Mailing-List: contact user-help@hbase.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@hbase.apache.org
Received-SPF: pass (nike.apache.org: local policy)
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws;
  s=s1024; d=yahoo.com;
  h=X-YMail-OSG:Received:X-Rocket-MIMEInfo:X-RocketYMMF:X-Mailer:References:Message-ID:Date:From:Reply-To:Subject:To:In-Reply-To:MIME-Version:Content-Type;
  b=cnVNFM+v3a5NtyRWQnkxHObxQX1mECx7iRvpecbtLHX9GqB+nJJGA+gN6dXKqbofg0s5HX+c3r8wNoaqMUBMXJPTBosVOU98n902WS3kFK4VZkDysGUAbYhwmQe2sgZVkRN77ITjdDe1FOj5rGwpzm8M/aSAFgArC/G2m2Q2sCU=;
References: 
 <CA+vHzD9A0=cmxPY=H6-YabR3DJHdZFQ=mGQJsORC3gvLS7UytA@mail.gmail.com>
	<CAOEq2C4pOH0SPg53PqEA1E8pRLtjkjk3BdDvooDQLQGgMyzzPg@mail.gmail.com>
	<1395208267.61135.YahooMailNeo@web140602.mail.bf1.yahoo.com>
 <CAOEq2C6u9Zo4t176PnOjqhV0XDhEYQcDoAretnCrKS22o5d0fQ@mail.gmail.com>
Message-ID: <1395382810.64534.YahooMailNeo@web140602.mail.bf1.yahoo.com>
Date: Thu, 20 Mar 2014 23:20:10 -0700 (PDT)
From: lars hofhansl <larsh@apache.org>
Reply-To: lars hofhansl <larsh@apache.org>
Subject: Re: How much time HBase will take if one Region Server down.
To: Demai Ni <nidmgg@gmail.com>,
  "user@hbase.apache.org" <user@hbase.apache.org>
In-Reply-To: 
 <CAOEq2C6u9Zo4t176PnOjqhV0XDhEYQcDoAretnCrKS22o5d0fQ@mail.gmail.com>
MIME-Version: 1.0
Content-Type: multipart/alternative;
 boundary="-118416272-766108868-1395382810=:64534"

---118416272-766108868-1395382810=:64534
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable

Depends. If a machine fails hard and HDFS 2.x is setup correctly I'd expect=
 this to be the ZK timeout (180s by default in 0.94, but can be lowered) + =
a few minutes to rejigger things.=0A=0A-- Lars=0A=0A=0A=0A_________________=
_______________=0A From: Demai Ni <nidmgg@gmail.com>=0ATo: "user@hbase.apac=
he.org" <user@hbase.apache.org>; lars hofhansl <larsh@apache.org> =0ASent: =
Wednesday, March 19, 2014 9:16 AM=0ASubject: Re: How much time HBase will t=
ake if one Region Server down.=0A =0A=0A=0ALars,=0A=0AI didn't get the 15~2=
0 min from the article. I played a bit with my clusters on 94 early last ye=
ar, and got a ball-park number. Well, my cluster was using all the default =
hbase configuration at that time, so it definitely wasn't well tuned.=A0 So=
 for HBase 94, what's a reasonable MTTR # from your experience? Thanks=0A=
=0ADemai=0A=0A=0A=0A=0AOn Tue, Mar 18, 2014 at 10:51 PM, lars hofhansl <lar=
sh@apache.org> wrote:=0A=0AMany of the improvement were made in HDFS, so be=
nefit 0.94 equally.=0A>Not sure where you read 15-20 minutes for 0.94 in th=
at article.=0A>=0A>-- Lars=0A>=0A>=0A>=0A>________________________________=
=0A>=A0From: Demai Ni <nidmgg@gmail.com>=0A>To: "user@hbase.apache.org" <us=
er@hbase.apache.org>=0A>Sent: Tuesday, March 18, 2014 10:59 AM=0A>Subject: =
Re: How much time HBase will take if one Region Server down.=0A>=0A>=0A>=0A=
>a good article to start with:=0A>http://hortonworks.com/blog/introduction-=
to-hbase-mean-time-to-recover-mttr/=0A>=0A>I would say 15~20 min for 94 or =
earlier; it can be reduced to a couple min=0A>with HBase 96+ and Hadoop .=
=0A>=0A>Demai=0A>=0A>=0A>=0A>On Tue, Mar 18, 2014 at 10:55 AM, Upendra Yada=
v <upendra1024@gmail.com>wrote:=0A>=0A>> How much time HBase will take if o=
ne Region Server down and try to read=0A>> Region that belong to this down =
Region Server.=0A>>=0A>> Means Is there service down for those all region t=
hat belongs to dead RS?=0A>>=0A>> How much time HMaster will take to assign=
 those region to live RS? What is=0A>> the dependencies that may increase r=
egion assignment in such situation.=0A>>=0A>> Thanks....=0A>>
---118416272-766108868-1395382810=:64534--