Re: Thin Client Protocol documentation

Prachi, see "Response (failure)" table. Failed response includes server protocol version and an error message. On Thu, Nov 30, 2017 at 4:55 AM, Prachi Garg wrote: > Pavel, > > If the connection handshake fails, what should be the message length in > the response. When I try

Re: Transport compression (not store compression)

Hello, everybody. I propose the following design for network compression. I suggest to implement it like SSL implementation that already works. I add compression filter to a chain of NIO filters. It changes the logic of TcpCommunicationSpi.safeTcpHandshake() and

Re: TC issues. IGNITE-3084. Spark Data Frame API

Valentin, > Oh, so this is because of upgrade to 2.2.0? Yes, we should upgrade spark module to jdk1.8 because switching to spark 2.2.0. > least we should consider not dropping previous version yet. Please, note, we can have IgniteRDD and Ignite Data Frame for a spark 2.1. The only thing we

[GitHub] ignite pull request #3114: IGNITE-6335 .NET: Thin client: cache binary mode

Github user asfgit closed the pull request at: https://github.com/apache/ignite/pull/3114 ---

[jira] [Created] (IGNITE-7076) NPE while stopping with GridDhtLockFuture

Alexander Belyak created IGNITE-7076: Summary: NPE while stopping with GridDhtLockFuture Key: IGNITE-7076 URL: https://issues.apache.org/jira/browse/IGNITE-7076 Project: Ignite Issue

[jira] [Created] (IGNITE-7075) Visor CMD: Add possibility to modify cache data instance by command

Vasiliy Sisko created IGNITE-7075: - Summary: Visor CMD: Add possibility to modify cache data instance by command Key: IGNITE-7075 URL: https://issues.apache.org/jira/browse/IGNITE-7075 Project:

Re: Enable/disable cache statistics in runtime

Alex, We have such issue in JIRA: https://issues.apache.org/jira/browse/IGNITE-369 I think you can update its description after we agreed on implementation details. -- Alexey Kuznetsov

Re: Thin Client Protocol documentation

Pavel, If the connection handshake fails, what should be the message length in the response. When I try to fail the handshake, I get 32 as the message length. Is this expected? -Prachi On Wed, Nov 29, 2017 at 8:20 AM, Pavel Tupitsyn wrote: > Hi Prachi, > > Flags

Subject: new committer: Yury Babak

The Project Management Committee (PMC) for Apache Ignite has invited Yury Babak to become a committer and we are pleased to announce that he has accepted. Yuri is among those who has been contributing to the newborn Machine Learning component in Ignite:

Re: Upgrade to Spark 2.2.0

Opps, It’s time for Ignite to say goodbye to Java 7 too. Hope this happens in the beginning of the next year. — Denis > On Nov 29, 2017, at 3:08 PM, Valentin Kulichenko > wrote: > > Denis, > > I just found out that with version 2.2.0 Spark dropped Java 7

Re: Documentation for GA Grid: GA library to Apache Ignite

Hi Turik, > If so, I thought a section called 'Genetic Algorithm (ie: GA Grid)' could > go under the Machine Learning topic. > https://apacheignite.readme.io/docs/machine-learning > This is a right approach. I’ve granted you the access to

Re: Enable/disable cache statistics in runtime

We cannot use the public API inside of ignite core, but we can definitely broadcast messages whenever appropriate. Discovery SPI should only be used whenever transactionality and ordering need to be enforced, which is not the case here. On Wed, Nov 29, 2017 at 2:21 PM, Alex Plehanov

Re: Optimization of SQL queries from Spark Data Frame to Ignite

Nikolay, Can you please create a separate ticket for the strategy implementation then? Any idea on how long will it take? As for querying a partition, both SqlQuery and SqlFieldQuery allow to specify set of partitions to work with (see setPartitions method). I think that should be enough. -Val

Re: TC issues. IGNITE-3084. Spark Data Frame API

Nikolay, Oh, so this is because of upgrade to 2.2.0? Then I'm not sure we should upgrade in the first place, or at least we should consider not dropping previous version yet. I sent a message to the original thread about upgrade, let's decide there and then come back to this issue. -Val On Tue,

Re: Upgrade to Spark 2.2.0

Denis, I just found out that with version 2.2.0 Spark dropped Java 7 support: https://spark.apache.org/docs/2.2.0/ I'm not sure it's a good idea to do an upgrade as long as we support Java 7. Probably we should find a way to support both 2.1.0 and 2.2.0 in parallel. What do you think? -Val On

