Allow a field inside the data to be used as a tie breaker for events
that have the same timestamp.
The field is optional by default.
If used, the tie-breaker always requires a non-null value since it is
used inside `search_after` which requires a non-null value.
Fix#56824
(cherry picked from commit e5719ecb474b32730d93afdbb6834a32b0b2df8b)
The shrink action creates a shrunken index with the target number of shards.
This makes the shrink action data stream aware. If the ILM managed index is
part of a data stream the shrink action will make sure to swap the original
managed index with the shrunken one as part of the data stream's backing
indices and then delete the original index.
(cherry picked from commit 99aeed6acf4ae7cbdd97a3bcfe54c5d37ab7a574)
Signed-off-by: Andrei Dan <andrei.dan@elastic.co>
This deprecates `Rounding#round` and `Rounding#nextRoundingValue` in
favor of calling
```
Rounding.Prepared prepared = rounding.prepare(min, max);
...
prepared.round(val)
```
because it is always going to be faster to prepare once. There
are going to be some cases where we won't know what to prepare *for*
and in those cases you can call `prepareForUnknown` and stil be faster
than calling the deprecated method over and over and over again.
Ultimately, this is important because it doesn't look like there is an
easy way to cache `Rounding.Prepared` or any of its precursors like
`LocalTimeOffset.Lookup`. Instead, we can just build it at most once per
request.
Relates to #56124
Adds assertions to Netty to make sure that its threads are not polluted by thread contexts (and
also that thread contexts are not leaked). Moves the ClusterApplierService to use the system
context (same as we do for MasterService), which allows to remove a hack from
TemplateUgradeService and makes it clearer that applying CS updates is fully executing under
system context.
* Convert to date/datetime the result of numeric aggregations (min, max)
in Painless scripts
(cherry picked from commit f1de99e2a6fbf3806c4f2b6b809738aa8faa2d75)
This adds new plugin level circuit breaker for the ML plugin.
`model_inference` is the circuit breaker qualified name.
Right now it simply adds to the breaker when the model is loaded (and possibly breaking) and removing from the breaker when the model is unloaded.
Before to determine if a field is meta-field, a static method of MapperService
isMetadataField was used. This method was using an outdated static list
of meta-fields.
This PR instead changes this method to the instance method that
is also aware of meta-fields in all registered plugins.
Related #38373, #41656Closes#24422
We want to validate the DataStreams on creation to make sure the future backing
indices would not clash with existing indices in the system (so we can
always rollover the data stream).
This changes the validation logic to allow for a DataStream to be created
with a backing index that has a prefix (eg. `shrink-foo-000001`) even if the
former backing index (`foo-000001`) exists in the system.
The new validation logic will look for potential index conflicts with indices
in the system that have the counter in the name greater than the data stream's
generation.
This ensures that the `DataStream`'s future rollovers are safe because for a
`DataStream` `foo` of generation 4, we will look for standalone indices in the
form of `foo-%06d` with the counter greater than 4 (ie. validation will fail if
`foo-000006` exists in the system), but will also allow replacing a
backing index with an index named by prefixing the backing index it replaces.
(cherry picked from commit 695b242d69f0dc017e732b63737625adb01fe595)
Signed-off-by: Andrei Dan <andrei.dan@elastic.co>
Deleting expired data can take a long time leading to timeouts if there
are many jobs. Often the problem is due to a few large jobs which
prevent the regular maintenance of the remaining jobs. This change adds
a job_id parameter to the delete expired data endpoint to help clean up
those problematic jobs.
This makes it easier to debug where such tasks come from in case they are returned from the get tasks API.
Also renamed the last occurrence of waitForCompletion to waitForCompletionTimeout in get async search request.
This PR adds the initial Java side changes to enable
use of the per-partition categorization functionality
added in elastic/ml-cpp#1293.
There will be a followup change to complete the work,
as there cannot be any end-to-end integration tests
until elastic/ml-cpp#1293 is merged, and also
elastic/ml-cpp#1293 does not implement some of the
more peripheral functionality, like stop_on_warn and
per-partition stats documents.
The changes so far cover REST APIs, results object
formats, HLRC and docs.
Backport of #57683
This is a major refactor of the underlying inference logic.
The main refactor is now we are separating the model configuration and
the inference interfaces.
