openmeetings-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Utkarsh Khokhar <utkarshkhok...@gmail.com>
Subject Re: Reg Openmeetings
Date Tue, 26 Nov 2013 07:04:43 GMT
Thanks for the changes Maxim.
I have a observation about a possible logical glitch with the new
configuration variable "redirect.url.for.external.users"

Eg. An OM User enters his meeting room using secureHash and also sends 2
different inviteHash for 2 different external users.

Inorder to ensure that all the 3 users can stay connected in the same room
throughout the call, one variable might not be sufficient. For instance if
I saved the last hash sent (be it secure or invite) as
"redirect.url.for.external.users" then only that url will get picked up
everytime.

Please correct me if I am wrong and missed something here.

Utkarsh


On Tue, Nov 26, 2013 at 8:31 AM, Maxim Solodovnik <solomax666@gmail.com>wrote:

> OK
> seems to be fixed https://issues.apache.org/jira/browse/OPENMEETINGS-842
> here is the patch: http://svn.apache.org/r1545517
>
> can anyone please test it?
>
> external users will be redirected to the URL set in admin->config:
> "redirect.url.for.external.users"
>
> the new config key and it's description is added to the docs
> http://openmeetings.apache.org/GeneralConfiguration.html
>
> The changes above will be available here:
>
> https://builds.apache.org/view/M-R/view/OpenMeetings/job/Openmeetings%202.x/
> (build
> #22)
>
>
>
>
> On Tue, Nov 26, 2013 at 4:38 AM, Rene' Rosenbaum <rene@meecoda.de> wrote:
>
> > Dear Maxime,
> > may you please let me know which files you modified? We are using a
> highly
> > customized OM based on v2.1 and thus, have to migrate your modifications
> > for now. We'll update quickly when 2.2 is stable enough for production.
> > We highly appreciate all your efforts to solve this problem,
> > Rene'
> >
> >
> > ~~~~~~~
> > *Dr.-Ing. Rene' Rosenbaum
> > meeCoda^IT  *  - Consulting and Services
> >  ~: Neue Reihe 15, 18182 Goorstorf, Germany
> >  #: ++49-(0)-1781408041
> >  @:info@meecoda.de  <mailto:info@meecoda.de>
> > //:www.meecoda.de  <http://www.meecoda.de>
> > ~~~~~~~~~~~~~~ +++ ~~~~~~~~~~~~~~~~
> >
> > On 11/25/2013 6:21 PM, Maxim Solodovnik wrote:
> >
> >> After fix I need you to test 2.2 build with this feature added.
> >> Will write into this thread.
> >>
> >>
> >> On Mon, Nov 25, 2013 at 10:46 PM, Rene' Rosenbaum <rene@meecoda.de>
> >> wrote:
> >>
> >>  Dear Maxim,
> >>> that would be great! I think a solution is of value for many adopters
> of
> >>> OM embedding the software in their websites. Please let me know when
> you
> >>> need any support or further information ...
> >>> Cheers,
> >>> Rene'
> >>> PS: We can sometimes reproduce the problem when suddenly switching off
> >>> the
> >>> network (plugging off). Works often with local installation, but less
> >>> with
> >>> second server machine.
> >>>
> >>>
> >>> ~~~~~~~
> >>> *Dr.-Ing. Rene' Rosenbaum
> >>> meeCoda^IT  *  - Consulting and Services
> >>>   ~: Neue Reihe 15, 18182 Goorstorf, Germany
> >>>   #: ++49-(0)-1781408041
> >>>   @:info@meecoda.de  <mailto:info@meecoda.de>
> >>> //:www.meecoda.de  <http://www.meecoda.de>
> >>> ~~~~~~~~~~~~~~ +++ ~~~~~~~~~~~~~~~~
> >>>
> >>> On 11/25/2013 4:24 PM, Maxim Solodovnik wrote:
> >>>
> >>>  I'll try to reproduce add code for this
> >>>>
> >>>>
> >>>> On Mon, Nov 25, 2013 at 10:19 PM, Rene' Rosenbaum <rene@meecoda.de>
> >>>> wrote:
> >>>>
> >>>>   Hi all,
> >>>>
> >>>>> first of all, thanks for all the good work of the team so far. We
are
> >>>>> using OM as a conferencing solution, part of a larger website. We
are
> >>>>> facing similar issues as described before: from time to time, a
> >>>>> conferencing session crashes*and the user is then referred to the
> start
> >>>>> page/dashboard of the OM backend*. As we are not working with the
> >>>>> backend
> >>>>> at all (all user management is done via the website), we have a
huge
> >>>>> problem here. We adapted the URL in the "onerror"-handler (in
