A voting-only master-eligible node is a node that can participate in master elections but will not act
as a master in the cluster. In particular, a voting-only node can help elect another master-eligible
node as master, and can serve as a tiebreaker in elections. High availability (HA) clusters require at
least three master-eligible nodes, so that if one of the three nodes is down, then the remaining two
can still elect a master amongst them-selves. This only requires one of the two remaining nodes to
have the capability to act as master, but both need to have voting powers. This means that one of
the three master-eligible nodes can be made as voting-only. If this voting-only node is a dedicated
master, a less powerful machine or a smaller heap-size can be chosen for this node. Alternatively, a
voting-only non-dedicated master node can play the role of the third master-eligible node, which
allows running an HA cluster with only two dedicated master nodes.
Closes#14340
Co-authored-by: David Turner <david.turner@elastic.co>
Today the `ClusterFormationFailureHelper` says `... discovery will continue
using ... from last-known cluster state` and lists all the nodes in the
last-known cluster state. In fact we ignore the master-ineligible nodes in the
last-known cluster state during discovery. This commit fixes this by listing
only the master-eligible nodes from the cluster state in this message.
If the master removes the relocating shard, but recovery isn't aware of
it, then we can enter an invalid state where ReplicationTracker does not
include the local shard.
Today we assert that a warning is logged after no more than
`discovery.cluster_formation_warning_timeout`, but the deterministic scheduler
adds a small amount of extra randomness to the timing of future events, causing
the following build to fail:
./gradlew :server:test --tests "org.elasticsearch.cluster.coordination.CoordinatorTests.testLogsWarningPeriodicallyIfClusterNotFormed" -Dtests.seed=DF35C28D4FA9EE2D
This commit adds an allowance for this extra time.
After two recent changes (#38824 and #33888), the _cat/indices API
no longer report information for active recovering indices and
non-replicated closed indices. It also misreport replicated closed
indices that are potentially not authorized for the user.
This commit changes how the cat action works by first using the
Get Settings API in order to resolve authorized indices. It then uses
the Cluster State, Cluster Health and Indices Stats APIs to retrieve
information about the indices.
Closes#39933
This refactors AggregatorTestCase to allow testing mock scripts.
The main change is to QueryShardContext. This was previously mocked,
but to get the ScriptService you have to invoke a final method
which can't be mocked.
Instead, we just create a mostly-empty QueryShardContext and populate
the fields that are needed for testing. It also introduces a few
new helper methods that can be overridden to change the default
behavior a bit.
Most tests should be able to override getMockScriptService() to supply
a ScriptService to the context, which is later used by the aggs.
More complicated tests can override queryShardContextMock() as before.
Adds a test to MaxAggregatorTests to test out the new functionality.
At the end of a peer recovery the primary wants to mark the replica as in-sync. For that the
persisted local checkpoint of the replica needs to have caught up with the global checkpoint on the
primary. If translog durability is set to ASYNC, this means that information about the persisted local
checkpoint can lag on the primary and might need to be explicitly fetched through a global
checkpoint sync action. Unfortunately, that action will only be triggered after 30 seconds, and, even
worse, will only run based on what the in-sync shard copies say (see
IndexShard.maybeSyncGlobalCheckpoint). As the replica has not been marked as in-sync yet, it is
not taken into consideration, and the primary might have its global checkpoint equal to the max seq
no, so it thinks nothing needs to be done.
Closes#43486
Long and Double ValuesSource set the current document on the script
before executing, but Bytes was missing this method call. That meant
it was possible to generate an OutOfBoundsException when using
a "value" script (field + script) on keyword or other bytes
fields.
This adds in the method call, and a few yaml tests to verify correct
behavior.
This commit performs the proper restore of network disruption.
Previously disruptionScheme.stopDisrupting() was called that does not
ensure that connectivity between cluster nodes is restored. The test
was checking that the cluster has green status, but it was not checking
that connectivity between nodes is restored.
Here we switch to internalCluster().clearDisruptionScheme(true) which
performs both checks before returning.
Similar to #42798Closes#42051
(cherry picked from commit cd1ed662f847a0055ede7dfbd325e214ec4d1490)
This commit replaces usages of Streamable with Writeable for the
AcknowledgedResponse and its subclasses, plus associated actions.
Note that where possible response fields were made final and default
constructors were removed.
This is a large PR, but the change is mostly mechanical.