Re: Enable/disable cache statistics in runtime

Hello, Dmitriy! Is it ok to use broadcast inside ignite-core? I didn't find any usage of broadcast inside ignite, except examples, benchmarks and tests. For now statisticsEnabled is only property, which enable/disable cache JMX metrics. 2017-11-30 0:58 GMT+03:00 Dmitriy Setrakyan

Documentation for GA Grid: GA library to Apache Ignite

Denis, Where do you want me to implement documenation for GA Grid? (https://issues.apache.org/jira/browse/IGNITE-6899) I have implemented examples, but believe more information is necessary.. IE: I see that when Persistent Store was implemented docs were placed here:

Re: Enable/disable cache statistics in runtime

On Wed, Nov 29, 2017 at 12:22 PM, Alex Plehanov wrote: > Hello, Igniters! > > We need the ability to enable or disable statistics for caches in runtime. > For now, there is a cache configuration property which seems modifiable in > runtime and this, at first sight, works

[jira] [Created] (IGNITE-7074) NPE when WAL path and WAL archive path are the same

Dan Ellis created IGNITE-7074: - Summary: NPE when WAL path and WAL archive path are the same Key: IGNITE-7074 URL: https://issues.apache.org/jira/browse/IGNITE-7074 Project: Ignite Issue Type:

Enable/disable cache statistics in runtime

Hello, Igniters! We need the ability to enable or disable statistics for caches in runtime. For now, there is a cache configuration property which seems modifiable in runtime and this, at first sight, works for local node. But there are no methods to enable or disable statistics for cache cluster

[GitHub] ignite pull request #3114: IGNITE-6335 .NET: Thin client: cache binary mode

GitHub user ptupitsyn opened a pull request: https://github.com/apache/ignite/pull/3114 IGNITE-6335 .NET: Thin client: cache binary mode You can merge this pull request into a Git repository by running: $ git pull https://github.com/gridgain/apache-ignite ignite-6335

[jira] [Created] (IGNITE-7073) Ignite Basic Tests: optimize tests execution time: GridUnsafeMapTest probably not required

Dmitriy Pavlov created IGNITE-7073: -- Summary: Ignite Basic Tests: optimize tests execution time: GridUnsafeMapTest probably not required Key: IGNITE-7073 URL: https://issues.apache.org/jira/browse/IGNITE-7073

Re: Contribute ApacheIgnite

Hello Yury, Thanks for your advice. I started to get more familiar with ML library, and if I will be stuck in the development process, I would be grateful if you could give me some ideas, advice. Thanks, Eugeniu. On Wed, Nov 29, 2017 at 12:01 PM, Yury Babak wrote: > Hello

[GitHub] ignite pull request #3113: Ignite 2.3.1.b1

GitHub user alamar opened a pull request: https://github.com/apache/ignite/pull/3113 Ignite 2.3.1.b1 You can merge this pull request into a Git repository by running: $ git pull https://github.com/gridgain/apache-ignite ignite-2.3.1.b1 Alternatively you can review and apply

Re: Thin Client Protocol documentation

Hi Prachi, Flags parameter in all cache operations is reserved for things like withSkipStore, withExpiryPolicy, withKeepBinary, withNoRetries, withPartitionRecover. See methods in IgniteCache interface. Thanks, Pavel On Wed, Nov 29, 2017 at 7:10 PM, Prachi Garg wrote: > Hi

Re: Thin Client Protocol documentation

Hi Pavel, In the operation request, what does the 'flags' parameter mean? Thanks, -Prachi On Wed, Nov 29, 2017 at 5:27 AM, Pavel Tupitsyn wrote: > Sergey, good point, done. > > On Wed, Nov 29, 2017 at 2:30 PM, Sergey Kozlov > wrote: > > > Pavel >

Re: Ignite node crashes after one query fetches many entries from cache

Ray, I think to avoid excessive GC and OOM you could try switching to a lazy result set: https://apacheignite-sql.readme.io/docs/performance-and-debugging#result-set-lazy-load - Nick On Wed, Nov 29, 2017, 7:19 AM Anton Vinogradov wrote: > Ray, > > Seems you're

[GitHub] ignite pull request #3112: Ignite 2.4.1 NPE on checkpoint fix

GitHub user sergey-chugunov-1985 opened a pull request: https://github.com/apache/ignite/pull/3112 Ignite 2.4.1 NPE on checkpoint fix You can merge this pull request into a Git repository by running: $ git pull https://github.com/gridgain/apache-ignite

Re: Ignite node crashes after one query fetches many entries from cache

Ray, Seems you're looking for org.apache.ignite.cache.query.SqlFieldsQuery#timeout? On Tue, Nov 28, 2017 at 5:30 PM, Alexey Kukushkin wrote: > Ignite Developers, > > I know community is developing an "Internal Problems Detection" feature >

