Re: Finding Key to Node Mapping

[Alper Tekinalp]

Hi Val.

The REST API won't work because in one type of our client we
constantly use that mapping. So what about configuring client so that
it will handle key to node mapping but the other things can be turned
off or limited. Which configurations should I switch off or decrease
number of etc?

Regards.

On Sat, Sep 24, 2016 at 1:18 AM, vkulichenko
 wrote:
> Hi Alper,
>
> No, this is not possible to know the mapping without connecting to the
> cluster, because affinity depends on topology. Good approach for such cases
> is using the REST API [1], but current implementation is limited though. So
> you will probably have to create your own endpoint and run it as a service
> on Ignite nodes. Will this work for you?
>
> [1] https://apacheignite.readme.io/docs/rest-api
>
> -Val
>
>
>
> --
> View this message in context: 
> http://apache-ignite-users.70518.x6.nabble.com/Finding-Key-to-Node-Mapping-tp7653p7926.html
> Sent from the Apache Ignite Users mailing list archive at Nabble.com.



-- 
Alper Tekinalp

Software Developer

Evam Stream Analytics

Atatürk Mah.Turgut Özal Bulvarı Gardenya 1 Plaza 42/B K:4 Ataşehir / İSTANBUL

Tlf : +90216 688 45 46 Fax : +90216 688 45 47 Gsm:+90 536 222 76 01

www.evam.com

Question about cron-based scheduler

[Level D]

Hi all,


I find the minimal scheduling time unit this scheduler supported  is 1 minute.


But I need a scheduling time unit less than 1 minute . 
Is there a way to make it happen?

Regards,


Zhou.

Out of memory

[javastuff....@gmail.com]

In my application I have 100s of different caches defined. Some are LOCAL,
NEAR and OFFHEAP, in total all are 10GB and LOCAL are about 1GB. 
Running application with 4GB heap, machine have 48 CPU and 40GB RAM. I am
starting ignite from java code. Under bigger load getting OOM. Attaching
screenshot from heap dump. 

Am I missing anything or need to disable any kind of metrics? I had enabled
task and cache events, which I am planing to disable. 

Any hints will be good.  

Thanks,
-Sam
  
 



--
View this message in context: 
http://apache-ignite-users.70518.x6.nabble.com/Out-of-memory-tp7995.html
Sent from the Apache Ignite Users mailing list archive at Nabble.com.

Re: Data Streamer error

[matt]

OK I'll give that a shot. Thanks Val!



--
View this message in context: 
http://apache-ignite-users.70518.x6.nabble.com/Data-Streamer-error-tp7725p7994.html
Sent from the Apache Ignite Users mailing list archive at Nabble.com.

Re: Is IgniteQueue transactional?

[vkulichenko]

Queues are not transactional, so this is expected behavior. What's your use
case?

-Val



--
View this message in context: 
http://apache-ignite-users.70518.x6.nabble.com/Is-IgniteQueue-transactional-tp7987p7993.html
Sent from the Apache Ignite Users mailing list archive at Nabble.com.

Re: How to avoid the event lost in the continuous query

[vkulichenko]

To run the example you only need to execute the main() method :) Remote
filter will be automatically deployed on all server nodes. The node embedded
in the example will have the local listener that will get notifications for
all updates that pass through the filter.

-Val



--
View this message in context: 
http://apache-ignite-users.70518.x6.nabble.com/How-to-avoid-the-event-lost-in-the-continuous-query-tp7904p7992.html
Sent from the Apache Ignite Users mailing list archive at Nabble.com.

Re: Ignite zeppelin jdbc connection

[vkulichenko]

Read-through means that when you do a get() on the cache, it will return the
value from the database if it doesn't exist in cache. But it sounds you're
invoking load() directly on the store. Is this the case? Can you show your
code?

-Val



--
View this message in context: 
http://apache-ignite-users.70518.x6.nabble.com/Ignite-zeppelin-jdbc-connection-tp7898p7991.html
Sent from the Apache Ignite Users mailing list archive at Nabble.com.

Ignite Rest Api

[Mohammad Shariq]

I am using example code from the ignite distribution to start an ignite
node...ExampleNodeStartup.java

Also I have included rest api lib in pom.xml.
But I am not seeing any response when I hit the url
localhost:8080/ignite?cmd=cache..

Although when I start my node using ignite.bat I do see the response...

Re: Result set processing in off heap mode

[Denis Magda]

Hi,

Right now result set processing can be done in Java heap only. There is a 
ticket to improve a situation around this by moving result set processing to 
off-heap
https://issues.apache.org/jira/browse/IGNITE-3870 


Until this functionality is not available I can suggest you the following 
options:
- tune Java garbage collector [1], queries rate and result set size.
- use a commercial JVM like Azul Zing that doesn’t have stop-the-world pauses 
at all. 

