Re: XFS maintainership

Re: XFS maintainership

[Eric Sandeen]

Dave Chinner wrote:
> It seems to me that SGI wants to maintain control without doing any
> of the work that having that control requires of them.  i.e. take
> without any give....
> 
> Point in case: we have a _critical_ 2.6.28 regression w.r.t.
> directory handling. The community triaged the bug, the community
> fixed the bug and the community reviewed the fix. It got checked
> into the SGI controlled dev tree 4 days ago.  Now we are waiting for
> SGI to stop playing "let's all be one happy family la-la-la" games
> and get off their backsides and *act as responsible maintainers* by
> pushing the fix to Linus ASAP.
> 
> Please, show us that SGI is really going to act as the maintainer of
> XFS. The only thing that will convince me right now that SGI should
> continue as XFS maintainer is this:
> 
> 	"Gesta non verba"

Ooh, bonus points for the Latin!

Since 12/10, when Melbourne got erased, there have only been 6 emails
from sgi to the list which were not from the short-timer skeleton crew
left in Melbourne.  3 of these had something to do with development.   2
were related to this question of maintainership.  1 was a test email.

Meanwhile almost 100 patches have been sent, reviewed, and in many cases
committed by hch & others to the staging trees on kernel.org.  The
proposed new maintainer crew has not participated in this process yet to
any apparent degree.  No questions, no reviews, no acks, no vetoes.
This is not a personal attack by any means, but it seems that it might
reflect the resources available for these tasks inside sgi.

>From my perspective, it certainly appears that much more xfs work is
being done outside sgi than inside sgi at this point in time.  This
*should* be a good thing for sgi, because one of your flagship storage
software offerings is being maintained & moving forward with very few
resource requirements from sgi.

But if sgi's role is simply to own and to veto and not to communicate,
collaborate, facilitate or contribute, sgi will likely find that they've
been left behind in short order.  The internet is famous for routing
around damage.

On the other hand, we're here to help, if you engage us.

-Eric (speaking for myself, not my employer, FWIW)

> Cheers,
> 
> Dave.
> 
> PS: I did say I was going to make myself unpopular :/

Perhaps only with some :)

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

Re: XFS maintainership

[Felix Blyakher]

On Jan 14, 2009, at 11:23 AM, Eric Sandeen wrote:

> Dave Chinner wrote:
>> It seems to me that SGI wants to maintain control without doing any
>> of the work that having that control requires of them.  i.e. take
>> without any give....
>>
>> Point in case: we have a _critical_ 2.6.28 regression w.r.t.
>> directory handling. The community triaged the bug, the community
>> fixed the bug and the community reviewed the fix. It got checked
>> into the SGI controlled dev tree 4 days ago.  Now we are waiting for
>> SGI to stop playing "let's all be one happy family la-la-la" games
>> and get off their backsides and *act as responsible maintainers* by
>> pushing the fix to Linus ASAP.
>>
>> Please, show us that SGI is really going to act as the maintainer of
>> XFS. The only thing that will convince me right now that SGI should
>> continue as XFS maintainer is this:
>>
>> 	"Gesta non verba"
>
> Ooh, bonus points for the Latin!
>
> Since 12/10, when Melbourne got erased, there have only been 6 emails
> from sgi to the list which were not from the short-timer skeleton crew
> left in Melbourne.  3 of these had something to do with  
> development.   2
> were related to this question of maintainership.  1 was a test email.

OK, we're all here, and listening, and learning, and getting to
know the process and responsibilities, setting up right environment,
all that transitional stuff.
Yes, we're overwhelmed at the moment, but not going to hide in
the bushes. Just need some time.

> Meanwhile almost 100 patches have been sent, reviewed, and in many  
> cases
> committed by hch & others to the staging trees on kernel.org.

Hmm, while not actively participating, I've been monitoring
all xfs channels I know of. I haven't seen 100 patches lately.
Where they all posted to this list?
Also, I think, I replied to one. Sure, it's not the right level
of activity, but as I said, it's still transition period here.

>  The
> proposed new maintainer crew has not participated in this process  
> yet to
> any apparent degree.  No questions, no reviews, no acks, no vetoes.
> This is not a personal attack by any means, but it seems that it might
> reflect the resources available for these tasks inside sgi.
>
>> From my perspective, it certainly appears that much more xfs work is
> being done outside sgi than inside sgi at this point in time.  This
> *should* be a good thing for sgi, because one of your flagship storage
> software offerings is being maintained & moving forward with very few
> resource requirements from sgi.
>
> But if sgi's role is simply to own and to veto and not to communicate,
> collaborate, facilitate or contribute,

No, that's definitely not on anybody's mind here.
I know, community were waiting for our comments on the process,
but I tried not to reply with promises, but rather wait till
first commit. As it's not there yet, here is my reply, not the
way I wanted, though.

