Tested. The leak is gone.<div><br></div><div><a href="http://www.varnish-cache.org">www.varnish-cache.org</a> updated, recompiled and restarted. I've added some ESI processing to the mix for good measure. :-)</div><div>
<br></div><div>Per.</div><div><br><div class="gmail_quote">On Sat, Mar 12, 2011 at 3:47 PM, Poul-Henning Kamp <span dir="ltr"><<a href="mailto:phk@phk.freebsd.dk">phk@phk.freebsd.dk</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
In message <AANLkTiknH8QZ7ApZFObmthsQrHp=<a href="mailto:bLJucKVVi31%2Bb7ea@mail.gmail.com">bLJucKVVi31+b7ea@mail.gmail.com</a>>, Per<br>
Buer writes:<br>
<br>
I think I've fixed this, a stupid oversight on my part.<br>
<br>
Please test again, when possible.<br>
<br>
Poul-Henning<br>
<br>
>--0015174c3d4addc4d1049e447a1e<br>
>Content-Type: text/plain; charset=ISO-8859-1<br>
<div><div></div><div class="h5">><br>
>Hi.<br>
><br>
>I've been running trunk on <a href="http://varnish-cache.org" target="_blank">varnish-cache.org</a> since yesterday (14:00) and<br>
>there a leak there. Turning off http_gzip_support seems to eliminate the<br>
>leak.<br>
><br>
>Reproducing the leak is quite simple. Just hammer varnish with a client that<br>
>doesn't send accept-encoding: gzip.<br>
><br>
><br>
>Per.<br>
><br>
>On Sat, Mar 12, 2011 at 12:45 AM, Dmitry Panov <<a href="mailto:dmitry.panov@yahoo.co.uk">dmitry.panov@yahoo.co.uk</a>>wrote:<br>
><br>
>> Hi Poul,<br>
>><br>
>> I think I found the reason. I tried the same test but with<br>
>> http_gzip_support=off and it worked fine. tsung does not support gzip so<br>
>> varnish had to gunzip objects for it and this is where the leak appears to<br>
>> be. Hope it helps.<br>
>><br>
>><br>
>><br>
>> On 11/03/2011 14:06, Poul-Henning Kamp wrote:<br>
>><br>
>>> In message<<a href="mailto:4D7A1DC5.7030405@yahoo.co.uk">4D7A1DC5.7030405@yahoo.co.uk</a>>, Dmitry Panov writes:<br>
>>><br>
>>> Both. The patch doesn't make any difference.<br>
>>>><br>
>>> Ok, that makes it my department :-)<br>
>>><br>
>>> Any indication what the leak depends on ?<br>
>>><br>
>>> number of requests ?<br>
>>><br>
>>> size of requests ?<br>
>>><br>
>>> hit rate ?<br>
>>><br>
>>><br>
>><br>
>> --<br>
>> Dmitry Panov<br>
>><br>
>><br>
>><br>
>> _______________________________________________<br>
>> varnish-dev mailing list<br>
>> <a href="mailto:varnish-dev@varnish-cache.org">varnish-dev@varnish-cache.org</a><br>
>> <a href="http://www.varnish-cache.org/lists/mailman/listinfo/varnish-dev" target="_blank">http://www.varnish-cache.org/lists/mailman/listinfo/varnish-dev</a><br>
>><br>
><br>
><br>
><br>
>--<br>
>Per Buer, Varnish Software<br>
>Phone: +47 21 98 92 61 / Mobile: +47 958 39 117 / Skype: per.buer<br>
>Varnish makes websites fly!<br>
>Want to learn more about Varnish?<br>
><a href="http://www.varnish-software.com/whitepapers" target="_blank">http://www.varnish-software.com/whitepapers</a><br>
><br>
</div></div>>--0015174c3d4addc4d1049e447a1e<br>
>Content-Type: text/html; charset=ISO-8859-1<br>
>Content-Transfer-Encoding: quoted-printable<br>
><br>
>Hi.<div><br></div><div>I've been running trunk on <a href=3D"<a href="http://var" target="_blank">http://var</a>=<br>
><a href="http://nish-cache.org" target="_blank">nish-cache.org</a>" target=3D"_blank"><a href="http://varnish-cache.org" target="_blank">varnish-cache.org</a></a> since yesterday (14=<br>
>:00) and there a leak there. Turning off http_gzip_support seems to elimina=<br>
>te the leak.</div><br>
><br>
><div><br></div><div>Reproducing the leak is quite simple. Just hammer varni=<br>
>sh with a client that doesn't send accept-encoding: gzip.</div><div><br=<br>
