You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@ignite.apache.org by VeenaMithare <v....@cmcmarkets.com> on 2020/05/02 18:07:52 UTC

REPLICATED CACHE ISSUES WHEN UPDATED THROUGH JDBC THIN CLIENT - 2.7.6

HI, 

I am running a 3 server node cluster on Ignite 2.7.6 .

1.  I have started a cache ( REPLICATED ) with native persistence enabled
with the below configuration . 


       cacheConfiguration.setRebalanceMode(CacheRebalanceMode.SYNC);
        cacheConfiguration.setCacheMode(CacheMode.REPLICATED);
       
cacheConfiguration.setWriteSynchronizationMode(CacheWriteSynchronizationMode.FULL_SYNC);
       
cacheConfiguration.setAtomicityMode(CacheAtomicityMode.TRANSACTIONAL);

I can see that my cache has been started using this mode : 
Started cache [name=SQL_PUBLIC_ABCDEF, id=-1243894080,
memoryPolicyName=null, mode=REPLICATED, atomicity=TRANSACTIONAL,
backups=2147483647, mvcc=false], encryptionEnabled=false]

If i connect to one of the servers through JDBC thin client ( DBeaver ) and
update a record in this cache - the change is visible only on the node where
the update was done. The other two nodes show the old value. 

Kindly guide if this is a bug or I am missing some configuration . 

2. Kindly note that if I create the same cache with CACHE_MODE : PARTITIONED
with BACKUPS =2 - an update from the DBeaver is propogated to all the nodes. 

Isnt the configuration in 1 and 2 the same ? Why does it show different
results ?

regards,
Veena.







--
Sent from: http://apache-ignite-users.70518.x6.nabble.com/

Re: REPLICATED CACHE ISSUES WHEN UPDATED THROUGH JDBC THIN CLIENT - 2.7.6

Posted by VeenaMithare <v....@cmcmarkets.com>.
Thanks Ilya, I will update the post with details a bit later. Stuck on
another issue on ignite right now.( Will send a separate post )



--
Sent from: http://apache-ignite-users.70518.x6.nabble.com/

Re: REPLICATED CACHE ISSUES WHEN UPDATED THROUGH JDBC THIN CLIENT - 2.7.6

Posted by Ilya Kasnacheev <il...@gmail.com>.
Hello!

This is very strange, I would assume that either you have three clusters
(each node has formed a cluster of its own) or that you have three
different caches.

Can you please elaborate what you are seeing? Can you provide logs from one
of the nodes? How do you update this cache?

Regards,
-- 
Ilya Kasnacheev


сб, 2 мая 2020 г. в 21:08, VeenaMithare <v....@cmcmarkets.com>:

> HI,
>
> I am running a 3 server node cluster on Ignite 2.7.6 .
>
> 1.  I have started a cache ( REPLICATED ) with native persistence enabled
> with the below configuration .
>
>
>        cacheConfiguration.setRebalanceMode(CacheRebalanceMode.SYNC);
>         cacheConfiguration.setCacheMode(CacheMode.REPLICATED);
>
>
> cacheConfiguration.setWriteSynchronizationMode(CacheWriteSynchronizationMode.FULL_SYNC);
>
> cacheConfiguration.setAtomicityMode(CacheAtomicityMode.TRANSACTIONAL);
>
> I can see that my cache has been started using this mode :
> Started cache [name=SQL_PUBLIC_ABCDEF, id=-1243894080,
> memoryPolicyName=null, mode=REPLICATED, atomicity=TRANSACTIONAL,
> backups=2147483647, mvcc=false], encryptionEnabled=false]
>
> If i connect to one of the servers through JDBC thin client ( DBeaver ) and
> update a record in this cache - the change is visible only on the node
> where
> the update was done. The other two nodes show the old value.
>
> Kindly guide if this is a bug or I am missing some configuration .
>
> 2. Kindly note that if I create the same cache with CACHE_MODE :
> PARTITIONED
> with BACKUPS =2 - an update from the DBeaver is propogated to all the
> nodes.
>
> Isnt the configuration in 1 and 2 the same ? Why does it show different
> results ?
>
> regards,
> Veena.
>
>
>
>
>
>
>
> --
> Sent from: http://apache-ignite-users.70518.x6.nabble.com/
>