Mailing-List: contact user-help@flink.apache.org; run by ezmlm
Precedence: bulk
MIME-Version: 1.0
In-Reply-To: <CAGKOEz25xE91BUgOeDJTWQO64GmrBqOfPY4iQ+13OfNeWjX4wg@mail.gmail.com>
References: <CAFoDxaOyKT+pN7V3V3hWjA5YuS4sv9KQR_xa06KgLJVZVd7bmw@mail.gmail.com>
 <CAGKOEz25xE91BUgOeDJTWQO64GmrBqOfPY4iQ+13OfNeWjX4wg@mail.gmail.com>
From: Moiz S Jinia <moiz.jinia@gmail.com>
Date: Thu, 25 May 2017 22:20:58 +0530
Message-ID: <CAFoDxaMRs5sJdsbNzP71ajaSqePb1_76NVSXa7Z7BCnabqD0_A@mail.gmail.com>
Subject: Re: Are timers in ProcessFunction fault tolerant?
To: Eron Wright <eronwright@gmail.com>
Cc: user@flink.apache.org
Content-Type: multipart/alternative; boundary="001a11429a50a454b405505c074d"
archived-at: Thu, 25 May 2017 16:51:23 -0000

--001a11429a50a454b405505c074d
Content-Type: text/plain; charset="UTF-8"

Awesome. Thanks.

On Thu, May 25, 2017 at 10:13 PM, Eron Wright <eronwright@gmail.com> wrote:

> Yes, registered timers are stored in managed keyed state and should be
> fault-tolerant.
>
> -Eron
>
> On Thu, May 25, 2017 at 9:28 AM, Moiz S Jinia <moiz.jinia@gmail.com>
> wrote:
>
>> With a checkpointed RocksDB based state backend, can I expect the
>> registered processing timers to be fault tolerant? (along with the managed
>> keyed state).
>>
>> Example -
>> A task manager instance owns the key k1 (from a keyed stream) that has
>> registered a processing timer with a timestamp thats a day ahead in the
>> future. If this instance is killed, and the key is moved to another
>> instance, will the onTimer trigger correctly on the other machine at the
>> expected time with the same keyed state (for k1)?
>>
>> Thanks,
>> Moiz
>>
>
>

--001a11429a50a454b405505c074d
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Awesome. Thanks.</div><div class=3D"gmail_extra"><br><div =
class=3D"gmail_quote">On Thu, May 25, 2017 at 10:13 PM, Eron Wright <span d=
ir=3D"ltr">&lt;<a href=3D"mailto:eronwright@gmail.com" target=3D"_blank">er=
onwright@gmail.com</a>&gt;</span> wrote:<br><blockquote class=3D"gmail_quot=
e" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">=
<div dir=3D"ltr">Yes, registered timers are stored in managed keyed state a=
nd should be fault-tolerant.=C2=A0<span class=3D"HOEnZb"><font color=3D"#88=
8888"><br><div><br></div><div>-Eron</div></font></span></div><div class=3D"=
HOEnZb"><div class=3D"h5"><div class=3D"gmail_extra"><br><div class=3D"gmai=
l_quote">On Thu, May 25, 2017 at 9:28 AM, Moiz S Jinia <span dir=3D"ltr">&l=
t;<a href=3D"mailto:moiz.jinia@gmail.com" target=3D"_blank">moiz.jinia@gmai=
l.com</a>&gt;</span> wrote:<br><blockquote class=3D"gmail_quote" style=3D"m=
argin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir=3D"l=
tr">With a checkpointed RocksDB based state backend, can I expect the regis=
tered processing timers to be fault tolerant? (along with the managed keyed=
 state).<div><br></div><div>Example -</div><div>A task manager instance own=
s the key k1 (from a keyed stream) that has registered a processing timer w=
ith a timestamp thats a day ahead in the future. If this instance is killed=
, and the key is moved to another instance, will the onTimer trigger correc=
tly on the other machine at the expected time with the same keyed state (fo=
r k1)?</div><div><br></div><div>Thanks,</div><div>Moiz</div></div>
</blockquote></div><br></div>
</div></div></blockquote></div><br></div>

--001a11429a50a454b405505c074d--