[GitHub] ignite pull request #3111: Ignite 2.4.1 mm blt aff fix

GitHub user sergey-chugunov-1985 opened a pull request: https://github.com/apache/ignite/pull/3111 Ignite 2.4.1 mm blt aff fix You can merge this pull request into a Git repository by running: $ git pull https://github.com/gridgain/apache-ignite ignite-2.4.1-mm-blt-affFix

[jira] [Created] (IGNITE-7072) IgniteCache.replace(k, v, nv) does not work in binary mode without classes

Pavel Tupitsyn created IGNITE-7072: -- Summary: IgniteCache.replace(k, v, nv) does not work in binary mode without classes Key: IGNITE-7072 URL: https://issues.apache.org/jira/browse/IGNITE-7072

[GitHub] ignite pull request #3110: Ignite 2.4.1 mm aff node

GitHub user sergey-chugunov-1985 opened a pull request: https://github.com/apache/ignite/pull/3110 Ignite 2.4.1 mm aff node You can merge this pull request into a Git repository by running: $ git pull https://github.com/gridgain/apache-ignite ignite-2.4.1-mm-affNode

Re: Thin Client Protocol documentation

Sergey, good point, done. On Wed, Nov 29, 2017 at 2:30 PM, Sergey Kozlov wrote: > Pavel > > Could you update the page by following: > > - String, date, UUID arrays allow to put NULL. Due to that every item in > the array written as type code byte (default item type or

[jira] [Created] (IGNITE-7071) Cache stop requests batching

Dmitriy Pavlov created IGNITE-7071: -- Summary: Cache stop requests batching Key: IGNITE-7071 URL: https://issues.apache.org/jira/browse/IGNITE-7071 Project: Ignite Issue Type: Bug

[jira] [Created] (IGNITE-7069) Ignite PDS compatibilty framework: implement configurable classpath, JVM options, node startup timeout

Dmitriy Pavlov created IGNITE-7069: -- Summary: Ignite PDS compatibilty framework: implement configurable classpath, JVM options, node startup timeout Key: IGNITE-7069 URL:

[jira] [Created] (IGNITE-7070) Ignite PDS compatibilty framework: implement configurable classpath, JVM options, node startup timeout

Dmitriy Pavlov created IGNITE-7070: -- Summary: Ignite PDS compatibilty framework: implement configurable classpath, JVM options, node startup timeout Key: IGNITE-7070 URL:

[GitHub] ignite pull request #3109: IGNITE-7047 NPE at org.jsr166.ConcurrentLinkedHas...

GitHub user apopovgg opened a pull request: https://github.com/apache/ignite/pull/3109 IGNITE-7047 NPE at org.jsr166.ConcurrentLinkedHashMap.replace You can merge this pull request into a Git repository by running: $ git pull https://github.com/gridgain/apache-ignite

[GitHub] ignite pull request #3108: Ignite-1.7.17

GitHub user sk0x50 opened a pull request: https://github.com/apache/ignite/pull/3108 Ignite-1.7.17 You can merge this pull request into a Git repository by running: $ git pull https://github.com/sk0x50/ignite ignite-1.7.17 Alternatively you can review and apply these changes

[GitHub] ignite pull request #3107: Ignite 2.1.8.b1

GitHub user ezhuravl opened a pull request: https://github.com/apache/ignite/pull/3107 Ignite 2.1.8.b1 You can merge this pull request into a Git repository by running: $ git pull https://github.com/gridgain/apache-ignite ignite-2.1.8.b1 Alternatively you can review and

Re: Optimization of SQL queries from Spark Data Frame to Ignite

Hi Nikolay, No, it is not possible to get this info from public API, neither we planned to expose it. See IGNITE-4509 and commit *fbf0e353* to get better understanding on how this was implemented. Vladimir. On Wed, Nov 29, 2017 at 2:01 PM, Николай Ижиков wrote: >

[jira] [Created] (IGNITE-7068) Incorrect error reporting for incompatible type in SQL WHERE clause

Kirill Shirokov created IGNITE-7068: --- Summary: Incorrect error reporting for incompatible type in SQL WHERE clause Key: IGNITE-7068 URL: https://issues.apache.org/jira/browse/IGNITE-7068 Project:

Re: Thin Client Protocol documentation

Pavel Could you update the page by following: - String, date, UUID arrays allow to put NULL. Due to that every item in the array written as type code byte (default item type or null type code) + type data. It should be detailed explained (looks like that the table should have an addtional

