Hi
Issue got resolved after doing the following changes.
Attached you the updated code,client.xml(will be in src/resources folder)
and pom.xml file.
srcandpomupdated.zip
<http://apache-ignite-users.70518.x6.nabble.com/file/t2737/srcandpomupdated.zip>
1. Removed the following dependency from pom.xml
<dependency>
<groupId>org.apache.ignite</groupId>
<artifactId>ignite-spring-boot-autoconfigure-ext</artifactId>
<version>1.0.0</version>
</dependency>
2. Modifying the IgniteConfig.class such that
changing the method igniteInstance to return Ignite
instance by calling
Ignition.start("Ignite-Client.xml").
Eariler it used to return IgniteConfiguration
by using the following piece of code.Also earlier
i haven't used IgniteClient.xml.
IgniteConfiguration cfg = new IgniteConfiguration();
// The node will be started as a client node.
cfg.setClientMode(true);
// Classes of custom Java logic will be transferred over the
wire
from this app.
cfg.setPeerClassLoadingEnabled(true);
// Setting up an IP Finder to ensure the client can locate the
servers.
TcpDiscoveryMulticastIpFinder ipFinder = new
TcpDiscoveryMulticastIpFinder();
ipFinder.setAddresses(Collections.singletonList("Ipaddress"));
cfg.setDiscoverySpi(new
TcpDiscoverySpi().setIpFinder(ipFinder));
// Starting the node
cfg.setSqlSchemas("PIE");
retrun cfg
Although this seems to be working.
I have following queries with this approach.Can you please
provide your inputs/suggestions.
We are going to expose this SpringBoot
App as a RestFul API where we are trying to
refresh the Ignite Cache by sending request to this
Application whenever the data in Oracle changes.
.
As we are trying to expose this APP as a API,
which means that this Ignite Client Node
instance launched from Spring boot Application will
be active for ever.Now my queries are
1. Is it ok to have a client node which
is continously in connected state forever.
If Ok,then how many such client node can exist.
2. Can this Application, be hosted on the
one of the server node with different port.
3. Also will client node connect automatically
to the cluster if the client node has lost
communication with server because if the client
node is down and if there is an API request
, then the application will throw an error.
4. Is this way of refreshing the cache is efficient.
If no, can you please suggest a better way.
5. Also take provide an example on the possibility
of having the REST API calling a
compute job which would
execute in a separate server
https://apacheignite.readme.io/docs/compute-grid
Also i have few more queries related to Ignite
Behaviour.It would be really helpful if you can
provide some inputs.
1. Currently we are using
ignite.cache("CacheName").loadCache(null) method
to load the cache initially and also increamentally as
this method loads data very fast almost 0.1 million
in 1 minute. For increamental load also we are using this
method
by passing an additional argument i.e. Custom
Sql to load/update only the records that have created/changed.
but what is happening is if the key doesn't
exist it is inserting the data but if the key is present
it is not updating its value.
Why is the value not updated if the key is present.
2. To overcome this, i have to build my own custom
cache store as mentioned in the following link
https://apacheignite.readme.io/v1.9/docs/data-loading.
But the problem with this approach is i need to use JDBC
result set to load the data which is very slow as
we have to insert row by row.
Is there a better way to update the keys.
3. How to load data faster with custom cache store
as loading the data which is very slow as
we have to insert row by row.
--
Sent from: http://apache-ignite-users.70518.x6.nabble.com/
