closes#16858
We are using `2.0.0` today but Azure team now recommends:
```xml
<dependency>
<groupId>com.microsoft.azure</groupId>
<artifactId>azure-storage</artifactId>
<version>4.0.0</version>
</dependency>
```
This new version fix the timeout issues we have seen with azure storage although #15080 adds a timeout support.
Azure storage client 2.0.0 was not passing correctly this value when it was calling Azure services.
Note that the timeout is a server side timeout and not client side timeout.
It means that it will raise only a timeout when:
* upload of blob is complete
* if azure service is not able to process the blob (and store it) within a given time range.
In which case it will raise an exception which elasticsearch can deal with:
```
java.io.IOException
at __randomizedtesting.SeedInfo.seed([91BC11AEF16E073F:6886FA5308FCE4D8]:0)
at com.microsoft.azure.storage.core.Utility.initIOException(Utility.java:643)
at com.microsoft.azure.storage.blob.BlobOutputStream.writeBlock(BlobOutputStream.java:444)
at com.microsoft.azure.storage.blob.BlobOutputStream.access$000(BlobOutputStream.java:53)
at com.microsoft.azure.storage.blob.BlobOutputStream$1.call(BlobOutputStream.java:388)
at com.microsoft.azure.storage.blob.BlobOutputStream$1.call(BlobOutputStream.java:385)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
at java.lang.Thread.run(Thread.java:745)
Caused by: com.microsoft.azure.storage.StorageException: Operation could not be completed within the specified time.
at com.microsoft.azure.storage.StorageException.translateException(StorageException.java:89)
at com.microsoft.azure.storage.core.StorageRequest.materializeException(StorageRequest.java:305)
at com.microsoft.azure.storage.core.ExecutionEngine.executeWithRetry(ExecutionEngine.java:175)
at com.microsoft.azure.storage.blob.CloudBlockBlob.uploadBlockInternal(CloudBlockBlob.java:1006)
at com.microsoft.azure.storage.blob.CloudBlockBlob.uploadBlock(CloudBlockBlob.java:978)
at com.microsoft.azure.storage.blob.BlobOutputStream.writeBlock(BlobOutputStream.java:438)
... 9 more
```
The following code was used to test this against Azure platform:
```java
public void testDumb() throws URISyntaxException, StorageException, IOException, InvalidKeyException {
String connectionString = "MY-AZURE-STRING";
CloudStorageAccount storageAccount = CloudStorageAccount.parse(connectionString);
CloudBlobClient client = storageAccount.createCloudBlobClient();
client.getDefaultRequestOptions().setTimeoutIntervalInMs(1000);
CloudBlobContainer container = client.getContainerReference("dumb");
container.createIfNotExists();
CloudBlockBlob blob = container.getBlockBlobReference("blob");
File sourceFile = File.createTempFile("sourceFile", ".tmp");
try {
int fileSize = 10000000;
byte[] buffer = new byte[fileSize];
Random random = new Random();
random.nextBytes(buffer);
logger.info("Generate local file");
FileOutputStream fos = new FileOutputStream(sourceFile);
fos.write(buffer);
fos.close();
logger.info("End generate local file");
FileInputStream fis = new FileInputStream(sourceFile);
logger.info("Start uploading");
blob.upload(fis, fileSize);
logger.info("End uploading");
}
finally {
if (sourceFile.exists()) {
sourceFile.delete();
}
}
}
```
With 2.0.0, the above code was not raising any exception. With 4.0.0, the exception is now thrown correctly.
The default timeout is 5 minutes. See https://github.com/Azure/azure-storage-java/blob/master/microsoft-azure-storage/src/com/microsoft/azure/storage/core/Utility.java#L352-L375Closes#12567.
Release notes from 2.0.0:
* Removed deprecated table AtomPub support.
* Removed deprecated constructors which take service clients in favor of constructors which take credentials.
* Added support for "Add" permissions on Blob SAS.
* Added support for "Create" permissions on Blob and File SAS.
* Added support for IP Restricted SAS and Protocol SAS.
* Added support for Account SAS to all services.
* Added support for Minute and Hour Metrics to FileServiceProperties and added support for File Metrics to CloudAnalyticsClient.
* Removed deprecated startCopyFromBlob() on CloudBlob. Use startCopy() instead.
* Removed deprecated Credentials and StorageKey classes. Please use the appropriate methods on StorageCredentialsAccountAndKey instead.
* Fixed a bug in table where a select on a non-existent field resulted in a null reference exception if the corresponding field in the TableEntity was not nullable.
* Fixed a bug in table where JsonParser was automatically closing the response stream before it was completely drained causing socket exhaustion.
* Fixed a bug in StorageCredentialsAccountAndKey.updateKey(String) which prevented valid keys from being set.
* Added CloudBlobContainer.listBlobs(final String, final boolean) method.
* Fixed a bug in blob where using AccessConditions on block blob uploads larger than 64MB done with the upload* methods or block blob uploads done openOutputStream with would fail if the blob did not already exist.
* Added support for setting a proxy per request. Proxy can be set on an OperationContext instance and will be used when that instance is passed to the request method.
* Added support for SAS to the Azure File service.
* Added support for Append Blob.
* Added support for Access Control Lists (ACL) to File Shares.
* Added support for getting and setting of CORS rules to File service.
* Added support for ShareStats to File Shares.
* Added support for copying an Azure File to another Azure File or a Block Blob asynchronously, and aborting Azure File copy operations asynchronously.
* Added support for copying a Blob to an Azure File asynchronously.
* Added support for setting a maximum quota property on a File Share.
* Removed deprecated AuthenticationScheme and its getter and setter. In the future only SharedKey will be used.
* Removed deprecated getter/setters for all request option properties on the service clients. Please use the default request options getter/setters instead.
