ns-3.31 release candidate next week?

classic Classic list List threaded Threaded
12 messages Options
Reply | Threaded
Open this post in threaded view
|

ns-3.31 release candidate next week?

Tom Henderson-2
Hi all, the ns-3.31 release has been delayed for some time, mainly
because of a significant amount of work in the wifi module, which now
seems to be wrapping up.  The delay has also been due to my hope to add
a TCP Cubic model; I thought that we were close but I understand that
some people are working on some issues with it and ns-3-dev.

We were able to quickly recover from some issues uncovered with
Config::Connect changes this week, thanks to response from several
maintainers.  I'm wondering whether we try to keep the momentum and
finish ns-3.31 now.

Tommaso suggested in some off-list mail to perhaps just merge all of the
pending (comments resolved) bug fixes in the merge request tracker (i.e.
not new features), and release based on that.  I am willing to do that,
and think we could rapidly get something together next week.  I went
through the tracker and labelled everything I think is a candidate with
the ns-3.31 milestone; some may require a small amount of work to close.
  Can other maintainers chime in on whether they agree to finalize
ns-3.31 now, and have a look at these merge requests?

- Tom
Reply | Threaded
Open this post in threaded view
|

Re: ns-3.31 release candidate next week?

Natale Patriciello
Il giorno ven 8 mag 2020 alle ore 19:28 Tom Henderson <[hidden email]>
ha scritto:
>
[cut]

> Tommaso suggested in some off-list mail to perhaps just merge all of the
> pending (comments resolved) bug fixes in the merge request tracker (i.e.
> not new features), and release based on that.  I am willing to do that,
> and think we could rapidly get something together next week.  I went
> through the tracker and labelled everything I think is a candidate with
> the ns-3.31 milestone; some may require a small amount of work to close.
>   Can other maintainers chime in on whether they agree to finalize
> ns-3.31 now, and have a look at these merge requests?


Hi, from my side I'd ask what about Michele's merge request on the
generic interface for the 3gpp channel. If that's ready, can we
include it?

Thanks

Nat
Reply | Threaded
Open this post in threaded view
|

Re: ns-3.31 release candidate next week?

Michele Polese
Hi,
it is ready from my side. I don't know if someone else has other comments,
in case I can try to address them quickly.
Thanks,
Michele

Il giorno lun 11 mag 2020 alle ore 12:35 Natale Patriciello <
[hidden email]> ha scritto:

> Il giorno ven 8 mag 2020 alle ore 19:28 Tom Henderson <[hidden email]>
> ha scritto:
> >
> [cut]
>
> > Tommaso suggested in some off-list mail to perhaps just merge all of the
> > pending (comments resolved) bug fixes in the merge request tracker (i.e.
> > not new features), and release based on that.  I am willing to do that,
> > and think we could rapidly get something together next week.  I went
> > through the tracker and labelled everything I think is a candidate with
> > the ns-3.31 milestone; some may require a small amount of work to close.
> >   Can other maintainers chime in on whether they agree to finalize
> > ns-3.31 now, and have a look at these merge requests?
>
>
> Hi, from my side I'd ask what about Michele's merge request on the
> generic interface for the 3gpp channel. If that's ready, can we
> include it?
>
> Thanks
>
> Nat
>


--
Michele Polese

Research Scientist
Institute for the Wireless Internet of Things
Northeastern University, Boston, MA

email: [hidden email]
web: polese.io - mmwave.dei.unipd.it
Reply | Threaded
Open this post in threaded view
|

Re: ns-3.31 release candidate next week?

Tommaso Pecorella
In reply to this post by Tom Henderson-2
I'm working on (and I'd like to push before 3.31) the followings:
- Refactor Ipv4L3Protocol::Send to fragment broadcast packets <https://gitlab.com/nsnam/ns-3-dev/-/merge_requests/107>
- Example showing multicast routing setup for flooding mesh topology <https://gitlab.com/nsnam/ns-3-dev/-/merge_requests/133>
They're both "ready", I just need JaredD to update the repos so I can push.