Re: Optimization of SQL queries from Spark Data Frame to Ignite

Valentin, > process the AST generated by Spark and convert it to Ignite SQL... > Does it make sense to you? Yes. I think it is a great approach. Let's implement such feature as the second step of Data Frame integration. 2017-11-29 3:23 GMT+03:00 Valentin Kulichenko

Re: Ignite Enhancement Proposal #7 (Internal problems detection)

Vladimir, At the moment policy looks like so: /** * Policy that defines how node will process the failures. Note that default * failure processing policy is defined by {@link IgniteConfiguration#DFLT_FLR_PLC} property. */ public enum FailureProcessingPolicy { /** Restart jvm. */

Re: Optimization of SQL queries from Spark Data Frame to Ignite

Hello, Vladimir. > partition pruning is already implemented in Ignite, so there is no need to do this on your own. Spark work with partitioned data set. It is required to provide data partition information to Spark from custom Data Source(Ignite). Can I get information about pruned partitions

Re: Ignite Enhancement Proposal #7 (Internal problems detection)

Dmitry, Thank you, but how FailureProcessingPolicy looks like? It is not clear how can I configure different reactions to different event types. On Wed, Nov 29, 2017 at 1:47 PM, Дмитрий Сорокин wrote: > Vladimir, > > These policies (policy, in fact) can be configured

[GitHub] ignite pull request #3097: IGNITE-6992 FIX Ignite MR problem with accessing ...

Github user asfgit closed the pull request at: https://github.com/apache/ignite/pull/3097 ---

Re: Ignite Enhancement Proposal #7 (Internal problems detection)

Vladimir, These policies (policy, in fact) can be configured in IgniteConfiguration by calling setFailureProcessingPolicy(FailureProcessingPolicy flrPlc) method. 2017-11-29 10:35 GMT+03:00 Vladimir Ozerov : > Denis, > > Yes, but can we look at proposed API before we dig

[jira] [Created] (IGNITE-7067) web agent throws SocketTimeoutException instead of userfriendly error message

Pavel Konstantinov created IGNITE-7067: -- Summary: web agent throws SocketTimeoutException instead of userfriendly error message Key: IGNITE-7067 URL: https://issues.apache.org/jira/browse/IGNITE-7067

Re: Contribute ApacheIgnite

Hello Eugeniu, First of all thanks for your interest, we appreciate it. Please feel free to ask any questions about ticket, module and whole project. >From my side I want to provide advice to take a look on Trainer and Model API. Basically org.apache.ignite.ml.Model is a result of any ML

[jira] [Created] (IGNITE-7066) Web console: add ability to select web agent for importing model from DB

Pavel Konstantinov created IGNITE-7066: -- Summary: Web console: add ability to select web agent for importing model from DB Key: IGNITE-7066 URL: https://issues.apache.org/jira/browse/IGNITE-7066

IGNITE-4908 is ready for review (Ignite.reentrantLock looks much slower than IgniteCache.lock)

Igniters, I’ve implemented new fast reentrant lock within an issue[1]. Could someone review prepared PR [2, 3, 4]? Some details are described below. The main idea: Current locks implementation is based on shared states in a cache, while the new lock uses IgniteCache#invoke* methods to update

[jira] [Created] (IGNITE-7065) Cannot use multiple Ignite Kafka Sink Connectors

Alexey Kukushkin created IGNITE-7065: Summary: Cannot use multiple Ignite Kafka Sink Connectors Key: IGNITE-7065 URL: https://issues.apache.org/jira/browse/IGNITE-7065 Project: Ignite

Re: Rework storage format to index-organized approach

On Wed, Nov 29, 2017 at 12:33 AM, Vladimir Ozerov wrote: > Alexey, > > This is completely unrelated activity. > But still is a great idea :)

Re: Rework storage format to index-organized approach

Alexey, This is completely unrelated activity. On Wed, Nov 29, 2017 at 11:16 AM, Dmitriy Setrakyan wrote: > On Wed, Nov 29, 2017 at 12:14 AM, Alexey Kuznetsov > wrote: > > > > > If we are going to rework indexes, could we also think about

Re: Rework storage format to index-organized approach

On Wed, Nov 29, 2017 at 12:14 AM, Alexey Kuznetsov wrote: > > If we are going to rework indexes, could we also think about supporting TTL > Indexes (like in Mongo DB [1])? > Awesome idea. Is there a ticket?

Re: Rework storage format to index-organized approach

Vova, If we are going to rework indexes, could we also think about supporting TTL Indexes (like in Mongo DB [1])? [1] https://docs.mongodb.com/manual/core/index-ttl/ On Tue, Nov 28, 2017 at 3:46 AM, Vladimir Ozerov wrote: > Igniters, > > I'd like to start a discussion