> sgi will likely find that they've
> been left behind in short order.  The internet is famous for routing
> around damage.
>
> On the other hand, we're here to help, if you engage us.

Thanks, we'll definitely need that, but expecting that to
be mutual. I have no doubt we'll contribute as well.

> -Eric (speaking for myself, not my employer, FWIW)

Felix (not speaking for sgi only)

>
>
>> Cheers,
>>
>> Dave.
>>
>> PS: I did say I was going to make myself unpopular :/
>
> Perhaps only with some :)
>
>
> _______________________________________________
> xfs mailing list
> xfs@oss.sgi.com
> http://oss.sgi.com/mailman/listinfo/xfs

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

Re: XFS maintainership

[Dave Chinner]

On Wed, Jan 14, 2009 at 12:05:01PM -0600, Felix Blyakher wrote:
> 
> On Jan 14, 2009, at 11:23 AM, Eric Sandeen wrote:
> 
> > Dave Chinner wrote:
> >> It seems to me that SGI wants to maintain control without doing any
> >> of the work that having that control requires of them.  i.e. take
> >> without any give....
> >>
> >> Point in case: we have a _critical_ 2.6.28 regression w.r.t.
> >> directory handling. The community triaged the bug, the community
> >> fixed the bug and the community reviewed the fix. It got checked
> >> into the SGI controlled dev tree 4 days ago.  Now we are waiting for
> >> SGI to stop playing "let's all be one happy family la-la-la" games
> >> and get off their backsides and *act as responsible maintainers* by
> >> pushing the fix to Linus ASAP.
> >>
> >> Please, show us that SGI is really going to act as the maintainer of
> >> XFS. The only thing that will convince me right now that SGI should
> >> continue as XFS maintainer is this:
> >>
> >> 	"Gesta non verba"
> >
> > Ooh, bonus points for the Latin!
> >
> > Since 12/10, when Melbourne got erased, there have only been 6 emails
> > from sgi to the list which were not from the short-timer skeleton crew
> > left in Melbourne.  3 of these had something to do with  
> > development.   2
> > were related to this question of maintainership.  1 was a test email.
> 
> OK, we're all here, and listening, and learning, and getting to
> know the process and responsibilities, setting up right environment,
> all that transitional stuff.
> Yes, we're overwhelmed at the moment, but not going to hide in
> the bushes. Just need some time.

Therein lies the problem. SGI is overwhelmed and playing catchup,
while there are some things that the community needs to be done
immediately. That's pretty much the story of the last six months
and the indications you are giving are that it will continue this
way for some time....

> > Meanwhile almost 100 patches have been sent, reviewed, and in many  
> > cases
> > committed by hch & others to the staging trees on kernel.org.
> 
> Hmm, while not actively participating, I've been monitoring
> all xfs channels I know of. I haven't seen 100 patches lately.
> Where they all posted to this list?

$ find incoming/xfs-linux/cur/ -newerct 20081210 -exec grep PATCH {} \; |grep ^Subject |grep -v [Rr]e: | wc -l
100

> >> From my perspective, it certainly appears that much more xfs work is
> > being done outside sgi than inside sgi at this point in time.  This
> > *should* be a good thing for sgi, because one of your flagship storage
> > software offerings is being maintained & moving forward with very few
> > resource requirements from sgi.
> >
> > But if sgi's role is simply to own and to veto and not to communicate,
> > collaborate, facilitate or contribute,
> 
> No, that's definitely not on anybody's mind here.
> I know, community were waiting for our comments on the process,

FWIW, I'm waiting for action, not comments...

Cheers,

Dave.
-- 
Dave Chinner
david@fromorbit.com

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

XFS maintainership...

[Chris Wedgwood]

On Wed, Jan 14, 2009 at 04:50:50PM +1100, Dave Chinner wrote:

> PS: I did say I was going to make myself unpopular :/

It depends with who.



As a Linux XFS user for what must be the best part of a decade now,
quite some time before it was merged --- I don't recall exactly when;
I've been happy with it, it's had a lot of good people behind it and
many incremental improvements and fixes.


I'd like to see that continue.  I'd really like to see:

 - the list move to vger' (please)

 - the maintainer-ship move to yourself and Christoph (if either or
   both of you are willing)

 - things to continue more or less as they have over the last 12-24
   months (or better, but not worse)


For various uses right now there is no viable stable alternative to
XFS.  It would be a shame to let XFS rot and degrade given the now
fairly large user base.  Most of which are probably completely unheard
of.

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs