Re: need few clarification for some of our use cases

Hi Navneet, Ignite supports various logging utilities (jul, log4j, slf4j etc.)[1]. I used to log4j SocketAppender for issues like this. Look at the Ignite transaction API[2], you got specific exception when transaction was not commited. Yes, you can use Cache Interceptor[3] for processing cache

need few clarification for some of our use cases

Hi All, I need few clarification for some of our use cases: - Can we redirect the apache ignite logs to remote server ? - How LRTs(pending transactions not committed) can be identified and acted upon ? - Does Apache ignite support DB triggers when Cache is updated. -- View

Re: Ignite on google dataproc

Hi, Please properly subscribe to the mailing list so that the community can receive email notifications for your messages. To subscribe, send empty email to user-subscr...@ignite.apache.org and follow simple instructions in the reply. openSourceAlways wrote > Can ignite be used on google

Re: Cluster hung after a node killed

logs.zip Attaching threaddump and log from 2 nodes. Lost logs from node 3 which was killed using "kill-9 " Let me know if you need more logs. In my understanding, after killing node 3 topology version update screwed-up and

Re: Can java code look up cache in server without Ignition.start(cfg)

Saji, No, you need to start the client to access the cluster. Note that you can start it once for the whole application and reuse everywhere. It's thread safe and designed to be used by multiple threads concurrently. -Val -- View this message in context:

Can java code look up cache in server without Ignition.start(cfg)

Explaining my scenario I have my Ignite Server started using ignite.bat with grid name FirstGrid I have a java class CacheOperations with putCache() and getCache() as methods and acts as client to get and put data to the cache in the Ignite server. In CacheOperations I am using Ignite

Re: Limit the number of results query

Do you mean OOM on the server or on the client? On server you indeed will need some spare space for any ongoing operations (not only queries, actually). You should make sure that this space is enough for you load. On client you should not get OOM because results are sent in pages (unless you call

Re: Checkingpointing

Can you query the existing data to understand what is loaded and what is not? -Val -- View this message in context: http://apache-ignite-users.70518.x6.nabble.com/Checkingpointing-tp8933p9005.html Sent from the Apache Ignite Users mailing list archive at Nabble.com.

Re: Remote Server Thread Not exit when Job finished,Cause out of memory

All executors should be stopped on Ignite.close, I hardly believe there are threads still running after that. However, you should reuse the already started client in any case. All jobs started by the stopping client will be cancelled when you call close(). -Val -- View this message in

Re: Problem with v for listening updates

I've created simple project with three unit tests here - https://github.com/andry-k/test-ignite-CQ. First with peer class loading enabled. Second with specified indexed types. And third which is working as expected (peer class loading is disabled and no indexed types are specified) -- View

Limit the number of results query

Hi, Is there any way to limit the number of results of ignite query so that results wont create out of memory even query misses to limit the results? When number of parallel queries are running on a node.. this is very crucial.. each query would have some results. i see query timeout also not

Re: Remote Server Thread Not exit when Job finished,Cause out of memory

Thanks Denis, your suggestion really broaden my ignite knowledge I have tried both of ComputeFibonacciContinuationExample and ComputeJobMasterLeaveAware. The executed jobs created by client node still not been destroyed. At the very beging, I excepted that when I call ignite.close(), threads

Re: IN Query

Hi, You can check it in 1.6.0 release or checkout master and check on it. On Tue, Nov 15, 2016 at 5:05 PM, Anil wrote: > Thank you. You saved my time. > > May i know working ignite version ? i see it is issue in h2 db itself. > > Thanks. > > On 15 November 2016 at 19:30,

Re: IN Query

Thank you. You saved my time. May i know working ignite version ? i see it is issue in h2 db itself. Thanks. On 15 November 2016 at 19:30, Vladislav Pyatkov wrote: > Hi Anil, > > You are right. I have checked this on not released version, but in 7.0.0 > indexes are not

Re: IN Query

Hi Anil, You are right. I have checked this on not released version, but in 7.0.0 indexes are not used by some strange reason. You can check the case in master or previous version, it worked earlier and will work after (but 7.0.0 have bug). On Tue, Nov 15, 2016 at 2:36 PM, Anil

Re: How can I detect DB writing abnormal in case of write-behind?