I am ready to push:
- Mpi examples <https://gitlab.com/nsnam/ns-3-dev/-/merge_requests/145> (just waiting from a green light from Peter)

Other ones that can be pushed at will are:
- Tag uses of NS_DEPRECATED by release <https://gitlab.com/nsnam/ns-3-dev/-/merge_requests/280> (Peter, feel free to push)
- network: refactor DelayJitterEstimation to more closely follow RFC 1889 and RFC 3550 <https://gitlab.com/nsnam/ns-3-dev/-/merge_requests/239>  (Peter, feel free to push)
- Avoid use of null pointer in IPv6 fragmentation timeout <https://gitlab.com/nsnam/ns-3-dev/-/merge_requests/227> (Tom, feel free to push)

Other ones that need some testing (just to be extra sure) - but that I feel that they should be included in 3.31 are:
- core: use variadic template for CreateObject <https://gitlab.com/nsnam/ns-3-dev/-/merge_requests/147>
- Use variadic template for ObjectFactory::Set <https://gitlab.com/nsnam/ns-3-dev/-/merge_requests/276>
- Use std::function for callbacks <https://gitlab.com/nsnam/ns-3-dev/-/merge_requests/278>
- core: avoid implicit Time conversions <https://gitlab.com/nsnam/ns-3-dev/-/merge_requests/240>

These are my picks.

We might decide to delay again the variadic templates but I'm a bit ruthless, so I'll follow what we decide to do. What we need to check is that:
- Python bindings are all ok, and
- The supported compilers are not complaining - this means valgrind tests.

There are more merge request marked with the 3.31 milestone, we might as well check them and try to fast-track them.

Cheers,

T.




> On 8 May 2020, at 19:28, Tom Henderson <[hidden email]> wrote:
>
> Hi all, the ns-3.31 release has been delayed for some time, mainly because of a significant amount of work in the wifi module, which now seems to be wrapping up.  The delay has also been due to my hope to add a TCP Cubic model; I thought that we were close but I understand that some people are working on some issues with it and ns-3-dev.
>
> We were able to quickly recover from some issues uncovered with Config::Connect changes this week, thanks to response from several maintainers.  I'm wondering whether we try to keep the momentum and finish ns-3.31 now.
>
> Tommaso suggested in some off-list mail to perhaps just merge all of the pending (comments resolved) bug fixes in the merge request tracker (i.e. not new features), and release based on that.  I am willing to do that, and think we could rapidly get something together next week.  I went through the tracker and labelled everything I think is a candidate with the ns-3.31 milestone; some may require a small amount of work to close.  Can other maintainers chime in on whether they agree to finalize ns-3.31 now, and have a look at these merge requests?
>
> - Tom

--------------------------------------------------------------

Thinking evolution:
 "To be is to do" - Socrates
 "To do is to be" - Sartre
 "Do Be Do Be Do" - Sinatra
 "Scooby Dooby Do" - Scooby Do
 "Yaba Daba Doo!" - Fred Flintstone

--------------------------------------------------------------

Tommaso Pecorella - Ph.D.

Assistant professor
Dpt. Ingegneria dell'Informazione
Università di Firenze

CNIT - Università di Firenze Unit

via di S. Marta 3
50139, Firenze
ITALY

email: [hidden email]
       [hidden email]

phone : +39-055-2758540
mobile: +39-320-4379803
fax   : +39-055-2758570








Reply | Threaded
Open this post in threaded view
|

Re: ns-3.31 release candidate next week?

Tom Henderson-2
In reply to this post by Tom Henderson-2
Hi all,

I have been working on the release in my available time.  I need to
mostly pause until mid-week, but I have updated a number of issues. The
Python bindings are presently not compiling (issue #215) and some people
have ideas on how to approach that, but no patch is available yet.

As time goes on, more issues and merge requests show up, and various
things are merged, and the release continues to be a moving target. 
Some merges in the past week have broken the build and regression tests,
which causes further delays.  I have labeled everything that I would
like to see in ns-3.31 with the 'ns-3.31' milestone in the issue and
merge requests trackers.  Gitlab provides a way to filter issues by
milestones.  Can maintainers please review the ns-3.31 list and, if I
have missed something important, add the milestone to it, but otherwise,
restrict this week's commits to closing these milestone issues and MRs only?

Thanks,

Tom

Reply | Threaded
Open this post in threaded view
|

Re: ns-3.31 release candidate next week?