> >>>>> hibRtmpConnection.lzx), but this didn't do the trick: the stated
URL
> is
> >>>>> not
> >>>>> called.
> >>>>>
> >>>>> ---------------------
> >>>>> <handler name="onerror" >
> >>>>>           <![CDATA[
> >>>>>                       ...
> >>>>> canvas.thishib.loaderVar.error.setAttribute('text',this.status);
> >>>>> canvas.setAttribute('loadingmessage','connection failed');
> >>>>>                       new lz.labelerrorPopup(canvas,{
> >>>>> errorlabelid:556});
> >>>>>                   }
> >>>>> *canvas.thishib.loaderVar._src.setAttribute('text','http:
> >>>>> //www.google.de/');*
> >>>>> //canvas.thishib.loaderVar._src.setAttribute('text',src);
> >>>>>               }
> >>>>>           ]]>
> >>>>>       </handler>
> >>>>> ---------------------
> >>>>>
> >>>>> Do you guys have an idea what we can do here to refer the user to
an
> >>>>> arbitrary URL and not the dashboard?
> >>>>>
> >>>>> Setup details:
> >>>>> - OM v2.1
> >>>>> - just using the conferencing part embedded via SOAP/REST into an
> >>>>> iframe
> >>>>> - complete setup of a conference via SOAP/REST (users, documents,
> etc.)
> >>>>>
> >>>>> Thanks for all your help,
> >>>>> Rene'
> >>>>> --
> >>>>>
> >>>>> ~~~~~~~
> >>>>> *Dr.-Ing. Rene' Rosenbaum
> >>>>> meeCoda^IT  *  - Consulting and Services
> >>>>>    ~: Neue Reihe 15, 18182 Goorstorf, Germany
> >>>>>    #: ++49-(0)-1781408041
> >>>>>    @:info@meecoda.de  <mailto:info@meecoda.de>
> >>>>> //:www.meecoda.de  <http://www.meecoda.de>
> >>>>> ~~~~~~~~~~~~~~ +++ ~~~~~~~~~~~~~~~~
> >>>>>
> >>>>> On 11/25/2013 10:02 AM, Maxim Solodovnik wrote:
> >>>>>
> >>>>>   I believe this will be resolved in 3.1.
> >>>>>
> >>>>>>
> >>>>>> On Mon, Nov 25, 2013 at 2:37 PM, Utkarsh Khokhar
> >>>>>> <utkarshkhokhar@gmail.com>wrote:
> >>>>>>
> >>>>>>    I am dealing with the exact same problem at my end.
> >>>>>>
> >>>>>>  Anyone who joins a room directly through secureHash or inviteHash
> >>>>>>> gets
> >>>>>>> disconnected at times, due to network issue, and taken back
to
> >>>>>>> dashboard.
> >>>>>>> I believe this is quite a fundamental problem. I am yet
to dive
> into
> >>>>>>> this
> >>>>>>> completely but it would be great to learn from others on
this forum
> >>>>>>> who
> >>>>>>> might have a prior insight on this.
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>> On Mon, Nov 25, 2013 at 12:51 PM, Maxim Solodovnik <
> >>>>>>> solomax666@gmail.com
> >>>>>>>
> >>>>>>>   wrote:
> >>>>>>>
> >>>>>>>> If I'm not mistaken, Sebastian told it is impossible
to reconnect
> >>>>>>>> user
> >>>>>>>> to
> >>>>>>>> the same scope ...
> >>>>>>>>
> >>>>>>>>
> >>>>>>>> On Mon, Nov 25, 2013 at 1:58 PM, sanjeev kumar <
> >>>>>>>> sanjeev1048@gmail.com
> >>>>>>>>
> >>>>>>>>   wrote:
> >>>>>>>>
> >>>>>>>>> Dear Sir
> >>>>>>>>>         I want add functionality to reconnect user
in the same
> >>>>>>>>> scope
> >>>>>>>>> while
> >>>>>>>>> his NetConnection lost due to NetConnection.connect.NetworkChange
> >>>>>>>>> or
> >>>>>>>>>
> >>>>>>>>>   Closed.
> >>>>>>>>>
> >>>>>>>>          when any event generates other than Success
it is going
> to
> >>>>>>>>
> >>>>>>>>> onerror
> >>>>>>>>> (in hibRtmpConnection.lzx ) handler there it is
disconnecting
> from
> >>>>>>>>> the
> >>>>>>>>> current scope and connecting to default scope.
> >>>>>>>>>        so how i can reconnect to same scope again.....
> >>>>>>>>> what i am doing , from onerror handler it self i
am calling
> >>>>>>>>>
> >>>>>>>>>   this.connect()
> >>>>>>>>>
> >>>>>>>>   but in the userlist multiple entries are coming for
same user ,
> >>>>>>>> so i
> >>>>>>>>
> >>>>>>>>>   want
> >>>>>>>>>
> >>>>>>>> to know correct way of reconnecting in to same scope
> automatically.
> >>>>>>>>
> >>>>>>>>  please help me.....
> >>>>>>>>>
> >>>>>>>>> Thanks & Regards
> >>>>>>>>> Sanjeev Kumar
> >>>>>>>>>
> >>>>>>>>>
> >>>>>>>>>   --
> >>>>>>>>>
> >>>>>>>> WBR
> >>>>>>>> Maxim aka solomax
> >>>>>>>>
> >>>>>>>>
> >>>>>>>>
> >>>>>>>>
> >>
> >
>
>
> --
> WBR
> Maxim aka solomax
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message