accumulo-notifications mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Josh Elser (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (ACCUMULO-3471) Adding a new tserver puts some tables offline for few minutes
Date Wed, 14 Jan 2015 06:08:34 GMT

    [ https://issues.apache.org/jira/browse/ACCUMULO-3471?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14276533#comment-14276533
] 

Josh Elser commented on ACCUMULO-3471:
--------------------------------------

ACCUMULO-1085 might be of interest. It's only in 1.7 right now though. I took a table, dropped
table.split.threshold wayy down, and made 24k tablets over 2 tservers (one physical machine).
Did a clean start up, and I'm sitting here painfully waiting for all of the tablets to get
assigned. Increasing the new property tserver.assignment.concurrent.max to 10 seems to help
the assignments speed along their merry way, but I'm still on order of 10 minutes to get all
of the tablets assigned.

The master log is showing about 10 assignments ack'ed every 500ms. I'm guessing I would see
the same things that Denis also saw on the tserver side.

> Adding a new tserver puts some tables offline for few minutes
> -------------------------------------------------------------
>
>                 Key: ACCUMULO-3471
>                 URL: https://issues.apache.org/jira/browse/ACCUMULO-3471
>             Project: Accumulo
>          Issue Type: Bug
>          Components: tserver
>    Affects Versions: 1.6.1
>         Environment: Ubuntu 12.04
>            Reporter: Denis Petrov
>             Fix For: 1.6.2, 1.7.0
>
>         Attachments: ACCUMULO-3471-balance-test.patch
>
>
> I run an Accumulo cluster with 15 tservers with about 6000 tablets on each (disks are
quite slow - each node has 2*4Tb SATA)
> When a new tserver added to the cluster, the rebalancing procedure starts.
> During this procedure some tablets are offline and unreachable during 5-10 minutes.
> It is visible in http://monitor:50095/tables and by timeouts on client side.
> The rebalancing caused by killing a tserver converges much faster then rebalancing caused
by adding a tserver.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message