Looks like the whole site is down at this point.

TK

On Fri, Sep 19, 2014 at 02:43:02PM +0000, Tim Keating via Digitalmars-d wrote:
> Looks like the whole site is down at this point.
[...]

Works for me. Are you behind a firewall?


T

-- 
The richest man is not he who has the most, but he who needs the least.

On Friday, 19 September 2014 at 14:51:28 UTC, H. S. Teoh via Digitalmars-d wrote:
>
> Works for me. Are you behind a firewall?
>
> T

Weird. I am not. Unless Time Warner is pulling a fast one on me.

Ah, well. Sorry for the false alarm.

Okay, this is weird.

This is only happening for me using Google Chrome on OS X 10.9. I can hit the site with Safari and Firefox just fine.

I thought maybe I had some malicious infected extension or something (I was seeing some activity there in the network trace in the Chrome developer tools), so I disabled all my extensions. Still happening.

And the error page looks very clearly like the output from the errorOut function in server.d. Which suggests to me that I'm getting to the server and it's legitimately sending me a 500 error. But there's no attached debugging information (debug_msg appears to be blank).

Scratching my head at this point.

TK

On Saturday, 20 September 2014 at 04:11:01 UTC, Tim Keating wrote:
> Okay, this is weird.
>
> This is only happening for me using Google Chrome on OS X 10.9. I can hit the site with Safari and Firefox just fine.
>
> I thought maybe I had some malicious infected extension or something (I was seeing some activity there in the network trace in the Chrome developer tools), so I disabled all my extensions. Still happening.
>
> And the error page looks very clearly like the output from the errorOut function in server.d. Which suggests to me that I'm getting to the server and it's legitimately sending me a 500 error. But there's no attached debugging information (debug_msg appears to be blank).
>
> Scratching my head at this point.
>
> TK

Works with Chrome 37 on Linux.

Am 20.09.2014 06:11, schrieb Tim Keating:
> Okay, this is weird.
>
> This is only happening for me using Google Chrome on OS X 10.9. I can
> hit the site with Safari and Firefox just fine.
>
> I thought maybe I had some malicious infected extension or something (I
> was seeing some activity there in the network trace in the Chrome
> developer tools), so I disabled all my extensions. Still happening.
>
> And the error page looks very clearly like the output from the errorOut
> function in server.d. Which suggests to me that I'm getting to the
> server and it's legitimately sending me a 500 error. But there's no
> attached debugging information (debug_msg appears to be blank).
>
> Scratching my head at this point.
>
> TK

Works for me with Chrome 37 on OS X 10.9.2. I've enabled some more debugging output, so hopefully it will output something useful next time.

BTW, it's also theoretically possible that the site was simply down for a few seconds. The gateway proxy currently still returns a 500 error instead of 502 in that case. There have been some RangeViolation errors in the log (without call trace unfortunately), which have caused the site to get restarted.

Mystery solved. With the debug info you enabled, I started getting meaningful stack data:

Sorry, we got a 400 error (Bad Request) while processing this request. We will look into this as soon as possible.

Internal error information:

object.Exception@../vibe/source/vibe/http/server.d(1534): Expected name=value.
----------------
./vibed-org(pure @safe bool std.exception.enforce!(bool).enforce(bool, lazy const(char)[], immutable(char)[], ulong)+0x6b) [0x7d9e4f]
./vibed-org(void vibe.http.server.parseCookies(immutable(char)[], ref vibe.http.common.CookieValueMap)+0x7b) [0x98e79f]
./vibed-org(bool vibe.http.server.handleRequest(vibe.core.stream.Stream, vibe.core.net.TCPConnection, vibe.http.server.HTTPServerListener, ref vibe.http.server.HTTPServerSettings, ref bool)+0xfec) [0x98ce84]
./vibed-org(void vibe.http.server.handleHTTPConnection(vibe.core.net.TCPConnection, vibe.http.server.HTTPServerListener)+0x17e) [0x98bdb6]
./vibed-org(void vibe.http.server.listenHTTPPlain(vibe.http.server.HTTPServerSettings).doListen(vibe.http.server.HTTPServerSettings, vibe.http.server.HTTPServerListener, immutable(char)[]).__lambda4(vibe.core.net.TCPConnection)+0x2c) [0x98bac8]
./vibed-org(void vibe.core.drivers.libevent2_tcp.onConnect(int, short, void*).ClientTask.execute()+0x4d4) [0x8f678c]
./vibed-org(_D4vibe4core4core27__T16makeTaskFuncInfoTDFZvZ16makeTaskFuncInfoFNbDFZvZS4vibe4core4core12TaskFuncInfo12callDelegateFPS4vibe4core4core12TaskFuncInfoZv+0xdb) [0x8e1727]
./vibed-org(void vibe.core.core.CoreTask.run()+0x173) [0x8ddb7f]
./vibed-org(void core.thread.Fiber.run()+0x2a) [0xa530ee]
./vibed-org(fiber_entryPoint+0x61) [0xa52ff9]
[(nil)]

1534 here looks like the enforce in line 1537 in the current HEAD. That's a cookie parsing error, so possibly I got a corrupted write to local storage or something & ended up with a malformed cookie. To test, I cleared cookies for the site, and voila!

I will say that possibly the whole site shouldn't barf if you get a bad cookie, though :-)

Thanks for the help!

TK

TK