This has the following benefits:
- we can store extra things with the model that are not
necessary for inference (i.e. treenode split information gain)
- we can optimize inference separate from model serialization and storage.
- The user is oblivious to the optimizations (other than seeing the benefits).
A major part of this commit is removing all inference related methods from the
trained model configurations (ensemble, tree, etc.) and moving them to a new class.
This new class satisfies a new interface that is ONLY for inference.
The optimizations applied currently are:
- feature maps are flattened once
- feature extraction only happens once at the highest level
(improves inference + feature importance through put)
- Only storing what we need for inference + feature importance on heap
#47711 and #47246 helped to validate that monitoring settings are
rejected at time of setting the monitoring settings. Else an invalid
monitoring setting can find it's way into the cluster state and result
in an exception thrown [1] on the cluster state application (there by
causing significant issues). Some additional monitoring settings have
been identified that can result in invalid cluster state that also
result in exceptions thrown on cluster state application.
All settings require a type of either http or local to be
applicable. When a setting is changed, the exporters are automatically
updated with the new settings. However, if the old or new settings lack
of a type setting an exception will be thrown (since exporters are
always of type 'http' or 'local'). Arguably we shouldn't blindly create
and destroy new exporters on each monitoring setting update, but the
lifecycle of the exporters is abit out the scope this PR is trying to
address.
This commit introduces a similar methodology to check for validity as
#47711 and #47246 but this time for ALL (including non-http) settings.
Monitoring settings are not useful unless there an exporter with a type
defined. The type is used as dependent setting, such that it must
exist to set the value. This ensures that when any monitoring settings
changes that they can only get added to cluster state if the type
exists. If the type exists (and the other validations pass) then the
exporters will get re-built and the cluster state remains valid.
Tests have been included to ensure that all dynamic monitoring settings
have the type as dependent settings.
[1]
org.elasticsearch.common.settings.SettingsException: missing exporter type for [found-user-defined] exporter
at org.elasticsearch.xpack.monitoring.exporter.Exporters.initExporters(Exporters.java:126) ~[?:?]
When we force delete a DF analytics job, we currently first force
stop it and then we proceed with deleting the job config.
This may result in logging errors if the job config is deleted
before it is retrieved while the job is starting.
Instead of force stopping the job, it would make more sense to
try to stop the job gracefully first. So we now try that out first.
If normal stop fails, then we resort to force stopping the job to
ensure we can go through with the delete.
In addition, this commit introduces `timeout` for the delete action
and makes use of it in the child requests.
Backport of #57680
rewrite config on update if either version is outdated, credentials change,
the update changes the config or deprecated settings are found. Deprecated
settings get migrated to the new format. The upgrade can be easily extended to
do any necessary re-writes.
fixes#56499
backport #57648
For a rolling/mixed cluster upgrade (add new version to existing cluster
then shutdown old instances), the watches that ship by default
with monitoring may not get properly updated to the new version.
Monitoring watches can only get published if the internal state is
marked as dirty. If a node is not master, will also get marked as
clean (e.g. not dirty).
For a mixed cluster upgrade, it is possible for the new node to be
added, not as master, the internal state gets marked as clean so
that no more attempts can be made to publish the watches. This
happens on all new nodes. Once the old nodes are de-commissioned
one of the new version nodes in the cluster gets promoted to master.
However, that new master node (with out intervention like restarting
the node or removing/adding exporters) will never attempt to re-publish
since the internal state was already marked as clean.
This commit adds a cluster state listener to mark the resource dirty
when a node is promoted to master. This will allow the new resource
to be published without any intervention.
In #55592 and #55416, we deprecated the settings for enabling and disabling
basic license features and turned those settings into no-ops. Since doing so,
we've had feedback that this change may not give users enough time to cleanly
switch from non-ILM index management tools to ILM. If two index managers
operate simultaneously, results could be strange and difficult to
reconstruct. We don't know of any cases where SLM will cause a problem, but we
are restoring that setting as well, to be on the safe side.
This PR is not a strict commit reversion. First, we are keeping the new
xpack.watcher.use_ilm_index_management setting, introduced when
xpack.ilm.enabled was made a no-op, so that users can begin migrating to using
it. Second, the SLM setting was modified in the same commit as a group of other
settings, so I have taken just the changes relating to SLM.
