W3C home > Mailing lists > Public > www-jigsaw@w3.org > November to December 1997

Thread overloading

From: Mark Friedman <mark@intraspect.com>
Date: Mon, 08 Dec 1997 11:29:45 -0800
Message-ID: <348C4AA9.61B00C41@intraspect.com>
To: www-jigsaw@w3.org, jigsaw@w3.org
I'm having a terrible problem with running out of client threads, even
after all HTTP connections have been closed by any browsers. I
previously sent the two attached messages but have received no response.
I could really use some help, please.

Thanks in advace.

-Mark

Message-ID: <34732B10.D1C08894@intraspect.com>
Date: Wed, 19 Nov 1997 10:08:16 -0800
From: Mark Friedman <mark@intraspect.com>
Reply-To: mark@intraspect.com
Organization: Intraspect
X-Mailer: Mozilla 4.03 [en] (WinNT; I)
MIME-Version: 1.0
To: Jigsaw Mailinglist <www-jigsaw@w3.org>
CC: jigsaw@w3.org
Subject: Re: Some Problems!
References: <01bcf4ef$ff56a200$1cea1b81@kopernikus.iaik.tu-graz.ac.at>
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable

I also have a problem with refused connections due to thread overloading.=
 In
my case the server regularly grinds to a halt whereby it never accepts an=
y
more connections and all I can do is kill it and restart it.

When I last described this problem the answer from the Jigsaw folks was t=
hat
there was a problemin the alpha-5 Jigsaw that they knew about that might =
cause
this problem and that that problem was fixed in beta-1. Unfortunately it
doesn't seem to have fixed this problem. Can anybody help me? What would =
make
Jigsaw connection threads stick around forever?

Thanks in advance.

-Mark

Wolfgang Platzer wrote:

> 2) The errlog also contains a lot of messages like:
>
> [http-server] kopernikus.iaik.tu-graz.ac.at/129.27.234.28 refused
> (overloaded).
>
> But when I point my browser to
>
> http://host/Admin/Statistics I always get:
>
> Thread counts:
>
> free -1
>  idle -1
>  total -1
>
> Current load:-1
>
> and http://host/Admin/Threads  results in:
>
>      socket-clients:1D R
>      socket-clients:2D R
>      socket-clients:3D R
>      socket-clients:4D R
>      socket-clients:5D R
>      socket-clients:6D R
>      socket-clients:8D R
>      socket-clients:9D R
>      socket-clients:11D R
>      socket-clients:12D R
>      socket-clients:13D R
>      socket-clients:14D R
>      socket-clients:15D R
>      socket-clients:16D R
>      socket-clients:17D R
>      socket-clients:18D R
>      socket-clients:19D R
>      socket-clients:20D R
>      socket-clients:21D R
>      socket-clients:22D R
>
> How can I see how many socket clients are in use at this time and how m=
any
> socket clients are free and wait for a new connection?
>
> Wolfgang Platzer
>
> --
> Dipl.-Ing. Wolfgang Platzer
> Technische Universitaet Graz - University of Technology Graz
> Institut f=FCr Angewandte Informationsverarbeitung
> und Kommunikationstechnologien
> Klosterwiesgasse 32/I, A-8010 Graz,
> Tel: ++43 316 873-5527,Fax: ++43 316 873-5520
> URL <http://www.iaik.tu-graz.ac.at/index.html>




Message-ID: <347487D8.3ADFC285@intraspect.com>
Date: Thu, 20 Nov 1997 10:56:24 -0800
From: Mark Friedman <mark@intraspect.com>
Reply-To: mark@intraspect.com
Organization: Intraspect
X-Mailer: Mozilla 4.03 [en] (WinNT; I)
MIME-Version: 1.0
To: Jigsaw Mailinglist <www-jigsaw@w3.org>, jigsaw@w3.org
Subject: Re: Some Problems!
References: <01bcf4ef$ff56a200$1cea1b81@kopernikus.iaik.tu-graz.ac.at> <34732B10.D1C08894@intraspect.com>
Content-Type: multipart/alternative; boundary="------------1F3B2FE02A578794C14523D7"


--------------1F3B2FE02A578794C14523D7
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

I've finally been able to catch with jdb my problem with running out of
client threads. I notice that all the socket-client threads are in the
following state:

  [1] java.net.SocketInputStream.read (SocketInputStream:92)
  [2] java.io.BufferedInputStream.fill (BufferedInputStream:135)
  [3] java.io.BufferedInputStream.read (BufferedInputStream:162)
  [4] w3c.www.mime.MimeParser.parse (pc 16)
  [5] w3c.jigsaw.http.Client.getNextRequest (pc 6)
  [6] w3c.jigsaw.http.Client.startConnection (pc 66)
  [7] w3c.jigsaw.http.socket.SocketClient.run (pc 86)
  [8] w3c.util.CachedThread.run (pc 10)

and I don't know if this also matters but there are 5 active-stream
threads which are all in the following state:

  [1] java.lang.Object.wait (Object:315)
  [2] w3c.util.CachedThread.waitForRunner (pc 37)
  [3] w3c.util.CachedThread.run (pc 1)

I would really appreciate any and all help. Thanks.

-Mark

--------------1F3B2FE02A578794C14523D7
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<HTML>
I've finally been able to catch with jdb my problem with running out of
client threads. I notice that all the socket-client threads are in the
following state:

<P><TT>&nbsp; [1] java.net.SocketInputStream.read (SocketInputStream:92)</TT>
<BR><TT>&nbsp; [2] java.io.BufferedInputStream.fill (BufferedInputStream:135)</TT>
<BR><TT>&nbsp; [3] java.io.BufferedInputStream.read (BufferedInputStream:162)</TT>
<BR><TT>&nbsp; [4] w3c.www.mime.MimeParser.parse (pc 16)</TT>
<BR><TT>&nbsp; [5] w3c.jigsaw.http.Client.getNextRequest (pc 6)</TT>
<BR><TT>&nbsp; [6] w3c.jigsaw.http.Client.startConnection (pc 66)</TT>
<BR><TT>&nbsp; [7] w3c.jigsaw.http.socket.SocketClient.run (pc 86)</TT>
<BR><TT>&nbsp; [8] w3c.util.CachedThread.run (pc 10)</TT>

<P>and I don't know if this also matters but there are 5 active-stream
threads which are all in the following state:

<P><TT>&nbsp; [1] java.lang.Object.wait (Object:315)</TT>
<BR><TT>&nbsp; [2] w3c.util.CachedThread.waitForRunner (pc 37)</TT>
<BR><TT>&nbsp; [3] w3c.util.CachedThread.run (pc 1)</TT>

<P>I would really appreciate any and all help. Thanks.

<P>-Mark</HTML>

--------------1F3B2FE02A578794C14523D7--
Received on Monday, 8 December 1997 14:36:28 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 9 April 2012 12:13:27 GMT