Mailing-List: contact user-help@cassandra.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@cassandra.apache.org
Received-SPF: pass (nike.apache.org: domain of joolski@gmail.com designates
 74.125.82.172 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=gamma;
        h=mime-version:in-reply-to:references:date:message-id:subject:from:to
         :content-type;
        b=ovtZBjeKDj/t2TpoKNY8Eyba1UtX0qygHVfXNbsbEZb0kgvwkuYWZko+/MDLG+FwYY
         k2M0ijWZZxpV9ICUOIZ2blH0R/LF0FHhkSwgaEV79p2428bLMYGTXC3vF8GBGxlt/+NH
         Qx6Q5a1gAObZ1xBmRHfcw5wIPL9aUw5Wfj2hg=
MIME-Version: 1.0
In-Reply-To: <AANLkTimAfITg7Nq-_HmoEs8yXYuHwIiFeYS9AEtO0hr9@mail.gmail.com>
References: <AANLkTil7IMR0j41c27lWfFryYJCcT74RwqQIU5eZD7ok@mail.gmail.com>
	<1OMH23-0003qP-UN@mail.eleven.de>
	<AANLkTiknv_Koj_e-7o6cYGHkbuWpVJ5Vua8-hT8AAuWT@mail.gmail.com>
	<AANLkTimAfITg7Nq-_HmoEs8yXYuHwIiFeYS9AEtO0hr9@mail.gmail.com>
Date: Wed, 9 Jun 2010 10:57:14 +0100
Message-ID: <AANLkTimz96GIOnWZlnaBC3WQVFEdDhVAJKdis7yt0OEA@mail.gmail.com>
Subject: Re: Inserting new data, where the key points to a tombstone record.
From: Jools <joolski@gmail.com>
To: user@cassandra.apache.org
Content-Type: multipart/alternative; boundary=0016e649ce743327ad048895ee4b

--0016e649ce743327ad048895ee4b
Content-Type: text/plain; charset=ISO-8859-1

On 9 June 2010 10:43, Sylvain Lebresne <sylvain@yakaz.com> wrote:

> > However, as a final point of clarification, is there a particular reason
> > that insert does not raise an exception when trying to insert over an
> > existing key, or when the key points to a tombstone record ?
>
> Inserting over an existing key is an update of the record and in the
> Cassandra
> data model, this is nothing exceptional, this is the normal way to do
> things,
> by design.
>
> There is however a technical reason why it would be hard to return an
> exception in the case you mention. The way Cassandra does inserts, it don't
> know when you insert something that another record with the same key
> exists.
> Writes only append stuffs. Checking that the key exists would slow down
> writes
> quite a bit.
>
>
I can see there are technical, and performance reasons for not checking that
a key exists. And from reading Martin's post it's clear that to make
certain scenarios work as expected one needs to alter the way the data is
organised.

I've got plenty of pointers now to allow me to alter my datasets to cover
these issues now, many thanks Martin and Sylvain for taking the time to
respond.

Best regards,

--Jools

--0016e649ce743327ad048895ee4b
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<br><br><div class=3D"gmail_quote">On 9 June 2010 10:43, Sylvain Lebresne <=
span dir=3D"ltr">&lt;<a href=3D"mailto:sylvain@yakaz.com">sylvain@yakaz.com=
</a>&gt;</span> wrote:<br><blockquote class=3D"gmail_quote" style=3D"margin=
:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<div class=3D"im">&gt; However, as a final point of clarification, is there=
 a particular reason<br>
&gt; that insert does not raise an exception when trying to insert over an<=
br>
&gt; existing key, or when the key points to a tombstone record ?<br>
<br>
</div>Inserting over an existing key is an update of the record and in the =
Cassandra<br>
data model, this is nothing exceptional, this is the normal way to do thing=
s,<br>
by design.<br>
<br>
There is however a technical reason why it would be hard to return an<br>
exception in the case you mention. The way Cassandra does inserts, it don&#=
39;t<br>
know when you insert something that another record with the same key exists=
.<br>
Writes only append stuffs. Checking that the key exists would slow down wri=
tes<br>
quite a bit.<br><br></blockquote><div><br></div><div>I can see there are te=
chnical, and performance reasons for not checking that a key exists. And fr=
om reading Martin&#39;s post it&#39;s clear that to make certain=A0scenario=
s work as expected one needs to alter the way the data is organised.</div>
<div><br></div><div>I&#39;ve got plenty of pointers now to allow me to alte=
r my datasets to cover these issues now, many thanks Martin and=A0Sylvain f=
or taking the time to respond.</div><div><br></div><div>Best regards,</div>
<div><br></div><div>--Jools</div><div><br></div><div><br></div><div><br></d=
iv><div>=A0=A0</div><div><br></div><div>=A0</div></div><br>

--0016e649ce743327ad048895ee4b--