* Remove duplicate ssl setup in sql/qa projects
* Fix enforcement of task instances
* Use static data for cert generation
* Move ssl testing logic into a plugin
* Document test cert creation
This PR replaces the marker interface with the method
FieldMapper#parsesArrayValue. I find this cleaner and it will help with the
fields retrieval work (#55363).
The refactor also ensures that only field mappers can declare they parse array
values. Previously other types like ObjectMapper could implement the marker
interface and be passed array values, which doesn't make sense.
Add `TRIM` function which combines the functionality of both
`LTRIM` and `RTRIM` by stripping both leading and trailing
whitespaces.
Refers to #41195
(cherry picked from commit 6c86c919e12f0c4cb5e39d129aa65ab3e274268f)
We test expected TLS failures by catching SSLException, but other
security providers ( i.e. BCFIPS ) might throw a different one. In
this case, BCFIPS throws org.bouncycastle.tls.TlsFatalAlert
* Move classes from build scripts to buildSrc
- move Run task
- move duplicate SanEvaluator
* Remove :run workaround
* Some little cleanup on build scripts on the way
As the datastream information is stored in the `ClusterState.Metadata` we exposed
the `Metadata` to the `AsyncWaitStep#evaluateCondition` method in order for
the steps to be able to identify when a managed index is part of a DataStream.
If a managed index is part of a DataStream the rollover target is the DataStream
name and the highest generation index is the write index (ie. the rolled index).
(cherry picked from commit 6b410dfb78f3676fce1b7401f1628c1ca6fbd45a)
Signed-off-by: Andrei Dan <andrei.dan@elastic.co>
Some BI tools (i.e. Tableau) would try to cast strings where the time
part is separated from the date part with a whitespace instead of `T`.
Adjust type conversion used by CAST to support this.
(cherry picked from commit 0e18321e7ad9f779c42855efbf93f171b9128a5e)
Add basic support for `TOP X` as a synonym to LIMIT X which is used
by [MS-SQL server](https://docs.microsoft.com/en-us/sql/t-sql/queries/top-transact-sql?view=sql-server-ver15),
e.g.:
```
SELECT TOP 5 a, b, c FROM test
```
TOP in SQL server also supports the `PERCENTAGE` and `WITH TIES`
keywords which this implementation doesn't.
Don't allow usage of both TOP and LIMIT in the same query.
Refers to #41195
(cherry picked from commit 2f5ab81b9ad884434d1faa60f4391f966ede73e8)
At some point, we changed the supported-type test to also catch
assertion errors. This has the side effect of also catching the
`fail()` call inside the try-catch, which silently smothered some
failures.
This modifies the test to throw at the end of the try-catch
block to prevent from accidentally catching itself.
Catching the AssertionError is convenient because there are other locations
that do throw an assertion in tests (due to hitting an assertion
before the exception is thrown) so I think we should keep it around.
Also includes a variety of fixes to other tests which were failing
but being silently smothered.
* [ML] mark forecasts for force closed/failed jobs as failed (#57143)
forecasts that are still running should be marked as failed/finished in the following scenarios:
- Job is force closed
- Job is re-assigned to another node.
Forecasts are not "resilient". Their execution does not continue after a node failure. Consequently, forecasts marked as STARTED or SCHEDULED should be flagged as failed. These forecasts can then be deleted.
Additionally, force closing a job kills the native task directly. This means that if a forecast was running, it is not allowed to complete and could still have the status of `STARTED` in the index.
relates to https://github.com/elastic/elasticsearch/issues/56419
* [ML] adds new for_export flag to GET _ml/inference API (#57351)
Adds a new boolean flag, `for_export` to the `GET _ml/inference/<model_id>` API.
This flag is useful for moving models between clusters.
* Add new circuitbreaker plugin and refactor CircuitBreakerService (#55695)
This commit lays the ground work for plugins supplying their own circuit breakers.
It adds a new interface: `CircuitBreakerPlugin`.
This interface provides methods for providing custom child CircuitBreaker objects. There are also facilities for allowing dynamic settings for the custom breakers.
With the refactor, circuit breakers are no longer replaced on setting changes. Instead, the two mutable settings themselves are `volatile`. Plugins that want to use their custom circuit breaker should keep a reference of their constructed breaker.
This adds a max_model_memory setting to forecast requests.
This setting can take a string value that is formatted according to byte sizes (i.e. "50mb", "150mb").
