Mailing-List: contact user-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@hadoop.apache.org
Received-SPF: pass (nike.apache.org: local policy)
Subject: Re: HA for hadoop-0.20.2
Mime-Version: 1.0 (Apple Message framework v1085)
Content-Type: text/plain; charset=utf-8
From: "ac@hsk.hk" <ac@hsk.hk>
In-Reply-To: 
 <CAOcnVr0ssPAmz4GSLig9QBBPZhfpanknnVuarorA8-3iqdqqsA@mail.gmail.com>
Date: Tue, 13 Nov 2012 19:25:52 +0800
Cc: "ac@hsk.hk" <ac@hsk.hk>
Content-Transfer-Encoding: quoted-printable
Message-Id: <A20961C8-4C10-4901-ABDC-F55311579FB7@hsk.hk>
References: 
 <CABbGW3xX7yV+ARX6nBgOEDaO37f4eDUn6VExnSD5=kTcL5kn_w@mail.gmail.com>
 <CAOcnVr0ssPAmz4GSLig9QBBPZhfpanknnVuarorA8-3iqdqqsA@mail.gmail.com>
To: user@hadoop.apache.org

Hi, A question, is 2.x ready for production deployment?  Thanks

On 13 Nov 2012, at 5:19 PM, Harsh J wrote:

> Hi,
>=20
> Why not just use the 2.x releases for HA-NNs? There is quite a wide
> delta between 0.20.x and 2.x, especially around the edit log areas
> after HDFS-1073.
>=20
> In any case, I think your question suits hdfs-dev@hadoop.apache.org
> more than the user lists, although I don't quite understand what
> you're attempting to do (or point).
>=20
> On Tue, Nov 13, 2012 at 2:18 PM, lei liu <liulei412@gmail.com> wrote:
>> I want to implement HA function for hadoop-0.20.2.
>>=20
>> When I learn the hadoop-2.0 code, I meet some question like this:
>>=20
>> Thera are below code in FSEditLogLoader.loadEditRecords method.
>>=20
>>   if (op.hasTransactionId()) {
>>            if (op.getTransactionId() > expectedTxId) {
>>              MetaRecoveryContext.editLogLoaderPrompt("There appears " =
+
>>                  "to be a gap in the edit log.  We expected txid " +
>>                  expectedTxId + ", but got txid " +
>>                  op.getTransactionId() + ".", recovery, "ignoring =
missing "
>> +
>>                  " transaction IDs");
>>            } else if (op.getTransactionId() < expectedTxId) {
>>              MetaRecoveryContext.editLogLoaderPrompt("There appears " =
+
>>                  "to be an out-of-order edit in the edit log.  We " +
>>                  "expected txid " + expectedTxId + ", but got txid " =
+
>>                  op.getTransactionId() + ".", recovery,
>>                  "skipping the out-of-order edit");
>>              continue;
>>            }
>>          }
>>=20
>> The method use transaction id to  guarantee same transaction log is =
not
>> applied to namespace more than once.
>>=20
>> But in hadoop-0.20.2, FSEditLog don't store the transaction id into =
edits
>> log file. So I want to know if  StandbyNN apply same transaction log  =
to
>> namespace more than once, that will lead to the namespace of =
StandbyNN is
>> corrupt?
>>=20
>> Please give me some advice=EF=BC=8CThanks.
>>=20
>>=20
>>=20
>> Best Regards
>>=20
>> LiuLei
>=20
>=20
>=20
> --=20
> Harsh J