Relates to #34389
Backport of #43414
Unsupported HTTP methods are detected during requests dispatching
which generates an appropriate error response. Sadly, this error is
never sent back to the client because the method of the original
request is checked again in DefaultRestChannel which throws again
an IllegalArgumentException that is never handled.
This pull request changes the DefaultRestChannel so that the latest
exception is swallowed, allowing the error message to be sent back
to the client. It also eagerly adds the objects to close to the toClose
list so that resources are more likely to be released if something
goes wrong during the response creation and sending.
We currently assert that adding deletion tombstones to Lucene must always succeed if it's not a
tragic exception, and the same should also hold true for NOOP tombstones. We rely on this
assumption, as without this, we have the risk of creating gaps in the history, which will break
operation-based recoveries and CCR.
Depending on git configuration, line feed on checked out files may be
platform dependent, which causes problems to some msearch tests as the
line separator must always be `/n`. With this change we move two files
to the test code so that we control exactly what line separator is used,
given that the corresponding tests fail on windows.
Closes#43464
Removes `@TestLogging` annotations in `*DisruptionIT` tests, so that the only
tests with annotations are those with open issues. Also adds links to the open
issues in the remaining cases.
Relates #43403
The current toXContent implementation can fail when the superclasses toXContent
is called (see #43423). This change makes sure that
DefaultShardOperationFailedException#toXContent is final and implementations
need to add special fields in #innerToXContent. All implementations should write
to self-contained xContent objects. Also adding a test for xContent deserialization
to CloseIndexResponseTests.
Closes#43423
Today when searching for an exclusive range the java date math parser rounds up the value
with the granularity of the operation. So when searching for values that are greater than
"now-2M" the parser rounds up the operation to "now-1M". This behavior was introduced when
we migrated to java date but it looks like a bug since the joda math parser rounds up values
but only when a rounding is used. So "now/M" is rounded to "now-1ms" (minus 1ms to get the largest inclusive value)
in the joda parser if the result should be exclusive but no rounding is applied if the input
is a simple operation like "now-1M". This change restores the joda behavior in order to have
a consistent parsing in all versions.
Closes#43277
Currently the fromXContent logic for reindex requests is implemented in
the rest action. This is inconsistent with other requests where the
logic is implemented in the request. Additionally, it requires access to
the rest action in order to parse the request. This commit moves the
logic and tests into the ReindexRequest.
AggregatorTestCase will NPE if only a single, null MappedFieldType
is provided (which is required to simulate an unmapped field). While
it's possible to test unmapped fields by supplying other, non-related
field types... that's clunky and unnecessary. AggregatorTestCase
just needs to filter out null field types when setting up.
Today the `DisruptibleMockTransport` always allows a connection to a node to be
established, and then fails requests sent to that node such as the subsequent
handshake. Since #42342, we log handshake failures on an open connection as a
warning, and this makes the test logs rather noisy. This change fails the
connection attempt first, avoiding these unrealistic warnings.
The test needed adaption after #43205, as the ReplicationTracker now distinguishes between the
knowledge of the persisted global checkpoint and the computed global checkpoint on the primary
Follow-up to #43205
Local and global checkpoints currently do not correctly reflect what's persisted to disk. The issue is
that the local checkpoint is adapted as soon as an operation is processed (but not fsynced yet). This
leaves room for the history below the global checkpoint to still change in case of a crash. As we rely
on global checkpoints for CCR as well as operation-based recoveries, this has the risk of shard
copies / follower clusters going out of sync.
This commit required changing some core classes in the system:
- The LocalCheckpointTracker keeps track now not only of the information whether an operation has
been processed, but also whether that operation has been persisted to disk.
- TranslogWriter now keeps track of the sequence numbers that have not been fsynced yet. Once
they are fsynced, TranslogWriter notifies LocalCheckpointTracker of this.
- ReplicationTracker now keeps track of the persisted local and persisted global checkpoints of all
shard copies when in primary mode. The computed global checkpoint (which represents the
minimum of all persisted local checkpoints of all in-sync shard copies), which was previously stored
in the checkpoint entry for the local shard copy, has been moved to an extra field.
- The periodic global checkpoint sync now also takes async durability into account, where the local
checkpoints on shards only advance when the translog is asynchronously fsynced. This means that
the previous condition to detect inactivity (max sequence number is equal to global checkpoint) is
not sufficient anymore.
- The new index closing API does not work when combined with async durability. The shard
verification step is now requires an additional pre-flight step to fsync the translog, so that the main
verify shard step has the most up-to-date global checkpoint at disposition.