Tommaso Pecorella
I totally agree.

We should re-enforce the policy that no merges should be done except from:
1) important bug fixes,
2) important bug fixes,
3) important bug fixes

during the tag preparation period. And by important bug fixes I would include only:
- wrong behaviour in models
- memory errors and bugs

Mind, I'm the first (but not the only one) breaking this rule.

I'll try to work on the open 3.31 issues.

Cheers,

T.



> On 1 Jun 2020, at 19:50, Tom Henderson <[hidden email]> wrote:
>
> Hi all,
>
> I have been working on the release in my available time.  I need to mostly pause until mid-week, but I have updated a number of issues. The Python bindings are presently not compiling (issue #215) and some people have ideas on how to approach that, but no patch is available yet.
>
> As time goes on, more issues and merge requests show up, and various things are merged, and the release continues to be a moving target.  Some merges in the past week have broken the build and regression tests, which causes further delays.  I have labeled everything that I would like to see in ns-3.31 with the 'ns-3.31' milestone in the issue and merge requests trackers.  Gitlab provides a way to filter issues by milestones.  Can maintainers please review the ns-3.31 list and, if I have missed something important, add the milestone to it, but otherwise, restrict this week's commits to closing these milestone issues and MRs only?
>
> Thanks,
>
> Tom
>

--------------------------------------------------------------

The nice thing about standards is that there are so many to choose from.
And if you really don't like all the standards you just have to wait another year until the one arises you are looking for.
-- A. Tanenbaum, "Introduction to Computer Networks"

--------------------------------------------------------------

Tommaso Pecorella - Ph.D.

Assistant professor
Dpt. Ingegneria dell'Informazione
Università di Firenze

CNIT - Università di Firenze Unit

via di S. Marta 3
50139, Firenze
ITALY

email: [hidden email]
       [hidden email]

phone : +39-055-2758540
mobile: +39-320-4379803
fax   : +39-055-2758570









Reply | Threaded
Open this post in threaded view
|

Re: ns-3.31 release candidate next week?

Tom Henderson-2
On 6/2/20 6:47 AM, Tommaso Pecorella wrote:
> I totally agree.
>
> We should re-enforce the policy that no merges should be done except from:
> 1) important bug fixes,
> 2) important bug fixes,
> 3) important bug fixes

I am interested in merging a couple more new things, especially those
that I have said I would handle at an earlier date (Michele has a patch
pending, and also several of us working on the Wi-Fi Bianchi example
would like to commit that before the release), but please coordinate
with me and others involved if you want to commit something other than
the listed milestone items.

- Tom

Reply | Threaded
Open this post in threaded view
|

Re: ns-3.31 release candidate next week?

Tommaso Pecorella
Can someone check and provide feedbacks on https://gitlab.com/nsnam/ns-3-dev/-/issues/199 <https://gitlab.com/nsnam/ns-3-dev/-/issues/199> ?

Thanks,

T.


> On 2 Jun 2020, at 16:36, Tom Henderson <[hidden email]> wrote:
>
> On 6/2/20 6:47 AM, Tommaso Pecorella wrote:
>> I totally agree.
>>
>> We should re-enforce the policy that no merges should be done except from:
>> 1) important bug fixes,
>> 2) important bug fixes,
>> 3) important bug fixes
>
> I am interested in merging a couple more new things, especially those that I have said I would handle at an earlier date (Michele has a patch pending, and also several of us working on the Wi-Fi Bianchi example would like to commit that before the release), but please coordinate with me and others involved if you want to commit something other than the listed milestone items.
>
> - Tom
>

--------------------------------------------------------------

``It's not worth doing something unless you were doing something that someone, somewere, would much rather you weren't doing.''
-- Terry Pratchett

--------------------------------------------------------------

Tommaso Pecorella - Ph.D.

Assistant professor
Dpt. Ingegneria dell'Informazione
Università di Firenze

CNIT - Università di Firenze Unit

via di S. Marta 3
50139, Firenze
ITALY

email: [hidden email]
       [hidden email]

phone : +39-055-2758540
mobile: +39-320-4379803
fax   : +39-055-2758570








Reply | Threaded
Open this post in threaded view
|

Status of DCE

Peter Kourzanov
In reply to this post by Tom Henderson-2
Hi,

I see the new release 3.31 is about to start taking shape.