The default value is `20mb`.
There is a HARD limit at `500mb` which will throw an error if used.
If the limit is larger than 40% the anomaly job's configured model limit, the forecast limit is reduced to be strictly lower than that value. This reduction is logged and audited.
related native change: https://github.com/elastic/ml-cpp/pull/1238
closes: https://github.com/elastic/elasticsearch/issues/56420
Implement TIME_PARSE(<time_str>, <pattern_str>) function
which allows to parse a time string according to the specified
pattern into a time object. The patterns allowed are those of
java.time.format.DateTimeFormatter.
Closes#54963
Co-authored-by: Andrei Stefan <astefan@users.noreply.github.com>
Co-authored-by: Patrick Jiang(白泽) <patrickjiang0530@gmail.com>
(cherry picked from commit 1fe1188d449cad7d0782a202372edc52a4014135)
Backport of #56878 to 7.x branch.
With this change the following APIs will be able to resolve data streams:
get index, get mappings and ilm explain APIs.
Relates to #53100
Allows geo fields (`geo_point`, `geo_shape`) to have missing values.
Fixes a bug where such missing values would result in an error.
Closes#57299
Backport of #57300
Backporting #56888 to 7.x branch.
Limit the creation of data streams only for namespaces that have a composable template with a data stream definition.
This way we ensure that mappings/settings have been specified and will be used at data stream creation and data stream rollover.
Also remove `timestamp_field` parameter from create data stream request and
let the create data stream api resolve the timestamp field
from the data stream definition snippet inside a composable template.
Relates to #53100
Previously, `CASE` and `IIF` when translated to painless scripts
(used in GROUP BY, HAVING, WHERE) a custom `caseFunction`
registered in the `InternalSqlScriptUtils` was used. This function
received and array of arbitrary length:
```[condition1, result1, condition2, result2, ... elseResult]```
Painless doesn't know of the context and therefore is evaluating
all conditions and results before invoking the `caseFunction` on them.
As a consequence, erroneous result expressions (i.e. division by 0)
where always evaluated despite of the guarding condition.
Replace the `caseFunction` with painless `<cond> ? <res1> : <res2>`
expressions to properly guard the result expressions and only evaluate
the one for which its guarding condition evaluates to true (or of course
the elseResult).
As a bonus, this approach includes performance benefits since we avoid
unnecessary evaluations of both conditions and result expressions.
Fixes: #49672
(cherry picked from commit 9584b345d89f797bfb658212b928b9812804f02f)
The ssl.trust setting for Watcher provides a list of hostnames that
should be automatically trusted for SSL hostname verification. It was
accidentally broken when we added the full ssl.* settings for email
notifications (see #45272)
This commit corrects this, so the setting is once again respected,
as long as none of the other ssl settings are configured for email
notifications.
Resolves: #52153
Backport of: #56090
This commit removes the compiler.java setting from the build. It was
originally added when Gradle was far behind support for the latest jdk,
but is no longer applicable as we don't have any need to update the
supported compile version before gradle supports the newer version. Note
that the runtime version changing support still exists here, this only
ensures we use the same jdk to compile as we use to run gradle.
Since #51888 the ML job stats endpoint has returned entries for
jobs that have a persistent task but not job config. Such
orphaned tasks caused monitoring to fail.
This change ignores any such corrupt jobs for monitoring purposes.
Backport of #57235
This PR removes the blocking call to insert ingest documents into a queue in the
coordinator. It replaces it with an offer call which will throw a rejection exception
in the event that the queue is full. This prevents deadlocks of the write threads
when the queue fills to capacity and there are more than one enrich processors
in a pipeline.
The searcher was randomly wrapping its reader as slow, parallel, or filtered.
This was causing casting issues in the normalizer tests. By removing the
wrapping, the problem goes away.
Closes#57164
If a job is NOT opened, forecasts should be able to be deleted, no matter their state.
This also fixes a bug with expanding forecast IDs. We should check for wildcard `*` and `_all` when expanding the ids
closes https://github.com/elastic/elasticsearch/issues/56419
Change the error message wording for comparisons against fields in
filtering (s/variables/fields).
(cherry picked from commit d9a1cb50940d0a98fd75b9c0123ca6e1d862f65d)
* Update the JLine dependency to 3.14.1
Update the JLine dependency from 3.10.0 to 3.14.1.
