You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@felix.apache.org by Felix Meschberger <fm...@gmail.com> on 2008/07/01 08:23:10 UTC

Re: telnetd discussion

Hi Dieter,

Dieter Wimberger schrieb:
> Felix:
> 
>> As I understand you are willing to donate the simple telnet code 
>> described in [1] to the Apache Felix project. To get around the 
>> licensing cliffs, we need to follow a few steps:
> 
> Right.
> 
>>  * First I suggest you create a JIRA and attach the source
>>    code (including the correct Apache License file header)
> 
> Done.
> 
> https://issues.apache.org/jira/browse/FELIX-615
> 
> I have added the headers and the POM, but it doesn't seem to build a 
> bundle so far.
> 
> Please note that this code is my copyright (corrected in the POM), and 
> never had anything to do with the VFI (I am not contracted by the VFI 
> nor bound by any contract that all code I write is "theirs").
> 

Thanks. I will be looking into this.

>>  * Second, your CLA has to be on file with the ASF (see [2])
> 
> I have sent the signed and scanned ICLA to the Apache Secretary address. 
> I suppose that this means it should be on file soon.
> Probably good, because I have already actually already contributed to 
> other Apache projects before.

ok.

> 
>>  * Third, we would have to file an Incubator IP Clearance for
>>    this code (don't worry, we at Apache Felix would have to
>>    take care of this), see also [3] FYI
> 
> Ok. I don't worry (have seen [3]).
> 
>> I think for the simple remote shell bundle, this should be enough.
> 
> The ball is on your side ;)

Taking it up and following up on it.

> 
>> If you decide to donate the full telnetd bundle to Apache Felix, we 
>> would also need a software grant (see [4]). Otherwise a JIRA is required
>> to provde the code and the IP Clearance is required for tracking the 
>> process.
> 
> The problem is the following:
> I am still developing telnetd-osgi, and we are actively using it. So the 
> question I always tried to find an answer for, is how we would proceed 
> at the practical end.
> 
> If you wish to make a branch and then go from there alone, that's ok 
> with me. I contribute the current state of the code with a software 
> grant under Apache 2 license and you go with it (probably there are some 
> details to be discussed, like ensuring that there is no problematic code 
> in the bundle; thinking for example about the SSH key generation class 
> that contains code parts under a BSD license, which may need replacement 
> or rewrite).
> 
> If we are looking at going from a single project that will be Felix in 
> the future, then there are additional things that need to be discussed, 
> because I have an agenda for the development and you might have another 
> one :) This is what I was trying to find out actually.

First off, IMHO forking your code into Apache Felix and keep two 
(almost) identical projects with their own lifecycle is not practical. 
Having said this, I think, that if you contribute your code, Apache 
Felix would be very wise to take your own agenda into account.

And, we also do not want to be a code sink. That is if some code is 
contributed and the Apache Felix PMC decides to accept the contribution, 
the PMC is also considering the future of the code.

BTW, a pure BSD license is generally not a problem. But you are right, 
we would have to look into the details.

Regards
Felix

Re: telnetd discussion

Posted by Didier Donsez <di...@imag.fr>.
Richard S. Hall wrote:

> Of course there are alternatives, but that is not really at issue here.
>
> We want alternatives, so the question to answer is: Is a full-featured 
> telnet bundle a good alternative?
>
> I say, yes.
>
I say yes too
Didier

> -> richard
>
> Didier Donsez wrote:
>
>> Hi all
>>
>> An alternative to telnet shell or ssh shell could be JMX over 
>> "everything".
>> I have commit a Shell mbean in my sandbox
>> http://svn.apache.org/repos/asf/felix/sandbox/donsez/mosgi.util/
>> http://svn.apache.org/repos/asf/felix/sandbox/donsez/mbean.shell/
>> http://svn.apache.org/repos/asf/felix/sandbox/donsez/mbean.shell.felix/
>>
>> I have also one mbean for the KF shell and one for the Equinox shell.
>>
>> I'm currently refactoring the client side to remove the dependencies 
>> with the JConsole (GPLv2 Classpath Exception).
>> Here is a snapshot of the current JConsole client (in the VisualVM).
>> http://www-adele.imag.fr/users/Didier.Donsez/dev/osgi/jconsole.osgi/jconsole.shell.png 
>>
>>
>> Best regards
>>
>> Didier
>>
>>
>>
>> Felix Meschberger wrote:
>>
>>> Hi Dieter,
>>>
>>> Dieter Wimberger schrieb:
>>>
>>>> If it is for a simple access to the Felix Shell Service that uses:
>>>> shs.executeCommand(String line,PrintStream out,PrintStream out);
>>>> then I don't see much sense to move this to Apache Felix, because 
>>>> the simple remote shell does the job effectively and is much more 
>>>> efficient. I think even if SSH is the argument we may simply try to 
>>>> make a "SSH version" (using jaramiko).
>>>
>>>
>>>
>>> I agree completely.
>>>
>>>>
>>>> So let's collect good reasons, because these would also be the 
>>>> foundation to argue an inclusion with/in the Felix PMC.
>>>
>>>
>>>
>>> Agreed. My personal opinion is, that it would be cool for Felix to 
>>> be able to provide such telnet. Is this enough ?
>>>
>>>
>>> Regards
>>> Felix
>>>
>>> PS: You CLA has been registered. Thanks.
>>>
>>>
>>
>>
>
>