The RemoteClusterService should close the current
RemoteClusterConnection and should build it again if
the seeds are changed, similarly to what is done when
the ping interval or the compression settings are changed.
Closes#37799
The fact that SearchPhaseContext extends ActionListener makes it hard
to reason about when the original listener is notified and to trace
those calls. Also, the corresponding onFailure and onResponse were
only needed in two places, one each, where they can be replaced by a
more intuitive call, like sendSearchResponse for onResponse.
Today the fielddata for global ordinals re-creates docvalues readers of each segment
when building the iterator of a single segment. This is required because the lookup of
global ordinals needs to access the docvalues's TermsEnum of each segment to retrieve
the original terms. This also means that we need to create NxN (where N is the number of segment in the index) docvalues iterators
each time we want to collect global ordinal values. This wasn't an issue in previous versions since docvalues readers are stateless
before 6.0 so they are reused on each segment but now that docvalues are iterators we need to create a new instance each time
we want to access the values. In order to avoid creating too many iterators this change splits
the global ordinals fielddata in two classes, one that is used to cache a single instance per directory reader and one
that is created from the cached instance that can be used by a single consumer. The latter creates the TermsEnum of each segment
once and reuse them to create the segment's iterator. This prevents the creation of all TermsEnums each time we want to access
the value of a single segment, hence reducing the number of docvalues iterator to create to Nx2 (one iterator and one lookup per segment).
This commit removes some very old test logging annotations that appeared
to be added to investigate test failures that are long since closed. If
these are needed, they can be added back on a case-by-case basis with a
comment associating them to a test failure.
It's possible for the passed in `IndexMetaData` to be null (for
instance, cluster state passed in does not have the index in its
metadata) which in turn can cause a `NullPointerException` when
evaluating the conditions for an index. This commit adds null protection
and unit tests for this case.
Resolves#43296
While investigating memory consumption of deeply nested aggregations for #43091
the memory used to keep track of the doc ids and buckets in the BestBucketsDeferringCollector
showed up as one of the main contributor. In my tests half of the memory held in the
BestBucketsDeferringCollector is associated to segments that don't have matching docs
in the selected buckets. This is expected on fields that have a big cardinality since each
bucket can appear in very few segments. By allocating the builders lazily this change
reduces the memory consumption by a factor 2 (from 1GB to 512MB), hence reducing the
impact on gcs for these volatile allocations. This commit also switches the PackedLongValues.Builder
with a RoaringDocIdSet in order to handle very sparse buckets more efficiently.
I ran all my tests on the `geoname` rally track with the following query:
````
{
"size": 0,
"aggs": {
"country_population": {
"terms": {
"size": 100,
"field": "country_code.raw"
},
"aggs": {
"admin1_code": {
"terms": {
"size": 100,
"field": "admin1_code.raw"
},
"aggs": {
"admin2_code": {
"terms": {
"size": 100,
"field": "admin2_code.raw"
},
"aggs": {
"sum_population": {
"sum": {
"field": "population"
}
}
}
}
}
}
}
}
}
}
````
This PR is a backport a of #43214 from v8.0.0
A number of the aggregation base classes have an abstract doEquals() and doHashCode() (e.g. InternalAggregation.java, AbstractPipelineAggregationBuilder.java).
Theoretically this is so the sub-classes can add to the equals/hashCode and don't need to worry about calling super.equals(). In practice, it's mostly just confusing/inconsistent. And if there are more than two levels, we end up with situations like InternalMappedSignificantTerms which has to call super.doEquals() which defeats the point of having these overridable methods.
This PR removes the do versions and just use equals/hashCode ensuring the super when necessary.
* Follow up to #42109:
* Adjust test to only check that interface lookup by name works not actually lookup IPs which is brittle since virtual interfaces can be destroyed/created by Docker while the tests are running
Co-authored-by: Jason Tedor <jason@tedor.me>
* Return 0 for negative "free" and "total" memory reported by the OS
We've had a situation where the MX bean reported negative values for the
free memory of the OS, in those rare cases we want to return a value of
0 rather than blowing up later down the pipeline.
In the event that there is a serialization or creation error with regard
to memory use, this adds asserts so the failure will occur as soon as
possible and give us a better location for investigation.
