Getting [Errno 10054] when trying clone,push,pull over HTTPS

[Surya Nyayapati]

My clone actually works over HTTP. here is the output.(I will also try your
method and will let you know)

C:\projects\test>hg clone --debug --traceback http://webtest03.vis/hg/test
using http://webtest03.vis/hg/test
sending capabilities command
Keyring URL: http://webtest03.vis/hg/test
Username not specified in .hg/hgrc. Keyring will not be used.
http authorization required
realm: webtest03.vis
user: surya
password:
Manually entered password. Url: http://webtest03.vis/hg/test, user: surya,
passwd: ********
destination directory: test
query 1; heads
sending batch command
Keyring URL: http://webtest03.vis/hg/test
Cached auth data found. Url: http://webtest03.vis/hg/test, user: surya,
passwd: ********
requesting all changes
sending getbundle command
Keyring URL: http://webtest03.vis/hg/test
Cached auth data found. Url: http://webtest03.vis/hg/test, user: surya,
passwd: ********
adding changesets
changesets: 1 chunks
add changeset a5459e265cc5
changesets: 2 chunks
add changeset 467ca47361a4
changesets: 3 chunks
add changeset bf7b5cb54be9
adding manifests
manifests: 1/3 chunks (33.33%)
manifests: 2/3 chunks (66.67%)
manifests: 3/3 chunks (100.00%)
adding file changes
adding test.txt revisions
files: 1/1 chunks (100.00%)
added 3 changesets with 3 changes to 1 files
updating the branch cache
updating to branch default
resolving manifests
 overwrite False partial False
 ancestor 000000000000 local 000000000000+ remote bf7b5cb54be9
 test.txt: remote created -> g
calling hook preupdate.eol: <function preupdate at 0x0000000002DBE668>
updating: test.txt 1/1 files (100.00%)
getting test.txt
1 files updated, 0 files merged, 0 files removed, 0 files unresolved
sending listkeys command
Keyring URL: http://webtest03.vis/hg/test
Cached auth data found. Url: http://webtest03.vis/hg/test, user: surya,
passwd: ********


On Thu, Oct 6, 2011 at 4:47 PM, Matt Mackall <mpm at selenic.com> wrote:

> On Thu, 2011-10-06 at 16:26 -0400, Surya Nyayapati wrote:
> > Matt,
> >  thank you so much for getting back to me.
> > here is the output.
> >
> > c:\projects\test>hg clone --debug --traceback
> https://webtest03.vis/hg/test
> > using https://webtest03.vis/hg/test
> > sending capabilities command
> > using auth.webtest03.vis.* for authentication
> > webtest03.vis certificate successfully verified
>
> Ok, so your client is perfectly happy with the initial SSL negotiation.
> But the very first request ("capabilities") never finishes:
>
> >   File "mercurial\httprepo.pyo", line 56, in _fetchcaps
> >   File "mercurial\httprepo.pyo", line 164, in _call
> >   File "mercurial\httprepo.pyo", line 114, in _callstream
> >   File "urllib2.pyo", line 391, in open
> >   File "urllib2.pyo", line 409, in _open
> >   File "urllib2.pyo", line 369, in _call_chain
> >   File "mercurial\url.pyo", line 360, in https_open
> >   File "mercurial\keepalive.pyo", line 259, in do_open
> > URLError: <urlopen error [Errno 10054] An existing connection was
> forcibly
> > close
> > d by the remote host>
> > abort: error: An existing connection was forcibly closed by the remote
> host
>
> ..because the remote side hangs up for some unknown reason.
>
> You should check ALL the webserver logs on your server for any errors.
>
> In general, trying to go from nothing to full SSL-based hgweb server is
> a great way to get a system that's impossible to debug because you've
> skipped all the intermediate steps and thus have no known-good
> intermediate steps to fall back to and no known-good configuration.
>
> I instead recommend a process like this:
>
> - try 'hg serve', test with a client and a webbrowser
> - enable push support in hg serve, test with a client
> - install a webserver, test with a browser
> - add a trivial CGI script, test with a browser
> - install hgweb, test with a client and a browser
> - enable auth in your server, test with a client and a browser
> - enable push support, test with a client
> - enable SSL, test with a client and a browser
>
> Now, when something breaks.. you know precisely where the problem is!
>
> --
> Mathematics is the supreme nostalgia of our time.
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mercurial-scm.org/pipermail/mercurial/attachments/20111006/00476762/attachment-0002.html>