[
https://issues.apache.org/jira/browse/PHOENIX-3994?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16084652#comment-16084652
]
Samarth Jain commented on PHOENIX-3994:
---------------------------------------
OK, confirmed that adding RpcSchedulerFactory in the preOpen() coprocessor hook
doesn't work. This is what I did.
I added the RpcSchedulerFactory in preOpen() of the Indexer co-processor and
BaseScannerRegionObserver.
{code}
@Override
public void preOpen(ObserverContext<RegionCoprocessorEnvironment> e) throws
IOException {
final RegionCoprocessorEnvironment env = e.getEnvironment();
env.getConfiguration().setClass(RSRpcServices.REGION_SERVER_RPC_SCHEDULER_FACTORY_CLASS,
PhoenixRpcSchedulerFactory.class, RpcSchedulerFactory.class);
super.preOpen(e);
}
{code}
and saw that the PhoenixRpcSchedulerFactory wasn't instantiated at all. So it
is too late. Looked briefly at the HBase code and saw that we get hold of the
RpcSchedulerFactory in the constructor of RSRpcServices which I am guessing
happens before preOpen() is called (kinda makes sense too since preOpen() on a
region is an RPC from HMaster which is handled/scheduled by region server
opening the region).
{code}
try {
Class<?> rpcSchedulerFactoryClass = rs.conf.getClass(
REGION_SERVER_RPC_SCHEDULER_FACTORY_CLASS,
SimpleRpcSchedulerFactory.class);
rpcSchedulerFactory = ((RpcSchedulerFactory)
rpcSchedulerFactoryClass.newInstance());
}
{code}
> Index RPC priority still depends on the controller factory property in
> hbase-site.xml
> -------------------------------------------------------------------------------------
>
> Key: PHOENIX-3994
> URL: https://issues.apache.org/jira/browse/PHOENIX-3994
> Project: Phoenix
> Issue Type: Bug
> Affects Versions: 4.11.0
> Reporter: Sergey Soldatov
> Assignee: Samarth Jain
> Priority: Critical
> Fix For: 4.12.0, 4.11.1
>
> Attachments: PHOENIX-3994_addendum.patch, PHOENIX-3994.patch,
> PHOENIX-3994_v2.patch, PHOENIX-3994_v3.patch
>
>
> During PHOENIX-3360 we tried to remove dependency on
> hbase.rpc.controllerfactory.class property in hbase-site.xml since it cause
> problems on the client side (if client is using server side configuration,
> all client request may go using index priority). Committed solution is using
> setting the controller factory programmatically for coprocessor environment
> in Indexer class, but it comes that this solution doesn't work because the
> environment configuration is not used for the coprocessor connection
> creation. We need to provide a better solution since this issue may cause
> accidental locks and failures that hard to identify and avoid.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
