phoenix-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Samarth Jain (JIRA)" <>
Subject [jira] [Commented] (PHOENIX-3994) Index RPC priority still depends on the controller factory property in hbase-site.xml
Date Wed, 12 Jul 2017 20:55:00 GMT


Samarth Jain commented on PHOENIX-3994:

[~vincentpoon] - from what I understand, every co-processor gets its own environment. And
once you modify the configuration in that environment, subsequent invocations of the co-processor
see it too. I verified that this indeed is the case. Which is why I made the change in UngroupedAggregateRegionObserver
to clone the config since we use that co-processor for various other queries and not just
server side upsert select.

Also, see this comment:

It would be good to verify too though that if one coprocessor changes the config encapsulated
in it's own environment, then other co-processors on the same region won't see it. It would
be good if [~apurtell] too could confirm. 

If that isn't the case, I would think its a pretty bad bug on HBase's part unless it is by
design which I would argue against too.

> Index RPC priority still depends on the controller factory property in hbase-site.xml
> -------------------------------------------------------------------------------------
>                 Key: PHOENIX-3994
>                 URL:
>             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,
> 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

View raw message