Resolves#42157
* Fix test passing in invalid memory value
* Fix another test passing in invalid memory value
* Also change mem check in MachineLearning.machineMemoryFromStats
* Add background documentation for why we prevent negative return values
* Clarify comment a bit more
This PR reverts #35230.
Previously, we reply on soft-deletes to fill the mismatch between the
version map and the Lucene index. This is no longer needed after #43202
where we rebuild the version map when opening an engine. Moreover,
PrunePostingsMergePolicy can prune _id of soft-deleted documents out of
order; thus the lookup result including soft-deletes sometimes does not
return the latest version (although it's okay as we only use a valid
result in an engine).
With this change, we use only live documents in Lucene to resolve the
indexing strategy. This is perfectly safe since we keep all deleted
documents after the local checkpoint in the version map.
Closes#42979
Backport of: https://github.com/elastic/elasticsearch/pull/43222
This commit replaces usages of Streamable with Writeable for the
SingleShardRequest / TransportSingleShardAction classes and subclasses of
these classes.
Note that where possible response fields were made final and default
constructors were removed.
Relates to #34389
Today the `ClusterFormationFailureHelper` does not include the local node in
the list of nodes it claims to have discovered. This means that it sometimes
reports that it has not discovered a quorum when in fact it has. This commit
adds the local node to the set of discovered nodes.
Today we suppress election attempts on master-eligible nodes that are not in
the voting configuration. In fact this restriction is not necessary: any
master-eligible node can safely become master as long as it has a fresh enough
cluster state and can gather a quorum of votes. Moreover, this restriction is
sometimes undesirable: there may be a reason why we do not want any of the
nodes in the voting configuration to become master.
The reason for this restriction is as follows. If you want to shut the master
down then you might first exclude it from the voting configuration. When this
exclusion succeeds you might reasonably expect that a new master has been
elected, since the voting config exclusion is almost always a step towards
shutting the node down. If we allow nodes outside the voting configuration to
be the master then the excluded node will continue to be master, which is
confusing.
This commit adjusts the logic to allow master-eligible nodes to attempt an
election even if they are not in the voting configuration. If such a master is
successfully elected then it adds itself to the voting configuration. This
commit also adjusts the logic that causes master nodes to abdicate when they
are excluded from the voting configuration, to avoid the confusion described
above.
Relates #37712, #37802.
With this change, we will rebuild the live version map and local
checkpoint using documents (including soft-deleted) from the safe commit
when opening an internal engine. This allows us to safely prune away _id
of all soft-deleted documents as the version map is always in-sync with
the Lucene index.
Relates #40741
Supersedes #42979
This commit modifies InternalTestCluster to allow using client() and
other operations inside a RestartCallback (onStoppedNode typically).
Restarting nodes are now removed from the map and thus all
methods now return the state as if the restarting node does not exist.
This avoids various exceptions stemming from accessing the stopped
node(s).
Adds `equals()` and `hashcode()` methods to `DiscoveryNodeRole` to compare
these objects' values for equality, and adds a field to allow us to distinguish
unknown roles from known ones with the same name and abbreviation, for clearer
test failures.
Relates #43175
With this change we only have to add one line to add a new version.
The intent is to make it less error prone and easier to write a script
to automate the process.
There are a few tests within NodeTests that submit items to the
threadpool and then close the node. The tests are designed to check
how running tasks are affected during node close. These tests can cause
CI failures since the submitted tasks may not be running when the node
is closed and then execute after the thread context is closed, which
triggers an unexpected exception. This change ensures the threads are
running so we avoid the unexpected exception and can test these cases.
The test of task submittal while a node is closing is also important so
an additional but muted test has been added that tests the case where a
task may be getting submitted while the node is closing and ensuring we
do not trigger anything unexpected in these cases.
Relates #42774
Relates #42577
This PR proposes to model big integers as longs (and big decimals as doubles)
in the context of dynamic mappings.
Previously, the dynamic mapping logic did not recognize big integers or
decimals, and would an error of the form "No matching token for number_type
[BIG_INTEGER]" when a dynamic big integer was encountered. It now accepts these
numeric types and interprets them as 'long' and 'double' respectively. This
allows `dynamic_templates` to accept and and remap them as another type such as
`keyword` or `scaled_float`.
Addresses #37846.
The test fails if querying the roles via a transport client, since the
transport client does not have the plugin necessary to interpret the additional
role correctly. This commit adds this plugin to the transport client used.