I had tried to enable cache statistics, but it did not work. my cache configure as following:

Re: Sql Query on binary objects

Hi, I mean, what are differences you wait? This SQL will be equivalent for binary, enough to add IgniteCache cache_customer = ignite.createCache(ccfg_customer).cache.withKeepBinary(); On Tue, Nov 15, 2016 at 2:44 PM, rishi007bansod

Re: IN Query

HI, i am still seeing no index used. Can you verify the below query please? explain select * from ( ( select * from Person p join table(joinId varchar(10) = ('anilkd1','anilkd2')) i on p.id = i.joinId) UNION (select * from Person p join table(name varchar(10) = ('Anil1', 'Anil5')) i on p.name =

Re: Sql Query on binary objects

Hi, If you use BinaryMarshaller server works with object in Binary format. What are you meant: "convert existing cache queries"? Could you please, explain what do you want to achieve? On Tue, Nov 15, 2016 at 2:06 PM, rishi007bansod wrote: > Is there any way we can

Re: Cluster hung after a node killed

Hi, I think, INFO log level would be enough (from all nodes). -- View this message in context: http://apache-ignite-users.70518.x6.nabble.com/Cluster-hung-after-a-node-killed-tp8965p8991.html Sent from the Apache Ignite Users mailing list archive at Nabble.com.

?????? C++ API can't build successfully in my linux environment

Hi, Igor Sapego: using the latest ignite that you gave yesterday, it build report error as follow: (1) make[3]: Entering directory `/data/smilehong/apache-ignite-1.7.0-src/modules/platforms/cpp/jni' CXX os/linux/src/utils.lo CXX src/java.lo src/java.cpp: In

Re: How can I detect DB writing abnormal in case of write-behind?

Hi, You should enable cache statistics: ... for get correct value from metrics. -- View this message in context: http://apache-ignite-users.70518.x6.nabble.com/How-can-I-detect-DB-writing-abnormal-in-case-of-write-behind-tp8954p8989.html Sent from the Apache Ignite Users

Re: IN Query

Hi, This is strange. i tried to couple of times and did not see any index in explain plan. Let me give a try again. Thanks for you response. On 15 November 2016 at 15:04, vdpyatkov wrote: > Hi Anil, > > I tred to reproduce your case, but got expected behavior (indexes

Re: IN Query

Hi Anil, I tred to reproduce your case, but got expected behavior (indexes are used). I slightly modified your SQL for the my cache configuration: select * from ( (select p.id, p.name from "simple_cache".Person p join table(joinId INT = (1, 3)) i on p.id = i.joinId) UNION (select p.id, p.name

Re: CacheInterceptor with client nodes

Thanks for the advice, Vladislav. I will try this aproach. 2016-11-15 11:43 GMT+03:00 Vladislav Pyatkov : > Hi Igor, > > Why don't you implement CacheStore for this behavior? > You should implement own CacheStore and set "Read-Through" as true[1]. > At this case all invokes

Re: CacheInterceptor with client nodes

Hi Igor, Why don't you implement CacheStore for this behavior? You should implement own CacheStore and set "Read-Through" as true[1]. At this case all invokes "get" over cache will be reflect on primary node (into CacheStore), each time when data can not be found on cache. If you want to

Re: Apache Spark & Ignite Integration

Thanks for sharing the jira ticket. Do you have inputs on the additional questions I asked about shared RDD implementation? They aren't related to the dataframe/dataset support. Looking forward for your thoughts. -- View this message in context:

Re: CacheInterceptor with client nodes

Hi, I need to call a native function via JNI on every "get" and "put" call to a specific cache. The call should be made on a server node, where this concrete enrty resides. 2016-11-15 2:34 GMT+03:00 vkulichenko : > Hi, > > What exactly are you trying to do? What's