Mailing-List: contact user-help@accumulo.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@accumulo.apache.org
Received-SPF: pass (athena.apache.org: domain of flawaetz@gmail.com designates
 209.85.192.48 as permitted sender)
MIME-Version: 1.0
In-Reply-To: <5319FECF.1010605@gmail.com>
References: 
 <CAPnhrds2J03XQysM5aU9aqa66-BqYo0N0uRxGaJs68bVigDxbQ@mail.gmail.com>
 <5319FECF.1010605@gmail.com>
From: Frans Lawaetz <flawaetz@gmail.com>
Date: Mon, 10 Mar 2014 11:03:04 -0400
Message-ID: 
 <CAH=haiVACUKCXWjzbYzXz682_wrUwLV-L-zZhKsKUWufZSh=Og@mail.gmail.com>
Subject: Re: Is it safe / advisable to increase Zookeeper timeout?
To: user@accumulo.apache.org
Content-Type: multipart/alternative; boundary=001a11c2c974c0cbd104f441e76a

--001a11c2c974c0cbd104f441e76a
Content-Type: text/plain; charset=ISO-8859-1

On Fri, Mar 7, 2014 at 12:15 PM, Josh Elser <josh.elser@gmail.com> wrote:

> On 3/7/14, 12:01 PM, Terry P. wrote:
>
>> Greetings folks,
>> It seems network woes will never go away for this Accumulo 1.4.2 project
>> :-(
>>
>> They rebooted one of the two "redundant switches" last night, but of
>> course zero redundancy actually took place and the Master lost his
>> zookeeper lock as did one of the Datanodes after 60 seconds and shut
>> itself down.
>>
>
> By datanode you mean tserver? Hadoop datanodes don't communicate with
> ZooKeeper.
>
>
>  The 60 second period is odd, because I see that
>> instance.zookeeper.timeout is actually set to 30s, but I do recall that
>> often by default zookeeper clients retry 2 times before bailing so maybe
>> that's why.
>>
>
> It won't always be 30s before it's seen; I've seen it much quicker too.
> I'm not sure about the retries off the top of my head.


Most likely you were seeing the effects of ACCUMULO-1572 in which a
ZooKeeper disconnect causes Accumulo failure before the expiration of the
session.  Fixed in 1.5.1 and to-be-released 1.4.5.  If you think you're
seeing something else it would be good to hear about it.

--001a11c2c974c0cbd104f441e76a
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><br><div class=3D"gmail_extra"><br><br><div class=3D"gmail=
_quote">On Fri, Mar 7, 2014 at 12:15 PM, Josh Elser <span dir=3D"ltr">&lt;<=
a href=3D"mailto:josh.elser@gmail.com" target=3D"_blank">josh.elser@gmail.c=
om</a>&gt;</span> wrote:<br>


<blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-=
left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;p=
adding-left:1ex"><div>On 3/7/14, 12:01 PM, Terry P. wrote:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-=
left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;p=
adding-left:1ex">
Greetings folks,<br>
It seems network woes will never go away for this Accumulo 1.4.2 project :-=
(<br>
<br>
They rebooted one of the two &quot;redundant switches&quot; last night, but=
 of<br>
course zero redundancy actually took place and the Master lost his<br>
zookeeper lock as did one of the Datanodes after 60 seconds and shut<br>
itself down.<br>
</blockquote>
<br></div>
By datanode you mean tserver? Hadoop datanodes don&#39;t communicate with Z=
ooKeeper.<div><br>
<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-=
left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;p=
adding-left:1ex">
The 60 second period is odd, because I see that<br>
instance.zookeeper.timeout is actually set to 30s, but I do recall that<br>
often by default zookeeper clients retry 2 times before bailing so maybe<br=
>
that&#39;s why.<br>
</blockquote>
<br></div>
It won&#39;t always be 30s before it&#39;s seen; I&#39;ve seen it much quic=
ker too. I&#39;m not sure about the retries off the top of my head.</blockq=
uote><div><br></div><div>Most likely you were seeing the effects of ACCUMUL=
O-1572 in which a ZooKeeper disconnect causes Accumulo failure before the e=
xpiration of the session. =A0Fixed in 1.5.1 and to-be-released 1.4.5. =A0If=
 you think you&#39;re seeing something else it would be good to hear about =
it.</div>

</div>
</div></div>

--001a11c2c974c0cbd104f441e76a--