>></div><div><br></div><div>Per.<br><br><div class=3D"gmail_quote">On Sat, M=<br>
>ar 12, 2011 at 12:45 AM, Dmitry Panov <span dir=3D"ltr"><<a href=3D"mail=<br>
><a href="mailto:to%3Admitry.panov@yahoo.co.uk">to:dmitry.panov@yahoo.co.uk</a>" target=3D"_blank"><a href="mailto:dmitry.panov@yahoo.co.uk">dmitry.panov@yahoo.co.uk</a></a>=<br>
>></span> wrote:<br><br>
><br>
><br>
><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=<br>
>x #ccc solid;padding-left:1ex">Hi Poul,<br><br>
><br><br>
>I think I found the reason. I tried the same test but with http_gzip_suppor=<br>
>t=3Doff and it worked fine. tsung does not support gzip so varnish had to g=<br>
>unzip objects for it and this is where the leak appears to be. Hope it help=<br>
>s.<div><br>
><br>
><br>
><br><br>
><br><br>
><br><br>
>On 11/03/2011 14:06, Poul-Henning Kamp wrote:<br><br>
><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=<br>
>x #ccc solid;padding-left:1ex"><br>
>In message<<a href=3D"mailto:<a href="mailto:4D7A1DC5.7030405@yahoo.co.uk">4D7A1DC5.7030405@yahoo.co.uk</a>" target=3D"_bl=<br>
>ank"><a href="mailto:4D7A1DC5.7030405@yahoo.co.uk">4D7A1DC5.7030405@yahoo.co.uk</a></a>>, Dmitry Panov writes:<br><br>
><br><br>
><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=<br>
>x #ccc solid;padding-left:1ex"><br>
>Both. The patch doesn't make any difference.<br><br>
></blockquote><br>
>Ok, that makes it my department :-)<br><br>
><br><br>
>Any indication what the leak depends on ?<br><br>
><br><br>
>number of requests ?<br><br>
><br><br>
>size of requests ?<br><br>
><br><br>
>hit rate ?<br><br>
><br><br>
></blockquote><br>
><br><br>
><br><br>
>-- <br></div><font color=3D"#888888"><br>
>Dmitry Panov</font><div><div></div><div><br><br>
><br><br>
><br><br>
>_______________________________________________<br><br>
>varnish-dev mailing list<br><br>
><a href=3D"mailto:<a href="mailto:varnish-dev@varnish-cache.org">varnish-dev@varnish-cache.org</a>" target=3D"_blank">varnish-=<br>
><a href="mailto:dev@varnish-cache.org">dev@varnish-cache.org</a></a><br><br>
><a href=3D"<a href="http://www.varnish-cache.org/lists/mailman/listinfo/varnish-dev" target="_blank">http://www.varnish-cache.org/lists/mailman/listinfo/varnish-dev</a>"=<br>
> target=3D"_blank"><a href="http://www.varnish-cache.org/lists/mailman/listinfo/varn=" target="_blank">http://www.varnish-cache.org/lists/mailman/listinfo/varn=</a><br>
>ish-dev</a><br><br>
></div></div></blockquote></div><br><br clear=3D"all"><br>-- <br>Per Buer,=<br>
>=A0Varnish Software<br>Phone: +47 21 98 92 61 / Mobile: +47 958 39 117 / Sk=<br>
>ype: per.buer<br>Varnish makes websites fly!<br>Want to learn more about Va=<br>
>rnish? <a href=3D"<a href="http://www.varnish-software.com/whitepapers" target="_blank">http://www.varnish-software.com/whitepapers</a>" target=3D"_b=<br>
>lank"><a href="http://www.varnish-software.com/whitepapers" target="_blank">http://www.varnish-software.com/whitepapers</a></a><br><br>
><br>
><br>
><br><br>
></div><br>
><br>
>--0015174c3d4addc4d1049e447a1e--<br>
<div><div></div><div class="h5">><br>
<br>
--<br>
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20<br>
phk@FreeBSD.ORG | TCP/IP since RFC 956<br>
FreeBSD committer | BSD since 4.3-tahoe<br>
Never attribute to malice what can adequately be explained by incompetence.<br>
</div></div></blockquote></div><br><br clear="all"><br>-- <br>Per Buer, Varnish Software<br>Phone: +47 21 98 92 61 / Mobile: +47 958 39 117 / Skype: per.buer<br>Varnish makes websites fly!<br>Want to learn more about Varnish? <a href="http://www.varnish-software.com/whitepapers" target="_blank">http://www.varnish-software.com/whitepapers</a><br>
<br>
</div>