* Removed getSubDirectoryReference() for blob directories and file directories. Use getDirectoryReference() instead.
* Removed getEntityClass() in TableQuery. Please use getClazzType() instead.
* Added client-side verification for lease duration and break periods.
* Deprecated the setters in table for timestamp as this property is only modifiable by the service.
* Deprecated startCopyFromBlob() on CloudBlob. Use startCopy() instead.
* Deprecated the Credentials and StorageKey classes. Please use the appropriate methods on StorageCredentialsAccountAndKey instead.
* Deprecated constructors which take service clients in favor of constructors which take credentials.
* Fixed a bug where the DateBackwardCompatibility flag was not applied if set on the CloudTableClient default request options.
* Changed library behavior to retry all exceptions thrown when parsing a response object.
* Changed behavior to stop removing query parameters passed in with the resource URI if that URI contains a SAS token. Some query parameters such as comp, restype, snapshot and api-version will still be removed.
* Added support for logging StringToSign to SharedKey and SAS.
* **Added a connect timeout to prevent hangs when establishing the network connection.**
* **Made performance enhancements to the BlobOutputStream class.**
* Fixed a bug where maximum execution time was ignored for file, queue, and table services.
* **Changed the socket timeout to be set to the service side timeout plus 5 minutes when maximum execution time is not set.**
* **Changed the socket timeout to default to 5 minutes rather than infinite when neither service side timeout or maximum execution time are set.**
* Fixed a bug where MD5 was calculated for commitBlockList even though UseTransactionalMD5 was set to false.
* Fixed a bug where selecting fields that did not exist returned an error rather than an EntityProperty with a null value.
* Fixed a bug where table entities with a single quote in their partition or row key could be inserted but not operated on in any other way.
* Fixed a bug for all listing API's where next() would sometimes throw an exception if hasNext() had not been called even if there were more elements to iterate on.
* Added sequence number to the blob properties. This is populated for page blobs.
* Creating a page blob sets its length property.
* Added support for page blob sequence numbers and sequence number access conditions.
* Fixed a bug in abort copy where the lease access condition was not sent to the service.
* Fixed an issue in startCopyFromBlob where if the URI of the source blob contained certain non-ASCII characters they would not be encoded appropriately. This would result in Authorization failures.
* Fixed a small performance issue in XML serialization.
* Fixed a bug in BlobOutputStream and FileOutputStream where flush added data to a request pool rather than immediately committing it to the Azure service.
* Refactored to remove the blob, queue, and file package dependency on table in the error handling code.
* Added additional client-side logging for REST requests, responses, and errors.
Closes#15976.
This commit removes the system property "es.useLinkedTransferQueue" that
defaulted to false and was used to control the queue implementation used
in a few places.
Closes#16786
Java NIO has the notion of gathering writes. These are writes that
gather data from multiple buffers into a single channel. These gathering
writes in Netty have been enabled by default with the possibility to
disable them using "es.netty.gathering". This flag was added in case
having gathering writes on by default did not work out. We have not
published this ability and sufficient time has passed to render
judgement that using gathering writes is okay.
Closes#16774
Elasticsearch should reject ids that are this long, to ensure a document
always remains retrievable for clients that impose a maximum URI length
Closes#16034
This commit removes the es.max-open-files flag as the same information
can be obtained from the cluster nodes info API, and is warn logged on
startup if it's set too low anyway.
Closes#16757
2.x has show so far that running with security manager is the way to go.
This commit make this non-optional. Users that need to pass their own rules
can still do this via the system configuration for the security manager. They
can even opt out of all security that way.
This commit moves IndicesRequestCache into o.e.indics and makes all API in this
class package private. All references to SearchReqeust, SearchContext etc. have been factored
out and relevant glue code has been added to IndicesService. The IndicesRequestCache is not a
simple class without any hard dependencies on ThreadPool nor SearchService or IndexShard. This now
allows to add unittests.
This commit also removes two settings `indices.requests.cache.clean_interval` and `indices.fielddata.cache.clean_interval`
in favor of `indices.cache.clean_interval` which cleans both caches.
This is a simple port of the mapper attachment plugin to the ingest
functionality, no new features. The only option is to limit
the number of chars to prevent indexing of huge documents.
Fields can be selected in the processor as well.
Close#16303
The cat API previously used the Content-Type header field for
determining the media type of the response. This is in opposition to the
HTTP spec which specifies the Accept header field for this purpose. This
commit replaces the use of the Content-Type header field with the Accept
header field in the cat API.
Closes#14421
This processor is useful when all elements of a json array need to be processed in the same way.
This avoids that a processor needs to be defined for each element in an array.
Also it is very likely that it is unknown how many elements are inside an json array.
Clinton Gormley
Areek Zillur
Alexander Kazakov
Christoph Büscher
Jason Tedor
Ryan Ernst
Daniel Mitterdorfer
Ed Winn
Lee Hinman
Yannick Welsch
Robin Clarke
Nicholas Knize
Andrew Cholakian
Martijn van Groningen
Christopher Taylor
Robert Muir
javanna
Nik Everett
Jim Ferenczi
DeDe Morton
Vector241-Eric
James Kerr
Geoff Wagstaff
Adrien Grand
Michael McCandless
Camilo Díaz Repka
Andrey Ryaguzov
David Pilato
Dominik Stadler
Itamar Syn-Hershko
Ali Beyad
evanfreed
Konstantin Delchev
Henrik Nordvik
Boaz Leskes
Robert
gaelL
Colin Goodheart-Smithe
Simon Willnauer
Simon Willnauer
Dongjoon Hyun
Lee Hinman
Alexander Reelsen
Spencer
Chris Moultrie
Andrej Kazakov
Yannick Welsch
Russell Cohen
Tal Levy