Relates #43175Fixes#43223
Currently the randomization of the q.b. in these tests can create query strings
that can cause caching to be disabled for this query if we query all fields and
there is a date field present. This is pretty much an anomaly that we shouldn't
generally test for in the "testToQuery" tests where cache policies are checked.
This change makes sure we don't create offending query strings so the cache
checks never hit these cases and adds a special test method to check this edge
case.
Closes#43112
roundUp parsers were losing the composite pattern information when new
JavaDateFormatter was created from methods withLocale or withZone.
The roundUp parser should be preserved when calling these methods. This is the same approach in withLocale/Zone methods as in daa2ec8a60/server/src/main/java/org/elasticsearch/common/time/JavaDateFormatter.javacloses#42835
Given the significant performance impact that NIOFS has when term dicts are
loaded off-heap this change enforces FstLoadMode#AUTO that loads term dicts
off heap only if the underlying index input indicates a memory map.
Relates to #43150
As the ValuesSourceType evolves, it is important to be
confident that new enum constants do not break
backwards-compatibility on the stream. Having dedicated
unit tests for this class will help be sure of that.
This change adds the terms index (`.tip`) to the list of extensions
that are memory-mapped by hybridfs. These files used to be accessed
only once to load the terms index on-heap but since #42838 they can
now be used to read the binary FST directly so it is benefical to
memory-map them instead of accessing them via NIO.
Fixes an issue where tests would sometimes hang for 5 seconds when restarting a node. The reason
is that the SeedHostsResolver is blockingly waiting on a result for the full 5 seconds when the
corresponding threadpool is shut down.
We are still using `FileSwitchDirectory` in the case a user configures file based pre-load of mmaps. This is trappy for multiple reasons if the both directories used by `FileSwitchDirectory` point to the same filesystem directory. One issue is LUCENE-8835 that cause issues like #37111 - unless LUCENE-8835 isn't fixed we should not use it in elasticsearch. Instead we use a similar trick as we use for HybridFS and subclass mmap directory directly.
When set to false, allowPartialSearchResults option does not check if the
shard failures have been reseted to null. The atomic array, that is used to record
shard failures, is filled with a null value if a successful request on a shard happens
after a failure on a shard of another replica. In this case the atomic array is not empty
but contains only null values so this shouldn't be considered as a failure since all
shards are successful (some replicas have failed but the retries on another replica succeeded).
This change fixes this bug by checking the content of the atomic array and fails the request only
if allowPartialSearchResults is set to false and at least one shard failure is not null.
Closes#40743
Currently suggesters return null values on empty shards. Usually this gets replaced
by results from other non-epmty shards, but if the index is completely epmty (e.g. after
creation) the search responses "suggest" is also "null" and we don't render a corresponding
output in the REST response. This is an irritating edge case that requires special handling on
the user side (see #42473) and should be fixed.
This change makes sure every suggester type (completion, terms, phrase) returns at least an
empty skeleton suggestion output, even for empty shards. This way, even if we don't find
any suggestions anywhere, we still return and output the empty suggestion.
Closes#42473
When a search on some indices takes a long time, it may cause problems to other indices that are being searched as part of the same search request and being written to as well, because their search context needs to stay open for a long time. This is especially a problem when searching against throttled and non-throttled indices as part of the same request. The problem can be generalized though: this may happen whenever read-only indices are searched together with indices that are being written to. Search contexts staying open for a long time is only an issue for indices that are being written to, in practice.
This commit splits the search in two sub-searches: one for read-only indices, and one for ordinary indices. This way the two don't interfere with each other. The split is done only when size is greater than 0, no scroll is provided and query_then_fetch is used as search type. Otherwise, the search executes like before. Note that the returned num_reduce_phases reflect the number of reduction phases that were run. If the search is split in two, there are three reductions: one non-final for each search, and a final one that merges the results of the previous two.
Closes#40900
The SearchRequest#crossClusterSearch method is currently used only as
part of cross cluster search request, when minimizing roundtrips.
It will soon be used also when splitting a search into two: one for
throttled and one for non throttled indices. It will probably be used
for other usecases as well in the future, hence it makes sense to generalize its name to subSearchRequest.
Though not in use in elasticsearch currently, it seems surprising that
ThreadPool.scheduler().scheduleAtFixedRate would hang. A recurring
scheduled task is never completed (except on failure) and we test for
exceptions using RunnableFuture.get(), which hangs for periodic tasks.
Fixed by checking that task is done before calling .get().
Today the master eagerly reroutes the cluster as part of processing node joins.
