Mailing-List: contact zookeeper-user-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: zookeeper-user@hadoop.apache.org
Received-SPF: neutral (athena.apache.org: local policy)
DomainKey-Signature: a=rsa-sha1; s=serpent; d=yahoo-inc.com; c=nofws; q=dns;
	h=received:from:to:date:subject:thread-topic:thread-index:
	message-id:references:in-reply-to:accept-language:
	content-language:x-ms-has-attach:x-ms-tnef-correlator:acceptlanguage:
	content-type:content-transfer-encoding:mime-version;
	b=RVIeXWkKVkWPrCU8Cuv7oIgA3eplYbmPJclqtotHqWSw2LtfQGodbdoqNf7Z70gG
From: Benjamin Reed <breed@yahoo-inc.com>
To: "zookeeper-user@hadoop.apache.org" <zookeeper-user@hadoop.apache.org>
Date: Thu, 18 Dec 2008 14:59:37 -0800
Subject: RE: What happens when a server loses all its state?
Thread-Topic: What happens when a server loses all its state?
Thread-Index: Aclgmi95aqVOvHXpTuiHTLo9X0iYRwAygLfQ
Message-ID: 
 <6990D2A1CAF07E40A7CFE68A5FAAA15317CF9BE2B6@SP1-EX07VS02.ds.corp.yahoo.com>
References: <C56D700D.15CB8%mahadev@yahoo-inc.com> <49484193.60300@sun.com>
 <6990D2A1CAF07E40A7CFE68A5FAAA15317C7FCF519@SP1-EX07VS02.ds.corp.yahoo.com>
 <4949828E.5070407@sun.com>
In-Reply-To: <4949828E.5070407@sun.com>
Accept-Language: en-US
Content-Language: en-US
acceptlanguage: en-US
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0

I have opened ZOOKEEPER-261 for this issue. it shouldn't be too hard to fix=
 and it would be nice to target for 3.1.

ben

-----Original Message-----
From: Thomas.Johnson@Sun.COM [mailto:Thomas.Johnson@Sun.COM]=20
Sent: Wednesday, December 17, 2008 2:52 PM
To: zookeeper-user@hadoop.apache.org
Subject: Re: What happens when a server loses all its state?

Thanks for all the responses.
Benjamin Reed wrote:
> Thomas,=20
>
> in the scenario you give you have two simultaneous failures with 3 nodes,=
 so it will not recover correctly. A is failed because it is not up. B has =
failed because it lost all its data.
>
> it would be good for ZooKeeper to not come up in that scenario. perhaps w=
hat we need is something similar to your safe state proposal. basically a s=
erver that has forgotten everything should not be allowed to vote in the le=
ader election. that would avoid your scenario. we just need to put a flag f=
ile in the data directory to say that the data is valid and thus can vote.
>
> ben
> ________________________________________
>  =20
Would this feature be something you'd consider implementing in the short=20
to medium term?