-- 
--------------------------------------------------------------
Didier DONSEZ

Laboratoire LIG, Equipe ADELE
Universite Joseph Fourier
Bat. C, 220 rue de la Chimie, Domaine Universitaire
BP 53, 38041 Grenoble Cedex 9, France
Tel : +33 4 76 63 55 49           Fax : +33 4 76 63 55 50
GPS : lat 45°11'38.3"N, lon 05°46'14.7"E, alt 223m

mailto:Didier.Donsez@imag.fr
URL: http://www-adele.imag.fr/users/Didier.Donsez
Map: http://www-adele.imag.fr/users/Didier.Donsez/map/map.html
--------------------------------------------------------------


Re: telnetd discussion

Posted by "Richard S. Hall" <he...@ungoverned.org>.
Of course there are alternatives, but that is not really at issue here.

We want alternatives, so the question to answer is: Is a full-featured 
telnet bundle a good alternative?

I say, yes.

-> richard

Didier Donsez wrote:
> Hi all
>
> An alternative to telnet shell or ssh shell could be JMX over 
> "everything".
> I have commit a Shell mbean in my sandbox
> http://svn.apache.org/repos/asf/felix/sandbox/donsez/mosgi.util/
> http://svn.apache.org/repos/asf/felix/sandbox/donsez/mbean.shell/
> http://svn.apache.org/repos/asf/felix/sandbox/donsez/mbean.shell.felix/
>
> I have also one mbean for the KF shell and one for the Equinox shell.
>
> I'm currently refactoring the client side to remove the dependencies 
> with the JConsole (GPLv2 Classpath Exception).
> Here is a snapshot of the current JConsole client (in the VisualVM).
> http://www-adele.imag.fr/users/Didier.Donsez/dev/osgi/jconsole.osgi/jconsole.shell.png 
>
>
> Best regards
>
> Didier
>
>
>
> Felix Meschberger wrote:
>
>> Hi Dieter,
>>
>> Dieter Wimberger schrieb:
>>
>>> If it is for a simple access to the Felix Shell Service that uses:
>>> shs.executeCommand(String line,PrintStream out,PrintStream out);
>>> then I don't see much sense to move this to Apache Felix, because 
>>> the simple remote shell does the job effectively and is much more 
>>> efficient. I think even if SSH is the argument we may simply try to 
>>> make a "SSH version" (using jaramiko).
>>
>>
>> I agree completely.
>>
>>>
>>> So let's collect good reasons, because these would also be the 
>>> foundation to argue an inclusion with/in the Felix PMC.
>>
>>
>> Agreed. My personal opinion is, that it would be cool for Felix to be 
>> able to provide such telnet. Is this enough ?
>>
>>
>> Regards
>> Felix
>>
>> PS: You CLA has been registered. Thanks.
>>
>>
>
>

Re: telnetd discussion

Posted by Didier Donsez <di...@imag.fr>.
Hi all

An alternative to telnet shell or ssh shell could be JMX over "everything".
I have commit a Shell mbean in my sandbox
http://svn.apache.org/repos/asf/felix/sandbox/donsez/mosgi.util/
http://svn.apache.org/repos/asf/felix/sandbox/donsez/mbean.shell/
http://svn.apache.org/repos/asf/felix/sandbox/donsez/mbean.shell.felix/

I have also one mbean for the KF shell and one for the Equinox shell.

