phoenix-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Samarth Jain (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (PHOENIX-3360) Secondary index configuration is wrong
Date Fri, 14 Apr 2017 06:46:41 GMT

     [ https://issues.apache.org/jira/browse/PHOENIX-3360?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Samarth Jain updated PHOENIX-3360:
----------------------------------
    Attachment: indexlogging.patch

[~mujtabachohan] and I tested locally with HBase 0.98 that we still need the hbase.rpc.controllerfactory.class
config on the server side for the index updates to use the IndexRpcServer.handler thread pool.


I used the attached patch to log the thread name that is executing the index updates. Only
with both the configs hbase.rpc.controllerfactory.class and hbase.region.server.rpc.scheduler.factory.class
in place, we saw the Indexer using the IndexRpcServer thread pool. 

{code}
2017-04-14 03:29:47,519 INFO  [0,queue=0,port=60020] index.Indexer -  Indexer# preBatchMutate
Thread: IndexRpcServer.handler=20,queue=0,port=60020
2017-04-14 03:29:47,519 INFO  [0,queue=0,port=60020] index.Indexer -  Indexer# preBatchMutateWithExceptions
Thread: IndexRpcServer.handler=20,queue=0,port=60020
2017-04-14 03:29:47,520 INFO  [0,queue=0,port=60020] index.Indexer -  Indexer# doPre Thread:
IndexRpcServer.handler=20,queue=0,port=60020
2017-04-14 03:29:47,521 INFO  [0,queue=0,port=60020] index.Indexer -  Indexer# postBatchMutate
Thread: IndexRpcServer.handler=20,queue=0,port=60020
2017-04-14 03:29:47,521 INFO  [0,queue=0,port=60020] index.Indexer -  Indexer# doPost Thread:
IndexRpcServer.handler=20,queue=0,port=60020
2017-04-14 03:29:47,521 INFO  [0,queue=0,port=60020] index.Indexer -  Indexer# doPostWithExceptions
Thread: IndexRpcServer.handler=20,queue=0,port=60020
2017-04-14 03:29:47,524 INFO  [0,queue=0,port=60020] index.Indexer -  Indexer# postPut Thread:
IndexRpcServer.handler=20,queue=0,port=60020
2017-04-14 03:29:47,524 INFO  [0,queue=0,port=60020] index.Indexer -  Indexer# doPost Thread:
IndexRpcServer.handler=20,queue=0,port=60020
2017-04-14 03:29:47,524 INFO  [0,queue=0,port=60020] index.Indexer -  Indexer# doPostWithExceptions
Thread: IndexRpcServer.handler=20,queue=0,port=60020
{code}

We also saw the call being dispatched by the PhoenixRpcScheduler. 

{code}
2017-04-14 03:29:47,815 INFO  [.reader=8,port=60020] ipc.PhoenixRpcScheduler - Call dispatched
with index priority: callId: 5381 service: ClientService methodName: Multi size: 449 connection:
10.225.197.8:36044 param: region= X,,1492134552092.abe122d44a69701b81d304da339522de., for
1 actions and 1st row key=sdfdsfsdfdsfs djf dsajkfhads kfjlh ljddsfdsfsf dfg dfjkghh connection:
10.225.197.8:36044
{code}

> Secondary index configuration is wrong
> --------------------------------------
>
>                 Key: PHOENIX-3360
>                 URL: https://issues.apache.org/jira/browse/PHOENIX-3360
>             Project: Phoenix
>          Issue Type: Bug
>            Reporter: Enis Soztutar
>            Assignee: William Yang
>            Priority: Critical
>             Fix For: 4.10.0
>
>         Attachments: ConfCP.java, indexlogging.patch, PHOENIX-3360.patch, PHOENIX-3360-v2.PATCH,
PHOENIX-3360-v3.PATCH, PHOENIX-3360-v4.PATCH
>
>
> IndexRpcScheduler allocates some handler threads and uses a higher priority for RPCs.
The corresponding IndexRpcController is not used by default as it is, but used through ServerRpcControllerFactory
that we configure from Ambari by default which sets the priority of the outgoing RPCs to either
metadata priority, or the index priority.
> However, after reading code of IndexRpcController / ServerRpcController it seems that
the IndexRPCController DOES NOT look at whether the outgoing RPC is for an Index table or
not. It just sets ALL rpc priorities to be the index priority. The intention seems to be the
case that ONLY on servers, we configure ServerRpcControllerFactory, and with clients we NEVER
configure ServerRpcControllerFactory, but instead use ClientRpcControllerFactory. We configure
ServerRpcControllerFactory from Ambari, which in affect makes it so that ALL rpcs from Phoenix
are only handled by the index handlers by default. It means all deadlock cases are still there.

> The documentation in https://phoenix.apache.org/secondary_indexing.html is also wrong
in this sense. It does not talk about server side / client side. Plus this way of configuring
different values is not how HBase configuration is deployed. We cannot have the configuration
show the ServerRpcControllerFactory even only for server nodes, because the clients running
on those nodes will also see the wrong values. 



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

Mime
View raw message