[1] https://apacheignite.readme.io/docs/jvm-and-system-tuning 

—
Denis

> On Sep 28, 2016, at 6:01 AM, Sri Ganesh V  wrote:
> 
> Hi,
> 
> We use cache in *off-heap *mode.
> 
> Recently we faced GC issue when there were multiple client connections
> executing SQL queries
> 
> [via JDBC connection - DriverManager
>   
> .getConnection("jdbc:ignite:cfg://cache=TEST_CACHE:distributedJoins=true@file:///C:\\ignite-client-config.xml")
> ]
> 
> The returned query result set is often huge and GC kicks in very often and
> there is a huge impact on the query response time. 
> 
> Also we see "Query execution is too long" warning messages when GC kicks in.
> There were scenarios when ignite server node went into non-responsive state
> and never recovered back.
> 
> We would like to know if "Result Set" processing can happen out of GC
> collectable heap like the way we can use cache in off-heap mode ??
> 
> Thanks.
> 
> 
> 
> --
> View this message in context: 
> http://apache-ignite-users.70518.x6.nabble.com/Result-set-processing-in-off-heap-mode-tp7988.html
> Sent from the Apache Ignite Users mailing list archive at Nabble.com.

Re: Getting a cluster-configured JCache CacheManager instance

[Josh Cummings]

Okay, thanks, I'll try that. Along similar lines, is there a way for me to
bring up an Ignite client, say programmatically, give it a name, and then
send just the grid name through the JCache API?

Not trying to make Ignite behave just like Hazelcast, but just to make my
thoughts clear with an example, there I can do this:

Properties properties = new Properties();
properties.setProperty("hazelcast.instance.name", "my-instance-name");
Caching.getCachingProvider().getCacheManager(someUri, someClassLoader,
properties);

And it will pick up a client I've already configured elsewhere in the
runtime by the name of "my-instance-name".

Is there a way to identify an already configured client through the JCache
API?

Thanks, again.



On Tue, Sep 27, 2016 at 4:40 PM, vkulichenko 
wrote:

> Hi Josh,
>
> Ignite uses Spring for configuration, so you can utilize Spring property
> placeholder.You can find a nice example here:
> https://www.mkyong.com/spring/spring-propertyplaceholderconfigurer-
> example/
>
> You can also use one of the shared IP finders provided by Ignite. Shared
> here means that each node that joins the topology will leave its
> coordinates
> in some shared storage, so that other nodes always know where to connect.
> With this approach you will not have to explicitly specify addresses at
> all.
> Here is the list of all available IP finders (only static IP based is not
> shared): https://apacheignite.readme.io/docs/cluster-config
>
> -Val
>
>
>
> --
> View this message in context: http://apache-ignite-users.
> 70518.x6.nabble.com/Getting-a-cluster-configured-JCache-
> CacheManager-instance-tp7970p7978.html
> Sent from the Apache Ignite Users mailing list archive at Nabble.com.
>



-- 

*JOSH CUMMINGS*

Principal Engineer

[image: Workfront] 

*O*  801.477.1234  |  *M*  8015562751

joshcummi...@workfront.com | www.workfront.com
Address   |  Twitter
  |  LinkedIn
  |  Facebook


[image: Workfront] 

Result set processing in off heap mode

[Sri Ganesh V]

Hi,

We use cache in *off-heap *mode.

Recently we faced GC issue when there were multiple client connections
executing SQL queries
 
[via JDBC connection - DriverManager

.getConnection("jdbc:ignite:cfg://cache=TEST_CACHE:distributedJoins=true@file:///C:\\ignite-client-config.xml")
]

The returned query result set is often huge and GC kicks in very often and
there is a huge impact on the query response time. 

Also we see "Query execution is too long" warning messages when GC kicks in.
There were scenarios when ignite server node went into non-responsive state
and never recovered back.

We would like to know if "Result Set" processing can happen out of GC
collectable heap like the way we can use cache in off-heap mode ??

Thanks.



--
View this message in context: 
http://apache-ignite-users.70518.x6.nabble.com/Result-set-processing-in-off-heap-mode-tp7988.html
Sent from the Apache Ignite Users mailing list archive at Nabble.com.

Is IgniteQueue transactional?

[akaptsan]

Does IgniteQueue support transactions?
 
Test scenario:
1. Start transaction
2. Put message to transactional queue
*3. Another thread peeks the queue and gets UNCOMMITED message!*
4. commit


  TransactionalQueue.java

  



--
View this message in context: 
http://apache-ignite-users.70518.x6.nabble.com/Is-IgniteQueue-transactional-tp7987.html
Sent from the Apache Ignite Users mailing list archive at Nabble.com.

Re: How to avoid the event lost in the continuous query

[ght230]

This example combines the code for remotefilter and locallistener.

Can you tell me how to run this example in detail?



--
View this message in context: 
http://apache-ignite-users.70518.x6.nabble.com/How-to-avoid-the-event-lost-in-the-continuous-query-tp7904p7985.html
Sent from the Apache Ignite Users mailing list archive at Nabble.com.