Mailing-List: contact user-help@cassandra.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@cassandra.apache.org
MIME-Version: 1.0
In-Reply-To: 
 <CAOafRRdGwEZoS2wgavvntsVLAXT5KqrS6PuKa387B3DreC3JvQ@mail.gmail.com>
References: 
 <CAOafRRd2ig+Gcb6gkR_D7LmpJuVcUTGixB8X-a_uBQtKrE=Big@mail.gmail.com>
 <CAK0tFt67TcALfEkfL4ZCWRGZHfjZ2q51=+KCzQDVMV-6x2RbFQ@mail.gmail.com>
 <CAOafRRdL2-CMhfnw7Lc7jone-pEdgbu0dmjnCF3ovEvwxH3tag@mail.gmail.com>
 <CAK0tFt507mQ2coTCtKMzqnycLHbVbyf+xn4S5seSsnzwrCjsQw@mail.gmail.com>
 <CAOafRRf6VKQSbsGR2LoWUTAp9QNj53+eFXnEeC7a-5ONahraCA@mail.gmail.com>
 <CAK0tFt4VE0TeULWYMGS07yo7Y4=q=fJhUEi1JPst_iJBBtswNg@mail.gmail.com>
 <CAAam9sstbbjjRwbru8bxXasvNz7Oqa+LLpQr2-RyosYKphNTzA@mail.gmail.com>
 <CAOafRRdGwEZoS2wgavvntsVLAXT5KqrS6PuKa387B3DreC3JvQ@mail.gmail.com>
From: Tyler Hobbs <tyler@datastax.com>
Date: Tue, 8 Sep 2015 11:25:16 -0500
Message-ID: 
 <CAAam9stSos8bV5nmsQDwgfOuSoCqMMwyzzx3b7dijN2gvU6egg@mail.gmail.com>
Subject: Re: who does generate timestamp during the write?
To: user@cassandra.apache.org
Content-Type: multipart/alternative; boundary=001a1134cd06d56ff9051f3ed044

--001a1134cd06d56ff9051f3ed044
Content-Type: text/plain; charset=UTF-8

On Sat, Sep 5, 2015 at 8:32 AM, ibrahim El-sanosi <ibrahimsabattt@gmail.com>
wrote:

> So in this scenario, the latest data that wrote to the replicas is [K1,
> V2] which should be the correct one, but it reads [K1,V1] because of divert
> clock.
>
> Can such scenario occur?
>

Yes, it most certainly can.  There are a couple of pieces of advice for
this.  First, run NTP on all of your servers.  Second, if clock drift of a
second or so would cause problems for your data model (like your example),
change your data model.  Usually this means creating separate rows for each
version of the value (by adding a timuuid to the primary key, for example),
but in some cases lightweight transactions may also be suitable.


-- 
Tyler Hobbs
DataStax <http://datastax.com/>

--001a1134cd06d56ff9051f3ed044
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div class=3D"gmail_extra"><br><div class=3D"gmail_quote">=
On Sat, Sep 5, 2015 at 8:32 AM, ibrahim El-sanosi <span dir=3D"ltr">&lt;<a =
href=3D"mailto:ibrahimsabattt@gmail.com" target=3D"_blank">ibrahimsabattt@g=
mail.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"><p class=
=3D"MsoNormal" style=3D"margin-bottom:12pt;line-height:normal"><span style=
=3D"font-size:12pt;font-family:&quot;Times New Roman&quot;,&quot;serif&quot=
;"></span></p>

<p class=3D"MsoNormal" style=3D"margin-bottom:12pt;line-height:normal"><spa=
n style=3D"font-size:12pt;font-family:&quot;Times New Roman&quot;,&quot;ser=
if&quot;">So in this scenario, the latest
data that wrote to the replicas is [K1, V2] which should be the correct one=
,
but it reads [K1,V1] because of divert clock. <br>
<br>
</span></p>

<p class=3D"MsoNormal" style=3D"margin-bottom:12pt;line-height:normal"><spa=
n style=3D"font-size:12pt;font-family:&quot;Times New Roman&quot;,&quot;ser=
if&quot;">Can such scenario occur?</span></p></blockquote></div><br></div><=
div class=3D"gmail_extra">Yes, it most certainly can.=C2=A0 There are a cou=
ple of pieces of advice for this.=C2=A0 First, run NTP on all of your serve=
rs.=C2=A0 Second, if clock drift of a second or so would cause problems for=
 your data model (like your example), change your data model.=C2=A0 Usually=
 this means creating separate rows for each version of the value (by adding=
 a timuuid to the primary key, for example), but in some cases lightweight =
transactions may also be suitable.<br></div><div class=3D"gmail_extra"><br =
clear=3D"all"><br>-- <br><div class=3D"gmail_signature"><font color=3D"#888=
888">Tyler Hobbs<span></span><br>
<a href=3D"http://datastax.com/" target=3D"_blank">DataStax</a><br></font><=
/div>
</div></div>

--001a1134cd06d56ff9051f3ed044--