I'm currently refactoring the client side to remove the dependencies 
with the JConsole (GPLv2 Classpath Exception).
Here is a snapshot of the current JConsole client (in the VisualVM).
http://www-adele.imag.fr/users/Didier.Donsez/dev/osgi/jconsole.osgi/jconsole.shell.png

Best regards

Didier



Felix Meschberger wrote:

> Hi Dieter,
>
> Dieter Wimberger schrieb:
>
>> If it is for a simple access to the Felix Shell Service that uses:
>> shs.executeCommand(String line,PrintStream out,PrintStream out);
>> then I don't see much sense to move this to Apache Felix, because the 
>> simple remote shell does the job effectively and is much more 
>> efficient. I think even if SSH is the argument we may simply try to 
>> make a "SSH version" (using jaramiko).
>
>
> I agree completely.
>
>>
>> So let's collect good reasons, because these would also be the 
>> foundation to argue an inclusion with/in the Felix PMC.
>
>
> Agreed. My personal opinion is, that it would be cool for Felix to be 
> able to provide such telnet. Is this enough ?
>
>
> Regards
> Felix
>
> PS: You CLA has been registered. Thanks.
>
>


-- 
--------------------------------------------------------------
Didier DONSEZ

Laboratoire LIG, Equipe ADELE
Universite Joseph Fourier
Bat. C, 220 rue de la Chimie, Domaine Universitaire
BP 53, 38041 Grenoble Cedex 9, France
Tel : +33 4 76 63 55 49           Fax : +33 4 76 63 55 50
GPS : lat 45°11'38.3"N, lon 05°46'14.7"E, alt 223m

mailto:Didier.Donsez@imag.fr
URL: http://www-adele.imag.fr/users/Didier.Donsez
Map: http://www-adele.imag.fr/users/Didier.Donsez/map/map.html
--------------------------------------------------------------


Re: telnetd discussion

Posted by Felix Meschberger <fm...@gmail.com>.
Hi Dieter,

Dieter Wimberger schrieb:
> If it is for a simple access to the Felix Shell Service that uses:
> shs.executeCommand(String line,PrintStream out,PrintStream out);
> then I don't see much sense to move this to Apache Felix, because the 
> simple remote shell does the job effectively and is much more efficient. 
> I think even if SSH is the argument we may simply try to make a "SSH 
> version" (using jaramiko).

I agree completely.

> 
> So let's collect good reasons, because these would also be the 
> foundation to argue an inclusion with/in the Felix PMC.

Agreed. My personal opinion is, that it would be cool for Felix to be 
able to provide such telnet. Is this enough ?


Regards
Felix

PS: You CLA has been registered. Thanks.


Re: telnetd discussion

Posted by Dieter Wimberger <di...@wimpi.net>.
Hi Felix,

> Hi Dieter,
>
> Dieter Wimberger schrieb:
>> Felix:
>>> As I understand you are willing to donate the simple telnet code  
>>> described in [1] to the Apache Felix project. To get around the  
>>> licensing cliffs, we need to follow a few steps:
>> Right.
>>> * First I suggest you create a JIRA and attach the source
>>>   code (including the correct Apache License file header)
>> Done.
>> https://issues.apache.org/jira/browse/FELIX-615
>> I have added the headers and the POM, but it doesn't seem to build  
>> a bundle so far.
>> Please note that this code is my copyright (corrected in the POM),  
>> and never had anything to do with the VFI (I am not contracted by  
>> the VFI nor bound by any contract that all code I write is "theirs").
>
> Thanks. I will be looking into this.

I will add the logging (through the LogService) and fix a bug that  
Craig found (killing the connection may cause a infinite loop).
I'll upload that version to the same issue asap.

> First off, IMHO forking your code into Apache Felix and keep two  
> (almost) identical projects with their own lifecycle is not  
> practical. Having said this, I think, that if you contribute your  
> code, Apache Felix would be very wise to take your own agenda into  
> account.
>
> And, we also do not want to be a code sink. That is if some code is  
> contributed and the Apache Felix PMC decides to accept the  
> contribution, the PMC is also considering the future of the code.

I agree, and there was never any intention to abuse Apache Felix as a  
code sink. That's why I have been trying to figure out the reasons why  
you would like to add this to the Apache Felix project.

If it is for a simple access to the Felix Shell Service that uses:
shs.executeCommand(String line,PrintStream out,PrintStream out);
then I don't see much sense to move this to Apache Felix, because the  
simple remote shell does the job effectively and is much more  
efficient. I think even if SSH is the argument we may simply try to  
make a "SSH version" (using jaramiko).

So let's collect good reasons, because these would also be the  
foundation to argue an inclusion with/in the Felix PMC.

Regards,
Dieter