However, it is not necessary to do this reroute straight away, and it is
sometimes preferable to defer it until later. For instance, when the master
wins its election it processes joins and performs a reroute, but it would be
better to defer the reroute until after the master has become properly
established.
This change defers this reroute into a separate task, and batches multiple such
tasks together.
#40741 introduced a merge policy that can drop the postings for the `_id`
field on soft deleted documents. The TermsSliceQuery assumes that every document
has has an entry in the postings for that field so it doesn't check if the terms
index exists or not. This change fixes this bug by checking if the terms index for
the `_id` field is null and ignore the segment entirely if it's the case. This should
be harmless since segments without an `_id` terms index should only contain soft deleted
documents.
Closes#42996
If linearizability checking fails with OOM (or other exception), we did
not get the serialized history written into the log, making it difficult
to debug in cases where the problem is hard to reproduce. Fixed to
always attempt dumping the serialized history.
Related to #42244
Both TransportAnalyzeAction and CategorizationAnalyzer have logic to build
custom analyzers for index-independent analysis. A lot of this code is duplicated,
and it requires the AnalysisRegistry to expose a number of internal provider
classes, as well as making some assumptions about when analysis components are
constructed.
This commit moves the build logic directly into AnalysisRegistry, reducing the
registry's API surface considerably.
Setting `auto` after the fuzzy operator (e.g. `"query": "foo~auto"`) in the `query_string`
does not take the length of the term into account when computing the distance and always use
a max distance of 1. This change fixes this disrepancy by ensuring that the term is passed when
the fuzziness is computed.
We respect allocation deciders, including the `MaxRetryAllocationDecider`, when
executing reroute commands. If you specify `?retry_failed=true` then the retry
counter is reset, but today this does not happen until after trying to execute
the reroute commands. This means that if an allocation has repeatedly failed,
but you want to take control and assign a shard to a particular node to work
around the repeated failures, you cannot execute the routing command in the
same call to `POST /_cluster/reroute` as the one that resets the failure
counter.
This commit fixes this by resetting the failure counter first, meaning that you
can now explicitly allocate a repeatedly-failed shard like this:
```
POST /_cluster/reroute?retry_failed=true
{
"commands": [
{
"allocate_replica": {
"index": "blahblah",
"shard": 2,
"node": "node-4"
}
}
]
}
```
Fixes#39546
This commit refactors put mapping request validation for reuse. The
concrete case that we are after here is the ability to apply effectively
the same framework to indices aliases requests. This commit refactors
the put mapping request validation framework to allow for that.
This commit fixes a test bug in the request validators random test. In
particular, an assertion was not properly nested in a guard that would
ensure that was at least one failure.
Relates #43000
When applying put mapping validators, we apply all the validators in the
collection. If a failure occurs, we collect that as a top-level
exception, and suppress any additional failures into the top-level
exception. However, if a request passes the validator after a top-level
exception has been collected, we would try to suppress a null exception
into the top-level exception. This is a violation of the
Throwable#addSuppressed API. This commit addresses this, and adds test
to cover the logic of collecting the failures when validating a put
mapping request.
Today we test for translog corruption by incrementing a byte by 1 somewhere in
a file, and verify that this leads to a `TranslogCorruptionException`.
However, we rely on _all_ corruptions leading to this exception in the
`RemoveCorruptedShardDataCommand`: this command fails if a translog file
corruption leads to a different kind of exception, and `EOFException` and
`NegativeArraySizeException` are both possible. This commit strengthens the
translog corruption detection tests by simulating the following:
- a random value is written
- the file is truncated
It also makes sure that we return a `TranslogCorruptionException` in all such
cases.
Fixes#42661
Backport of #42744
David Kyle
Yannick Welsch
David Turner
Nhat Nguyen
Jake Landis
Tanguy Leroux
Zachary Tong
Przemysław Witek
Armin Braun
Jim Ferenczi
Andrey Ershov
Martijn van Groningen
Luca Cavanna
Christoph Büscher
Yu
Andrei Stefan
Tim Brooks
sandmannn
synical
Jason Tedor
Lee Hinman
Christos Soulios
Paul Sanwald
Igor Motov
Henning Andersen
Alpar Torok
Jay Modi
Julie Tibshirani
Przemyslaw Gomulka
Simon Willnauer
Tal Levy
Alan Woodward
Alexander Reelsen
Jim Ferenczi
Vigya Sharma