(cherry picked from commit c2d9b74046fa5ddb54604da3afa7887cc38548a1)
Backport of #55548
Adds equivalence for keyword field to the wildcard field. Regex, fuzzy, wildcard and prefix queries are all supported.
All queries use an approximation query backed by an automaton-based verification queries.
Closes#54275
Fix delete_expired_data/nightly maintenance when
many model snapshots need deleting (#57041)
The queries performed by the expired data removers pull back entire
documents when only a few fields are required. For ModelSnapshots in
particular this is a problem as they contain quantiles which may be
100s of KB and the search size is set to 10,000.
This change makes the search more efficient by only requesting the
fields needed to work out which expired data should be deleted.
In #51089 where SamlAuthenticatorTests were refactored, we missed
to update one test case which meant that a single key would be
used both for signing and encryption in the same run. As explained
in #51089, and due to FIPS 140 requirements, BouncyCastle FIPS
provider will block RSA keys that have been used for signing from
being used for encryption and vice versa
This commit changes testNoAttributesReturnedWhenTheyCannotBeDecrypted
to always use the specific keys we have added for encryption.
This change ensures that we stop the maintenance service on all nodes
when a data node is restarted. This ensures that we don't send
update_by_query requests on the node that is restarted.
This commit also raises the log level to trace for some packages
in order to investigate the failures to acquire a shard lock
after a restart.
Relates #56765
- Use opensaml to sign and encrypt responses/assertions/attributes
instead of doing this manually
- Use opensaml to build response and assertion objects instead of
parsing xml strings
- Always use different keys for signing and encryption. Due to FIPS
140 requirements, BouncyCastle FIPS provider will block
RSA keys that have been used for signing from being used for
encryption and vice versa. This change adds new encryption specific
keys to be used throughout the tests.
Move the JDBC functionality integration tests from `:sql:qa` to a separate
module `:sql:qa:jdbc`. This way the tests are isolated from the rest of the
integration tests and they only depend to the `:sql:jdbc` module, thus
removing the danger of accidentally pulling in some dependency that may
hide bugs.
Moreover this is a preparation for #56722, so that we can run those tests
between different JDBC and ES node versions and ensure forward
compatibility.
Move the rest of existing tests inside a new `:sql:qa:server` project, so that
the `:sql:qa` becomes the parent project for both and one can run all the integration
tests by using this parent project.
(cherry picked from commit c09f4a04484b8a43934fe58fbc41bd90b7dbcc76)
Our FIPS 140 testing depends on setting the appropriate java policy
in order to configure the JVM in FIPS mode. Some tests (
discovery-ec2 and ccr qa ) also needed to set a custom policy file
to grant a specific permission, which overwrote the FIPS related
policy and tests would fail. This change ensures that when a
custom policy needs to be set in these tests, the permissions that
are necessary for FIPS are also set.
Resolves: #51685, #52034
Field mapping detection is done via grok patterns.
This commit adds well-known text (WKT) formatted geometry detection.
If everything is a `POINT`, then a `geo_point` mapping is preferred.
Otherwise, if all the fields are WKT geometries a `geo_shape` mapping is preferred.
This does **NOT** detect other types of formatted geometries (geohash, comma delimited points, etc.)
closes https://github.com/elastic/elasticsearch/issues/56967
* Fix temp dir locked errors
The tests involving a temporary directory (containing the JDBC JAR) fail
on Windows because they can't be deleted, due to still being in use.
This commit forces a premature closing of the JAR file, which mitigates
the failure by giving the JVM more time to collect any open FDs.
(Calling the System.gc() in the tests is another working alternative
fix.)
The stream-based JAR access is taken care by disabling the cache usage
(cherry picked from commit 04f97333a015404a68e8f19223f33aadeb396687)
The original implementation utilized `bbox` as the index mapping type. This would not work as it would have to be `envelope`. But, given that `envelope` and `polygon` are tessellated in the same way, we choose to use `polygon` as the geo_shape type. This is for easier support other places in the stack (a la kibana maps)
Merging logic is currently split between FieldMapper, with its merge() method, and
MappedFieldType, which checks for merging compatibility. The compatibility checks
are called from a third class, MappingMergeValidator. This makes it difficult to reason
about what is or is not compatible in updates, and even what is in fact updateable - we
have a number of tests that check compatibility on changes in mapping configuration
that are not in fact possible.
