W3C home > Mailing lists > Public > public-webapps-testsuite@w3.org > March 2013

Troubleshooting w3c-test.org WebSocket timeout errors [was: What's the plan for reviewing and approving Opera's WebSocket tests?]

From: Michael[tm] Smith <mike@w3.org>
Date: Tue, 5 Mar 2013 14:11:39 +0900
To: Ken Reed <kenre@microsoft.com>, James Graham <jgraham@opera.com>
Cc: "public-webapps-testsuite@w3.org" <public-webapps-testsuite@w3.org>
Message-ID: <20130305051138.GG16125@sideshowbarker>
"Michael[tm] Smith" <mike@w3.org>, 2013-03-05 11:54 +0900:

> So I'm seeing an error in the logs that I suspect must be the cause of the
> test failures:
> 
>   "mod_pywebsocket: web_socket_transfer_data raised exception for /ws/echo: Connection read error"
> 
> The full error message is this attached.
> 
> I'm trying to troubleshoot it now. I'll let you all know if I make any
> progress.

Ken, I do see that error has been logged a number of times from an address
that appears to be a Microsoft address, so it assume must be yours. 

However, I'm also seeing that it's been logged many times for requests that
I'm pretty sure are coming from James's machine. So if James is not seeing
any timeouts or other problems on his side, then that'd seem to indicate
the error is not the cause of the timeouts you're seeing.

Myself, I can't consistently reproduce the error. It occurs sometimes when
I run a test, but only rarely. And it seems unrelated to whether the test
passes or times out. That is, my browser will show that the test has
passed, but the log entry for that same request will show that error.

So I'm back at the point where I don't really know what could be causing
the timeout errors on the client side.

If others on the list who are in a Microsoft IP range could try some of the
tests in http://w3c-test.org/webapps/WebSockets/tests/submissions/Opera/
and let me know if you get timeouts, I would very much appreciate it. E.g.,
http://w3c-test.org/webapps/WebSockets/tests/submissions/Opera/binary/001.html

I ask because it's possible that Ken has been running into some
rate-limiting set up in the W3C MIT network, and if so it would probably
also affect others in the same IP range.

  --Mike

-- 
Michael[tm] Smith http://people.w3.org/mike
Received on Tuesday, 5 March 2013 05:11:52 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 22 April 2014 14:15:59 UTC