Re: error while working solr 6.2 cdcr

2016-11-10 Thread Neeraj Bhatt 
Hi Eric

With one shard and one replica  I am getting " 08976 ERROR
(qtp110456297-13) [c:multi_dc_poc s:shard1 r:core_node1
x:multi_dc_poc_shard1_replica1] o.a.s.h.RequestHandlerBase
org.apache.solr.common.SolrException: Unknown action: null
at
org.apache.solr.handler.CdcrRequestHandler.handleRequestBody(CdcrRequestHandler.java:173)"

in target logs


Any suggestion ?

Thanks

On Thu, Nov 10, 2016 at 8:17 AM, Neeraj Bhatt 
wrote:

> Hi Eric
>
> Source have only target zookeeper ensemble (only 3 ips)
>   192.169.35.53:2181,192.169.35.249:2181,192.
> 169.35.251:2181> , because of some mail formatting ip were repeated(see
> sent item attached image). Error in log files is
> Action LASTPROCESSEDVERSION sent to non-leader replica at
> org.apache.solr.handler.CdcrRequestHandler.handleLastProcess
> edVersionAction(CdcrRequestHandler.java:498)
> I will start with empty collection with  3 nodes 3 shards one replication
> factor, It seems from log that request from souce is going to non leader
> replica.
>
> Please suggest
>
> Thanks
>
>
>
> On Wed, Nov 9, 2016 at 9:23 PM, Erick Erickson 
> wrote:
>
>> Why do you have 6 zookeepers listed in the zkHost parameter? My guess
>> is that you've listed the ZK addresses for both the source and target
>> clusters. This should only be the ZK nodes for the target cluster and
>> only present on the source.
>>
>> The source already knows its own ZK addresses, it needs to know about
>> the target cluster in order to know what nodes it should send updates
>> to.
>>
>> Best,
>> Erick
>>
>> On Wed, Nov 9, 2016 at 4:44 AM, Neeraj Bhatt 
>> wrote:
>> > Hello All
>> >
>> > I have a source cluster (3 nodes, 3 shards and 3 replica) and same for
>> > target cluster. On posting of documents to source I am getting below
>> error
>> > in target solr log files when i turned log to debug level
>> >
>> >
>> >
>> >
>> > *6538 ERROR (qtp110456297-20) [c:multi_dc_poc s:shard1 r:core_node3
>> > x:multi_dc_poc_shard1_replica2] o.a.s.h.RequestHandlerBase
>> > org.apache.solr.common.SolrException: Action LASTPROCESSEDVERSION sent
>> to
>> > non-leader replicaat
>> > org.apache.solr.handler.CdcrRequestHandler.handleLastProcess
>> edVersionAction(CdcrRequestHandler.java:498)*
>> >
>> > *should I reconfigure my replicas to one in target ?*
>> >
>> > solr config of source and target is same as given in wiki
>> > https://cwiki.apache.org/confluence/pages/viewpage.action?pa
>> geId=62687462
>> >
>> > *source *
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> > *> > class="solr.CdcrRequestHandler"> > > name="replica"> > > name="zkHost">192.169.35.53:2181
>> > ,192.169.35.249:2181
>> > ,192.169.35.251:2181
>> > 
>> >  > > name="source">multi_dc_poc > > name="target">multi_dc_poc
>> > > name="replicator">> > name="threadPoolSize">9> > name="schedule">1000> > name="batchSize">128> > name="updateLogSynchronizer">> > name="schedule">1000  
>> *
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> > *source and target update log  > > class="solr.CdcrUpdateLog">  > > name="dir">${solr.ulog.dir:}  > > name="numVersionBuckets">${solr.ulog.numVersionBuckets:65536}
>> > *
>> >
>> >
>> > *target solr config*
>> >
>> > 
>> >   
>> > disabled
>> >   
>> > 
>> >
>> > 
>> >   
>> > cdcr-processor-chain
>> >   
>> > 
>> >
>> > 
>> >   
>> >   
>> > 
>> >
>> > *target status*
>> >
>> > 0> > name="QTime">0> > name="process">stopped> > name="buffer">disabled
>> >
>> > *source status*
>> >
>> > 0> > name="QTime">0> > name="process">started> > name="buffer">enabled
>>
>
>

Re: error while working solr 6.2 cdcr

2016-11-09 Thread Neeraj Bhatt 
Hi Eric

Source have only target zookeeper ensemble (only 3 ips)
  192.169.35.53:2181,192.169.35.249:2181,
192.169.35.251:2181> , because of some mail formatting ip were repeated(see
sent item attached image). Error in log files is
Action LASTPROCESSEDVERSION sent to non-leader replica at
org.apache.solr.handler.CdcrRequestHandler.handleLastProcess
edVersionAction(CdcrRequestHandler.java:498)
I will start with empty collection with  3 nodes 3 shards one replication
factor, It seems from log that request from souce is going to non leader
replica.