Maybe it's a good idea to summarize the status of the upcoming DCE release
11 as well?

I am mostly curious about what kernel versions (Linux and FreeBSD), ns3,
ns3-DCE versions and other components are supposed to run in combination on
which (Ubuntu, etc) release? Which combinations were tested with which
apps?

The reason I am asking is because I had mixed success with DCE 10 on recent
Ubuntu machines and unsure which Docker image should be used for best
results (would prefer to move on from 17.04).

Kind regards,
Peter




On Fri, 8 May 2020, 19:28 Tom Henderson <[hidden email]> wrote:

> Hi all, the ns-3.31 release has been delayed for some time, mainly
> because of a significant amount of work in the wifi module, which now
> seems to be wrapping up.  The delay has also been due to my hope to add
> a TCP Cubic model; I thought that we were close but I understand that
> some people are working on some issues with it and ns-3-dev.
>
> We were able to quickly recover from some issues uncovered with
> Config::Connect changes this week, thanks to response from several
> maintainers.  I'm wondering whether we try to keep the momentum and
> finish ns-3.31 now.
>
> Tommaso suggested in some off-list mail to perhaps just merge all of the
> pending (comments resolved) bug fixes in the merge request tracker (i.e.
> not new features), and release based on that.  I am willing to do that,
> and think we could rapidly get something together next week.  I went
> through the tracker and labelled everything I think is a candidate with
> the ns-3.31 milestone; some may require a small amount of work to close.
>   Can other maintainers chime in on whether they agree to finalize
> ns-3.31 now, and have a look at these merge requests?
>
> - Tom
>
Reply | Threaded
Open this post in threaded view
|

Re: Status of DCE

Matt
Hi,

The valid combinations should be the one on the CI: hajime setup an
extensive testsuite based on docker.
https://github.com/direct-code-execution/ns-3-dce/blob/master/.circleci/config.yml

As for the rest to be frank I don't know. I can run the testsuite on a
minimal config with libc 3.31 but I have not tested with a linux
kernel. Tom and I updated waf for python3 but that's all mostly. I
dont use DCE anymore but I can merge patches. The Linux Kernel Library
has started its upstreaming process so once it is done, DCE could
integrate more easily with linux.

Cheers
Matt

2020年6月11日(木) 10:22 Peter Kourzanov <[hidden email]>:

>
> Hi,
>
> I see the new release 3.31 is about to start taking shape.
>
> Maybe it's a good idea to summarize the status of the upcoming DCE release
> 11 as well?
>
> I am mostly curious about what kernel versions (Linux and FreeBSD), ns3,
> ns3-DCE versions and other components are supposed to run in combination on
> which (Ubuntu, etc) release? Which combinations were tested with which
> apps?
>
> The reason I am asking is because I had mixed success with DCE 10 on recent
> Ubuntu machines and unsure which Docker image should be used for best
> results (would prefer to move on from 17.04).
>
> Kind regards,
> Peter
>
>
>
>
> On Fri, 8 May 2020, 19:28 Tom Henderson <[hidden email]> wrote:
>
> > Hi all, the ns-3.31 release has been delayed for some time, mainly
> > because of a significant amount of work in the wifi module, which now
> > seems to be wrapping up.  The delay has also been due to my hope to add
> > a TCP Cubic model; I thought that we were close but I understand that
> > some people are working on some issues with it and ns-3-dev.
> >
> > We were able to quickly recover from some issues uncovered with
> > Config::Connect changes this week, thanks to response from several
> > maintainers.  I'm wondering whether we try to keep the momentum and
> > finish ns-3.31 now.
> >
> > Tommaso suggested in some off-list mail to perhaps just merge all of the
> > pending (comments resolved) bug fixes in the merge request tracker (i.e.
> > not new features), and release based on that.  I am willing to do that,
> > and think we could rapidly get something together next week.  I went
> > through the tracker and labelled everything I think is a candidate with
> > the ns-3.31 milestone; some may require a small amount of work to close.
> >   Can other maintainers chime in on whether they agree to finalize
> > ns-3.31 now, and have a look at these merge requests?
> >
> > - Tom
> >
Reply | Threaded
Open this post in threaded view
|

Re: Status of DCE

Peter Kourzanov
Hi,
Just to return to the DCE briefly. I think the primary benefit of the DCE
is actually the ability to use networking (i.e., kernel) stacks that are
out there in the wild, so it would indeed make a lot of sense to track
recent versions of Linux and FreeBSD more closely.

