Mailing-List: contact user-help@cassandra.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@cassandra.apache.org
Received-SPF: pass (athena.apache.org: domain of tobias.hauth@gmail.com
 designates 209.85.213.48 as permitted sender)
MIME-Version: 1.0
In-Reply-To: 
 <CA+4UHyNg4XYZxdgUxvhczP=14QET1m5E3=7p-zv5S1apj_CdSg@mail.gmail.com>
References: 
 <CAHspM=jo4cL8KgUXmSvk=yusa_LqWdvjhoVBGE_XX0nGxiD7Ng@mail.gmail.com>
 <CA+4UHyNg4XYZxdgUxvhczP=14QET1m5E3=7p-zv5S1apj_CdSg@mail.gmail.com>
From: Tobias Hauth <tobias.hauth@gmail.com>
Date: Fri, 6 Mar 2015 09:32:45 -0800
Message-ID: 
 <CAE_BpsZ5Wa5aKOa5N72QRA=AXcyp9y7piHNANT+4xU9vkA7Lwg@mail.gmail.com>
Subject: Re: Downgrade Cassandra from 2.1.x to 2.0.x
To: user@cassandra.apache.org
Content-Type: multipart/alternative; boundary=001a11c20266eb50760510a214d3

--001a11c20266eb50760510a214d3
Content-Type: text/plain; charset=UTF-8

Hi Roni,

We had the same problem and had to build a new, separate cluster. We read
from both clusters in the beginning and copied the data in the background
by paging over our data in the old 2.1.3 cluster. This allowed us to make
the downgrade seemless. Just watch out for 2.0.12's bug with nodetool
cleanup and make sure you build the cluster with enough nodes so you don't
need to add another node before version 2.0.13 gets released.

Best,
Tobias

On Fri, Mar 6, 2015 at 7:59 AM, Kai Wang <depend@gmail.com> wrote:

> AFAIK downgrading is not officially supported.
>
> how much data do you have? If all possible I would dump all my data out
> and bulk load them into the 2.0.x cluster. This is the only way I feel safe.
> On Mar 6, 2015 5:55 AM, "Roni Balthazar" <ronibalthazar@gmail.com> wrote:
>
>> Hi there,
>>
>> What is the best way to downgrade a C* 2.1.3 cluster to the stable 2.0.12?
>> I know it's not supported, but we are getting too many issues with the
>> 2.1.x...
>> It is leading us to think that the best solution is to use the stable
>> version.
>> Is there a safe way to do that?
>>
>> Cheers,
>>
>> Roni
>>
>

--001a11c20266eb50760510a214d3
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Hi Roni,<div><br></div><div>We had the same problem and ha=
d to build a new, separate cluster. We read from both clusters in the begin=
ning and copied the data in the background by paging over our data in the o=
ld 2.1.3 cluster. This allowed us to make the downgrade seemless. Just watc=
h out for 2.0.12&#39;s bug with nodetool cleanup and make sure you build th=
e cluster with enough nodes so you don&#39;t need to add another node befor=
e version 2.0.13 gets released.</div><div><br></div><div>Best,</div><div>To=
bias</div></div><div class=3D"gmail_extra"><br><div class=3D"gmail_quote">O=
n Fri, Mar 6, 2015 at 7:59 AM, Kai Wang <span dir=3D"ltr">&lt;<a href=3D"ma=
ilto:depend@gmail.com" target=3D"_blank">depend@gmail.com</a>&gt;</span> wr=
ote:<br><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border=
-left:1px #ccc solid;padding-left:1ex"><p dir=3D"ltr">AFAIK downgrading is =
not officially supported.</p>
<p dir=3D"ltr">how much data do you have? If all possible I would dump all =
my data out and bulk load them into the 2.0.x cluster. This is the only way=
 I feel safe.</p><div class=3D"HOEnZb"><div class=3D"h5">
<div class=3D"gmail_quote">On Mar 6, 2015 5:55 AM, &quot;Roni Balthazar&quo=
t; &lt;<a href=3D"mailto:ronibalthazar@gmail.com" target=3D"_blank">ronibal=
thazar@gmail.com</a>&gt; wrote:<br type=3D"attribution"><blockquote class=
=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padd=
ing-left:1ex">Hi there,<br>
<br>
What is the best way to downgrade a C* 2.1.3 cluster to the stable 2.0.12?<=
br>
I know it&#39;s not supported, but we are getting too many issues with the =
2.1.x...<br>
It is leading us to think that the best solution is to use the stable versi=
on.<br>
Is there a safe way to do that?<br>
<br>
Cheers,<br>
<br>
Roni<br>
</blockquote></div>
</div></div></blockquote></div><br></div>

--001a11c20266eb50760510a214d3--