Please suggest

Thanks



On Wed, Nov 9, 2016 at 9:23 PM, Erick Erickson 
wrote:

> Why do you have 6 zookeepers listed in the zkHost parameter? My guess
> is that you've listed the ZK addresses for both the source and target
> clusters. This should only be the ZK nodes for the target cluster and
> only present on the source.
>
> The source already knows its own ZK addresses, it needs to know about
> the target cluster in order to know what nodes it should send updates
> to.
>
> Best,
> Erick
>
> On Wed, Nov 9, 2016 at 4:44 AM, Neeraj Bhatt 
> wrote:
> > Hello All
> >
> > I have a source cluster (3 nodes, 3 shards and 3 replica) and same for
> > target cluster. On posting of documents to source I am getting below
> error
> > in target solr log files when i turned log to debug level
> >
> >
> >
> >
> > *6538 ERROR (qtp110456297-20) [c:multi_dc_poc s:shard1 r:core_node3
> > x:multi_dc_poc_shard1_replica2] o.a.s.h.RequestHandlerBase
> > org.apache.solr.common.SolrException: Action LASTPROCESSEDVERSION sent
> to
> > non-leader replicaat
> > org.apache.solr.handler.CdcrRequestHandler.handleLastProcess
> edVersionAction(CdcrRequestHandler.java:498)*
> >
> > *should I reconfigure my replicas to one in target ?*
> >
> > solr config of source and target is same as given in wiki
> > https://cwiki.apache.org/confluence/pages/viewpage.action?pa
> geId=62687462
> >
> > *source *
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> > * > class="solr.CdcrRequestHandler">  > name="replica">  > name="zkHost">192.169.35.53:2181
> > ,192.169.35.249:2181
> > ,192.169.35.251:2181
> > 
> >   > name="source">multi_dc_poc  > name="target">multi_dc_poc
>  > name="replicator"> > name="threadPoolSize">9 > name="schedule">1000 > name="batchSize">128 > name="updateLogSynchronizer"> > name="schedule">1000  
> *
> >
> >
> >
> >
> >
> >
> >
> > *source and target update log   > class="solr.CdcrUpdateLog">   > name="dir">${solr.ulog.dir:}   > name="numVersionBuckets">${solr.ulog.numVersionBuckets:65536}
> > *
> >
> >
> > *target solr config*
> >
> > 
> >   
> > disabled
> >   
> > 
> >
> > 
> >   
> > cdcr-processor-chain
> >   
> > 
> >
> > 
> >   
> >   
> > 
> >
> > *target status*
> >
> > 0 > name="QTime">0 > name="process">stopped > name="buffer">disabled
> >
> > *source status*
> >
> > 0 > name="QTime">0 > name="process">started > name="buffer">enabled
>

Re: error while working solr 6.2 cdcr

2016-11-09 Thread Erick Erickson 
Why do you have 6 zookeepers listed in the zkHost parameter? My guess
is that you've listed the ZK addresses for both the source and target
clusters. This should only be the ZK nodes for the target cluster and
only present on the source.

The source already knows its own ZK addresses, it needs to know about
the target cluster in order to know what nodes it should send updates
to.

Best,
Erick

On Wed, Nov 9, 2016 at 4:44 AM, Neeraj Bhatt  wrote:
> Hello All
>
> I have a source cluster (3 nodes, 3 shards and 3 replica) and same for
> target cluster. On posting of documents to source I am getting below error
> in target solr log files when i turned log to debug level
>
>
>
>
> *6538 ERROR (qtp110456297-20) [c:multi_dc_poc s:shard1 r:core_node3
> x:multi_dc_poc_shard1_replica2] o.a.s.h.RequestHandlerBase
> org.apache.solr.common.SolrException: Action LASTPROCESSEDVERSION sent to
> non-leader replicaat
> org.apache.solr.handler.CdcrRequestHandler.handleLastProcessedVersionAction(CdcrRequestHandler.java:498)*
>
> *should I reconfigure my replicas to one in target ?*
>
> solr config of source and target is same as given in wiki
> https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=62687462
>
> *source *
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> * class="solr.CdcrRequestHandler">  name="replica">  name="zkHost">192.169.35.53:2181
> ,192.169.35.249:2181
> ,192.169.35.251:2181
> 
>   name="source">multi_dc_poc  name="target">multi_dc_poc name="replicator"> name="threadPoolSize">9 name="schedule">1000 name="batchSize">128 name="updateLogSynchronizer"> name="schedule">1000  *
>
>
>
>
>
>
>
> *source and target update log   class="solr.CdcrUpdateLog">   name="dir">${solr.ulog.dir:}   name="numVersionBuckets">${solr.ulog.numVersionBuckets:65536}
> *
>
>
> *target solr config*
>
> 
>   
> disabled
>   
> 
>
> 
>   
> cdcr-processor-chain
>   
> 
>
> 
>   
>   
> 
>
> *target status*
>
> 0 name="QTime">0 name="process">stopped name="buffer">disabled
>
> *source status*
>
> 0 name="QTime">0 name="process">started name="buffer">enabled

error while working solr 6.2 cdcr

2016-11-09 Thread Neeraj Bhatt 
Hello All

I have a source cluster (3 nodes, 3 shards and 3 replica) and same for
target cluster. On posting of documents to source I am getting below error
in target solr log files when i turned log to debug level




*6538 ERROR (qtp110456297-20) [c:multi_dc_poc s:shard1 r:core_node3
x:multi_dc_poc_shard1_replica2] o.a.s.h.RequestHandlerBase
org.apache.solr.common.SolrException: Action LASTPROCESSEDVERSION sent to
non-leader replicaat
org.apache.solr.handler.CdcrRequestHandler.handleLastProcessedVersionAction(CdcrRequestHandler.java:498)*

*should I reconfigure my replicas to one in target ?*

solr config of source and target is same as given in wiki
https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=62687462

*source *




















*  192.169.35.53:2181
,192.169.35.249:2181
,192.169.35.251:2181

 multi_dc_poc multi_dc_poc910001281000  *







*source and target update log${solr.ulog.dir:}  ${solr.ulog.numVersionBuckets:65536}
*


*target solr config*


  
disabled
  



  
cdcr-processor-chain
  



  
  


*target status*

00stoppeddisabled

*source status*

00startedenabled