Mailing-List: contact user-help@flink.apache.org; run by ezmlm
Precedence: bulk
MIME-Version: 1.0
In-Reply-To: <1504668939165-0.post@n4.nabble.com>
References: <CAB7Edug5tC+JhuQNm4ZvkE55foRxcwdm-gh_YcKv4OQeodFkSw@mail.gmail.com>
 <1504668939165-0.post@n4.nabble.com>
From: Fabian Hueske <fhueske@gmail.com>
Date: Fri, 8 Sep 2017 17:38:25 +0200
Message-ID: <CAAdrtT1EUSx2U=YsEbdSa2j6F9xuySp5M==TiV7f3YKA4JhOag@mail.gmail.com>
Subject: Re: Fwd: HA : My job didn't restart even if task manager restarted.
To: sunny yun <seonhee.yun@gmail.com>
Cc: user <user@flink.apache.org>, Till Rohrmann <trohrmann@apache.org>
Content-Type: multipart/alternative; boundary="94eb2c091ec0f009f40558af5f43"
archived-at: Fri, 08 Sep 2017 15:39:06 -0000

--94eb2c091ec0f009f40558af5f43
Content-Type: text/plain; charset="UTF-8"

Hi,

sorry for the late response!
I'm not familiar with the details of the failure recovery but Till (in CC)
knows the code in depth.
Maybe he can figure out what's going on.

Best, Fabian

2017-09-06 5:35 GMT+02:00 sunny yun <seonhee.yun@gmail.com>:

> I am still struggling to solve this problem.
> I have no doubt that the JOB should automatically restart after restarting
> the TASK MANAGER in YARN MODE. Is it a misunderstood?
>
> Problem seems that *JOB MANAGER still try to connect to old TASK MANAGER
> even after new TASK MANAGER container be created.*
> When I killed TM on node#2 then new TM container is created on node#3, but
> JM still tries to connect to TM on node#2 according to the log file. (It
> was
> not a log I posted before, when I found it while continuing the test.
> Normally the TM be created on the same node after killed.)
> So new TM don't know JOB info and JM show us JOB with fail status.
>
> If anyone has succeeded in the same situation(YARN + TM FAILURE), please
> just tell me.
> That will be big help to me.
>
>
>
> --
> Sent from: http://apache-flink-user-mailing-list-archive.2336050.
> n4.nabble.com/
>

--94eb2c091ec0f009f40558af5f43
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div><div><div>Hi,<br><br></div>sorry for the late respons=
e!<br></div>I&#39;m not familiar with the details of the failure recovery b=
ut Till (in CC) knows the code in depth.</div><div>Maybe he can figure out =
what&#39;s going on.<br></div><div><br></div>Best, Fabian<br></div><div cla=
ss=3D"gmail_extra"><br><div class=3D"gmail_quote">2017-09-06 5:35 GMT+02:00=
 sunny yun <span dir=3D"ltr">&lt;<a href=3D"mailto:seonhee.yun@gmail.com" t=
arget=3D"_blank">seonhee.yun@gmail.com</a>&gt;</span>:<br><blockquote class=
=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padd=
ing-left:1ex">I am still struggling to solve this problem.<br>
I have no doubt that the JOB should automatically restart after restarting<=
br>
the TASK MANAGER in YARN MODE. Is it a misunderstood?<br>
<br>
Problem seems that *JOB MANAGER still try to connect to old TASK MANAGER<br=
>
even after new TASK MANAGER container be created.*<br>
When I killed TM on node#2 then new TM container is created on node#3, but<=
br>
JM still tries to connect to TM on node#2 according to the log file. (It wa=
s<br>
not a log I posted before, when I found it while continuing the test.<br>
Normally the TM be created on the same node after killed.)<br>
So new TM don&#39;t know JOB info and JM show us JOB with fail status.<br>
<br>
If anyone has succeeded in the same situation(YARN + TM FAILURE), please<br=
>
just tell me.<br>
That will be big help to me.<br>
<br>
<br>
<br>
--<br>
Sent from: <a href=3D"http://apache-flink-user-mailing-list-archive.2336050=
.n4.nabble.com/" rel=3D"noreferrer" target=3D"_blank">http://apache-flink-u=
ser-<wbr>mailing-list-archive.2336050.<wbr>n4.nabble.com/</a><br>
</blockquote></div><br></div>

--94eb2c091ec0f009f40558af5f43--