DevTools: remote debugging should work in the IPv6-only environment.

DevTools: remote debugging should work in the IPv6-only environment.

BUG=599915
TBR=sky (for trivial browser change)

Committed: https://crrev.com/ff8289e0123c7829ae54581e96e70bfdfe9882ab
Cr-Commit-Position: refs/heads/master@{#384702}

[pfeldman, 2016-04-01 18:33:45]

pfeldman@chromium.org changed reviewers:
+ mmenke@chromium.org

[mmenke, 2016-04-01 18:41:21]

https://codereview.chromium.org/1846953007/diff/1/chrome/browser/devtools/rem...
File chrome/browser/devtools/remote_debugging_server.cc (right):

https://codereview.chromium.org/1846953007/diff/1/chrome/browser/devtools/rem...
chrome/browser/devtools/remote_debugging_server.cc:53: return socket;
Where's the RemoteDebuggingClient?  Would it be better to just listen on both
IPv6 and IPv4?

[mmenke, 2016-04-01 18:42:16]

https://codereview.chromium.org/1846953007/diff/1/chrome/browser/devtools/rem...
File chrome/browser/devtools/remote_debugging_server.cc (right):

https://codereview.chromium.org/1846953007/diff/1/chrome/browser/devtools/rem...
chrome/browser/devtools/remote_debugging_server.cc:53: return socket;
On 2016/04/01 18:41:21, mmenke wrote:
> Where's the RemoteDebuggingClient?  Would it be better to just listen on both
> IPv6 and IPv4?

Also, it's possible the port could be used by another app on IPv4, but could be
free on IPv6.

[pfeldman, 2016-04-01 18:50:22]

On 2016/04/01 18:42:16, mmenke wrote:
>
https://codereview.chromium.org/1846953007/diff/1/chrome/browser/devtools/rem...
> File chrome/browser/devtools/remote_debugging_server.cc (right):
> 
>
https://codereview.chromium.org/1846953007/diff/1/chrome/browser/devtools/rem...
> chrome/browser/devtools/remote_debugging_server.cc:53: return socket;
> On 2016/04/01 18:41:21, mmenke wrote:
> > Where's the RemoteDebuggingClient?  Would it be better to just listen on
both
> > IPv6 and IPv4?
> 
> Also, it's possible the port could be used by another app on IPv4, but could
be
> free on IPv6.

[pfeldman, 2016-04-01 18:51:40]

> Also, it's possible the port could be used by another app on IPv4, but could
be
> free on IPv6.

We will end up with the port bound on ipv6 instead of failing then. That's a
good point, but I am not sure whether we want to explicitly pass the ip format
to the app.

[pfeldman, 2016-04-01 18:53:11]

> Would it be better to just listen on both IPv6 and IPv4?

Oh, I missed this one. This specific request is for the remote debugging support
in the IPv6-only environment. We haven't seen requests for the mixed
requirements accessibility. I'd rather we don't make things more complex before
it is needed.

[mmenke, 2016-04-01 18:55:28]

On 2016/04/01 18:53:11, pfeldman wrote:
> > Would it be better to just listen on both IPv6 and IPv4?
> 
> Oh, I missed this one. This specific request is for the remote debugging
support
> in the IPv6-only environment. We haven't seen requests for the mixed
> requirements accessibility. I'd rather we don't make things more complex
before
> it is needed.

Where's the client code?  Can't really evaluate the correctness of this without
it.  If you don't want to pass whether we're talking IPv4 or IPv6, seems like
you should try all ports on IPv4, and then try all on IPv6, and the client
should also try IPv4 first, and then IPv6 (To handle the opposite case -
something already listening to localhost on IPv6, when we're waiting using the
same port on IPv4)

[mmenke, 2016-04-01 19:11:18]

LGTM

[pfeldman, 2016-04-01 19:13:40]

Description was changed from

==========
DevTools: remote debugging should work in the IPv6-only environment.

BUG=599915
==========

to

==========
DevTools: remote debugging should work in the IPv6-only environment.

BUG=599915
TBR=sky (for trivial browser change)
==========

[pfeldman, 2016-04-01 19:13:40]

pfeldman@chromium.org changed reviewers:
+ sky@chromium.org

[pfeldman, 2016-04-01 19:13:49]

The CQ bit was checked by pfeldman@chromium.org

[pfeldman, 2016-04-01 19:14:28]

pfeldman@chromium.org changed reviewers:
+ samuong@chromium.org

[commit-bot: I haz the power, 2016-04-01 19:14:32]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1846953007/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1846953007/1

[commit-bot: I haz the power, 2016-04-01 22:00:00]

Description was changed from

==========
DevTools: remote debugging should work in the IPv6-only environment.

BUG=599915
TBR=sky (for trivial browser change)
==========

to

==========
DevTools: remote debugging should work in the IPv6-only environment.

BUG=599915
TBR=sky (for trivial browser change)
==========

[commit-bot: I haz the power, 2016-04-01 22:00:01]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-04-01 22:01:19]

Description was changed from

==========
DevTools: remote debugging should work in the IPv6-only environment.

BUG=599915
TBR=sky (for trivial browser change)
==========

to

==========
DevTools: remote debugging should work in the IPv6-only environment.

BUG=599915
TBR=sky (for trivial browser change)

Committed: https://crrev.com/ff8289e0123c7829ae54581e96e70bfdfe9882ab
Cr-Commit-Position: refs/heads/master@{#384702}
==========

[commit-bot: I haz the power, 2016-04-01 22:01:20]

Patchset 1 (id:??) landed as
https://crrev.com/ff8289e0123c7829ae54581e96e70bfdfe9882ab
Cr-Commit-Position: refs/heads/master@{#384702}

[samuong, 2016-04-05 21:24:28]

lgtm, i think we'll do the same thing in chromedriver, and then make it listen
on both ipv4 and ipv6 in a separate change