541392 - ssltunnel should have more detailed errors

Status: NEW

(Testing :: Mochitest, enhancement)

Description

[Serge Gautherie (:sgautherie)]

[Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.9.1.8pre) Gecko/20100122 SeaMonkey/2.0.3pre] (comm-1.9.1-win32-unittest/1264165841) (W2Ksp4)
{
INFO | automation.py | SSL tunnel pid: 852
...
failed to bind socket
Shutting down...
}

(I'm not sure why I'm getting this atm, but:)

It would be very helpful if the error reported: server_addr, listen_port, the result code, etc.

Comment 1

[(not currently active) Ted Mielczarek]

See also bug 541117.

Comment 2

[Serge Gautherie (:sgautherie)]

(In reply to comment #0)
> (I'm not sure why I'm getting this atm, but:)

(Ftr, my SeaMonkey (maybe ChatZilla related?) connections were interfering...
It works again: "Server listening on port 4443 with cert pgo server certificate".)

Comment 3

[Honza Bambas (:mayhemer)]

Serge, didn't it happen that you had been running another instance of ssltunnel?

However, I'll add more detailed logging. I have added some new recently just to investigate an intermittent bug, but I plane to keep it, just clean that up and some more with it.

Comment 4

[Serge Gautherie (:sgautherie)]

(In reply to comment #3)
> Serge, didn't it happen that you had been running another instance of
> ssltunnel?

That happened with some other bug(s) in the past.
But not this time: I checked and eventually found that going offline with my SeaMonkey v1.5a "fixed" it.
(That's why the socket details would (have) make finding the "culprit" application much easier ;->)

Thanks for taking this bug :-)

Comment 5

[Serge Gautherie (:sgautherie)]

(In reply to comment #1)
> See also bug 541117.

I'm not sure which code that bug wants "fixed".
But a comment there references bug 541000 which relates to nsServerSocket.cpp which has the same need for details:
http://mxr.mozilla.org/mozilla-central/search?string=failed+to+bind&case=on&find=%5C.c

Comment 6

[Mark Hammond [:markh] [:mhammond]]

Another vote for more detailed error messages.  I just struck the same problem and it took alot of spelunking to determine that VMWare version 8 has a service related to "shared VMs" that by default grabs a couple of the same ports.  If the error message clearly said that it failed to bind a specific port (and even suggested I look for other processes which are bound to the port) I think I would have guessed the root cause of the problem much quicker and not have annoyed the guys in #ateam about failure of my mochi tests to work with SLL enabled sites :)