You are viewing a plain text version of this content. The canonical link for it is here.
Posted to modproxy-dev@apache.org by Graham Leggett <mi...@sharp.fm> on 2001/04/26 00:16:18 UTC
More on the Akamai weirdness
Hi all,
I've been doing some sniffing and tracing on the Akamai issue, but have
come no closer to understanding what the problem is.
The problem does in fact happen with both prefork and threaded MPMs, so
threadsafe issues don't seem to be the problem.
The problem is very weird. On a non-Akamai site, an strace shows the
process accept the connection, then send the headers, the receive the
headers, then receive the response, writing it away to the browser.
On an Akamai site, the proxy suddenly hangs on one of the network reads
of the response. It sometimes happens right after the headers have been
read, sometimes after some of the body has been read. There is no
segfault reported, and sniffing the network shows that the traffic from
Akamai suddenly stops dead in it's tracks, probably because incoming
data is not being acknowledged. Usually this happens all at once - as if
all the processes have broken at once. (This is seen with Netscape v4.7,
which does not support HTTP/1.1 keepalives and therefore fetches 4 http
objects simultaneously).
I need to get some sleep at this point - if anyone else can look at this
problem and get a fresh perspective on it it would be great...
Regards,
Graham
--
-----------------------------------------
minfrin@sharp.fm "There's a moon
over Bourbon Street
tonight..."
Re: More on the Akamai weirdness
Posted by Chuck Murcko <ch...@topsail.org>.
Graham, I seem to have observed that it's only the initial (in your
case, four) connections that hang. The others never get tried.
One interesting thing I noticed in my logs yesterday:
[Mon Apr 30 03:00:39 2001] [error] [client 172.16.86.108] client denied
by server configuration:
proxy:http://a772.g.akamai.net/7/772/51/bf8b48bb5bd500/www.apple.com/macosx/
images/action_fonts.gif, referer: http://www.apple.com/macosx/
[Mon Apr 30 03:00:42 2001] [error] [client 172.16.86.108] client denied
by server configuration:
proxy:http://a772.g.akamai.net/7/772/51/9dc9a6eb4945c8/www.apple.com/macosx/
images/action_preferences.gif, referer: http://www.apple.com/macosx/
[Mon Apr 30 03:00:47 2001] [error] [client 172.16.86.108] client denied
by server configuration:
proxy:http://a772.g.akamai.net/7/772/51/2b944e645acd67/www.apple.com/macosx/
images/action_customizefinder.gif, referer: http://www.apple.com/macosx/
[Mon Apr 30 03:00:48 2001] [error] [client 172.16.86.108] client denied
by server configuration:
proxy:http://a772.g.akamai.net/7/772/51/281b67b5119734/www.apple.com/macosx/
images/action_genie.gif, referer: http://www.apple.com/macosx/
Note that I am also trying four simultaneous connections from the
browser here, and I have a wide open proxy config (double checked that
first).
Chuck
On Saturday, April 28, 2001, at 03:05 PM, Graham Leggett wrote:
> Graham Leggett wrote:
>
>> On an Akamai site, the proxy suddenly hangs on one of the network reads
>> of the response. It sometimes happens right after the headers have been
>> read, sometimes after some of the body has been read.
>
> Actually, this has all the hallmarks of memory getting stomped on.
> Suddenly the FTP_DIR filter is showing odd corruption of memory here and
> there, and it looks like the same thing is causing it.
>
> This didn't happen until recently. When I get some backwidth I'll work
> backwards through the CVS log to try and find which commit broke things.
>
> Regards,
> Graham
> --
> -----------------------------------------
> minfrin@sharp.fm "There's a moon
> over Bourbon Street
> tonight..."
>
>
Chuck Murcko
Topsail Group
http://www.topsail.org/
Re: More on the Akamai weirdness
Posted by Graham Leggett <mi...@sharp.fm>.
Graham Leggett wrote:
> On an Akamai site, the proxy suddenly hangs on one of the network reads
> of the response. It sometimes happens right after the headers have been
> read, sometimes after some of the body has been read.
Actually, this has all the hallmarks of memory getting stomped on.
Suddenly the FTP_DIR filter is showing odd corruption of memory here and
there, and it looks like the same thing is causing it.
This didn't happen until recently. When I get some backwidth I'll work
backwards through the CVS log to try and find which commit broke things.
Regards,
Graham
--
-----------------------------------------
minfrin@sharp.fm "There's a moon
over Bourbon Street
tonight..."
Re: More on the Akamai weirdness
Posted by Chuck Murcko <ch...@topsail.org>.
On Wednesday, April 25, 2001, at 06:16 PM, Graham Leggett wrote:
> Hi all,
>
> I've been doing some sniffing and tracing on the Akamai issue, but have
> come no closer to understanding what the problem is.
Avi says they (Akamai) should just look like origin servers when an
inbound proxy hits an Akamai border server.
In the remote chance that we actually have stumbled on something broken
on the Akamai side, we'll just work with them to fix the problem.
Chuck Murcko
Topsail Group
http://www.topsail.org/