Regarding FreeBSD, I managed to get 12.1.0 compile and link with the latest
DCE, it would be possible to share some updates here, in addition to a
mix-stack example in ns-3-dce (mix ns3/freebsd) which was surprisingly
missing.

Kind regards,
Peter

On Mon, Jun 22, 2020 at 11:03 PM Matt <[hidden email]> wrote:

> Hi,
>
> The valid combinations should be the one on the CI: hajime setup an
> extensive testsuite based on docker.
>
> https://github.com/direct-code-execution/ns-3-dce/blob/master/.circleci/config.yml
>
> As for the rest to be frank I don't know. I can run the testsuite on a
> minimal config with libc 3.31 but I have not tested with a linux
> kernel. Tom and I updated waf for python3 but that's all mostly. I
> dont use DCE anymore but I can merge patches. The Linux Kernel Library
> has started its upstreaming process so once it is done, DCE could
> integrate more easily with linux.
>
> Cheers
> Matt
>
> 2020年6月11日(木) 10:22 Peter Kourzanov <[hidden email]>:
> >
> > Hi,
> >
> > I see the new release 3.31 is about to start taking shape.
> >
> > Maybe it's a good idea to summarize the status of the upcoming DCE
> release
> > 11 as well?
> >
> > I am mostly curious about what kernel versions (Linux and FreeBSD), ns3,
> > ns3-DCE versions and other components are supposed to run in combination
> on
> > which (Ubuntu, etc) release? Which combinations were tested with which
> > apps?
> >
> > The reason I am asking is because I had mixed success with DCE 10 on
> recent
> > Ubuntu machines and unsure which Docker image should be used for best
> > results (would prefer to move on from 17.04).
> >
> > Kind regards,
> > Peter
> >
> >
> >
> >
> > On Fri, 8 May 2020, 19:28 Tom Henderson <[hidden email]> wrote:
> >
> > > Hi all, the ns-3.31 release has been delayed for some time, mainly
> > > because of a significant amount of work in the wifi module, which now
> > > seems to be wrapping up.  The delay has also been due to my hope to add
> > > a TCP Cubic model; I thought that we were close but I understand that
> > > some people are working on some issues with it and ns-3-dev.
> > >
> > > We were able to quickly recover from some issues uncovered with
> > > Config::Connect changes this week, thanks to response from several
> > > maintainers.  I'm wondering whether we try to keep the momentum and
> > > finish ns-3.31 now.
> > >
> > > Tommaso suggested in some off-list mail to perhaps just merge all of
> the
> > > pending (comments resolved) bug fixes in the merge request tracker
> (i.e.
> > > not new features), and release based on that.  I am willing to do that,
> > > and think we could rapidly get something together next week.  I went
> > > through the tracker and labelled everything I think is a candidate with
> > > the ns-3.31 milestone; some may require a small amount of work to
> close.
> > >   Can other maintainers chime in on whether they agree to finalize
> > > ns-3.31 now, and have a look at these merge requests?
> > >
> > > - Tom
> > >
>
Reply | Threaded
Open this post in threaded view
|

Re: Status of DCE

Tom Henderson-2
On 7/22/20 5:01 AM, Peter Kourzanov wrote:

> Hi,
> Just to return to the DCE briefly. I think the primary benefit of the
> DCE is actually the ability to use networking (i.e., kernel) stacks that
> are out there in the wild, so it would indeed make a lot of sense to
> track recent versions of Linux and FreeBSD more closely.
>
> Regarding FreeBSD, I managed to get 12.1.0 compile and link with the
> latest DCE, it would be possible to share some updates here, in addition
> to a mix-stack example in ns-3-dce (mix ns3/freebsd) which was
> surprisingly missing.
>

Peter, can you please submit any pull requests to the DCE GitHub site?
I will check with Matt about possibly making a new release sometime this
summer.

Thanks,
Tom