This commit refactors the compatibility logic so that it all sits on FieldMapper, and
makes it called at merge time. It adds a new FieldMapperTestCase base class that
FieldMapper tests can extend, and moves the compatibility testing machinery from
FieldTypeTestCase to here.
Relates to #56814
Previously `COUNT(DISTINCT <literal>)` was returning the same result
as `COUNT(<literal>)` which is not correct as it should always return 1
if there is at least one matching row (bucket if there is a GROUP BY),
or 0 otherwise.
(cherry picked from commit 7f7d7562d43034907f432d39d0d66f490d78f4a8)
Throttling nightly cleanup as much as we do has been over cautious.
Night cleanup should be more lenient in its throttling. We still
keep the same batch size, but now the requests per second scale
with the number of data nodes. If we have more than 5 data nodes,
we don't throttle at all.
Additionally, the API now has `requests_per_second` and `timeout` set.
So users calling the API directly can set the throttling.
This commit also adds a new setting `xpack.ml.nightly_maintenance_requests_per_second`.
This will allow users to adjust throttling of the nightly maintenance.
The version number componenent can't equal or exceed the revision
multiplier.
This fixes a the VersionTests unit test.
(cherry picked from commit 7d2331a2818ae20024c5c3617cd4433f90e9c098)
* Adds support for MIN, MAX, AVG, SUM aggregates acting on literals.
SELECT SUM(1) FROM index
and
SELECT SUM(1), AVG(2)
work both on indices and as local execution.
(cherry picked from commit efb72907c0391612c4a2b6256e327060b4167912)
WatcherIndexTemplateRegistry as of https://github.com/elastic/elasticsearch/pull/52962
requires all nodes to be on 7.7.0 before it allows the version 11 index template to be
installed.
While in a mixed cluster, nothing prevents Watcher from running on the new
host before the all of the nodes are on 7.7.0. This will result in the
.watcher-history-11* index without the proper mappings. Without the proper
mapping a single document (for a large watch) can exceed the default 1000 field
limit and cause error to show in the logs.
This commit ensures the same logic for writing to the index is applied as for
installing the template. In a mixed cluster, the `10` index template will continue
to be written. Only once all of nodes are on 7.7.0+ will the `11` index template
be installed and used.
closes#56732
In DF analytics classification, it is possible to use no samples
of a class if its cardinality is too low.
This commit fixes this by ensuring the target sample count can never be zero.
Backport of #56783
* JDBC: fix access to the Manifest for non-entry JAR
The JDBC driver will attempt to read its version from the Manifest file
embedded into its JAR. The URL pointing to the JAR can be provided in a
few ways.
So far, accessing the Manfiest was attempted by getting a URLConnection
out of the URL and then getting an input stream out of this connection.
For file JAR URLs, this only works however if the URL points to the
driver as a JAR file entry (i.e. <sub-url>!/jdbc-driver.jar!/). If
that's not the case, the JarURLConnection will throw an IOException.
This commit fixes that: in case the URL points to a JAR entry
(jar:file:<path>/jdbc-driver.jar!/), the manifest is read directly with
JarURLConnection#getManifest().
(cherry picked from commit 2175b7b01cf5fcf3ab2bb21404a9bd454a8df3f0)
Co-authored-by: Elastic Machine <elasticmachine@users.noreply.github.com>
In some cases the Enrich processor factory may be called before it is
ready to create processors. While these calls are usually made in error,
the response from the Enrich processor is an NPE which is almost always
an unhelpful error when debugging an issue.
Yannick Welsch
Costin Leau
Andrei Dan
Nik Everett
Dan Hermann
Andrei Stefan
Benjamin Trent
Przemysław Witek
Mayya Sharipova
David Kyle
Luca Cavanna
David Roberts
Jake Landis
Dimitris Athanasiou
Tanguy Leroux
Hendrik Muhs
William Brafford
Ioannis Kakavas
William Brafford
Mark Vieira
Rene Groeschke
Marios Trivyzas
Igor Motov
Julie Tibshirani
Mark Tozzi
Zachary Tong
Martijn van Groningen
Tim Vernum
Ryan Ernst
Ryan Ernst
James Baiera
Lee Hinman
Tal Levy
Bogdan Pintea
markharwood
Jim Ferenczi
Armin Braun
Alan Woodward