VDSM - top 10 with patches with no activity for more than 30 days

thoughts on how to trim these?
(in openstack gerrit they auto-abandon patches with no activity for a 
couple of weeks - author can revive them back when they are relevant)

  preferred_email             | count
  ----------------------------+------
  fsimonce(a)redhat.com         | 34
  smizrahi(a)redhat.com         | 23
  lvroyce(a)linux.vnet.ibm.com  | 13
  ewarszaw(a)redhat.com         | 12
  wudxw(a)linux.vnet.ibm.com    | 12
  xuhj(a)linux.vnet.ibm.com     | 11
  shaohef(a)linux.vnet.ibm.com  | 6
  lilei(a)linux.vnet.ibm.com    | 6
  zhshzhou(a)linux.vnet.ibm.com | 6
  shuming(a)linux.vnet.ibm.com  | 5

----- Original Message -----
>  From: "Itamar Heim" <iheim(a)redhat.com&gt;
>  To: vdsm-devel(a)lists.fedorahosted.org
>  Sent: Wednesday, February 20, 2013 5:39:21 PM
>  Subject: [vdsm] VDSM - top 10 with patches with no activity for more than 30	days
>  
>  thoughts on how to trim these?
>  (in openstack gerrit they auto-abandon patches with no activity for a
>  couple of weeks - author can revive them back when they are relevant)
>  
>    preferred_email             | count
>    ----------------------------+------
>    fsimonce(a)redhat.com         | 34
>    smizrahi(a)redhat.com         | 23
>    lvroyce(a)linux.vnet.ibm.com  | 13
>    ewarszaw(a)redhat.com         | 12
>    wudxw(a)linux.vnet.ibm.com    | 12
>    xuhj(a)linux.vnet.ibm.com     | 11
>    shaohef(a)linux.vnet.ibm.com  | 6
>    lilei(a)linux.vnet.ibm.com    | 6
>    zhshzhou(a)linux.vnet.ibm.com | 6
>    shuming(a)linux.vnet.ibm.com  | 5
>  _______________________________________________ 
Review day? Anyone thinks a monthly review day will
help?

On 03/01/2013 01:51 AM, Doron Fediuck wrote:
>  ----- Original Message -----
> > From: "Itamar Heim" <iheim(a)redhat.com&gt;
> > To: vdsm-devel(a)lists.fedorahosted.org
> > Sent: Wednesday, February 20, 2013 5:39:21 PM
> > Subject: [vdsm] VDSM - top 10 with patches with no activity for more than 30	days
> >
> > thoughts on how to trim these?
> > (in openstack gerrit they auto-abandon patches with no activity for a
> > couple of weeks - author can revive them back when they are relevant)
> >
> >    preferred_email             | count
> >    ----------------------------+------
> >    fsimonce(a)redhat.com         | 34
> >    smizrahi(a)redhat.com         | 23
> >    lvroyce(a)linux.vnet.ibm.com  | 13
> >    ewarszaw(a)redhat.com         | 12
> >    wudxw(a)linux.vnet.ibm.com    | 12
> >    xuhj(a)linux.vnet.ibm.com     | 11
> >    shaohef(a)linux.vnet.ibm.com  | 6
> >    lilei(a)linux.vnet.ibm.com    | 6
> >    zhshzhou(a)linux.vnet.ibm.com | 6
> >    shuming(a)linux.vnet.ibm.com  | 5
> > _______________________________________________
>  Review day? Anyone thinks a monthly review day will
>  help? 
+1 for review day.
>  _______________________________________________
>  vdsm-devel mailing list
>  vdsm-devel(a)lists.fedorahosted.org
>  https://lists.fedorahosted.org/mailman/listinfo/vdsm-devel

[Adam Litke]

On Thu, 2013-02-28 at 12:51 -0500, Doron Fediuck wrote:
>  ----- Original Message -----
>  > From: "Itamar Heim" <iheim(a)redhat.com&gt;
>  > To: vdsm-devel(a)lists.fedorahosted.org
>  > Sent: Wednesday, February 20, 2013 5:39:21 PM
>  > Subject: [vdsm] VDSM - top 10 with patches with no activity for more than 30	days
>  > 
>  > thoughts on how to trim these?
>  > (in openstack gerrit they auto-abandon patches with no activity for a
>  > couple of weeks - author can revive them back when they are relevant)
>  > 
>  >   preferred_email             | count
>  >   ----------------------------+------
>  >   fsimonce(a)redhat.com         | 34
>  >   smizrahi(a)redhat.com         | 23
>  >   lvroyce(a)linux.vnet.ibm.com  | 13
>  >   ewarszaw(a)redhat.com         | 12
>  >   wudxw(a)linux.vnet.ibm.com    | 12
>  >   xuhj(a)linux.vnet.ibm.com     | 11
>  >   shaohef(a)linux.vnet.ibm.com  | 6
>  >   lilei(a)linux.vnet.ibm.com    | 6
>  >   zhshzhou(a)linux.vnet.ibm.com | 6
>  >   shuming(a)linux.vnet.ibm.com  | 5
>  > _______________________________________________
>  
>  Review day? Anyone thinks a monthly review day will
>  help? 
We've discussed this in the past and part of the reason for the backlog
is that folks like Saggi and Federico like to use gerrit to store
work-in-progress patches that don't need review.  They may not be
working on those patches at the moment but want them in gerrit for them
to come back to.  If we want to allow this use of gerrit then we will
always have some stale patches lying around.

-- 
Adam Litke <agl(a)linux.vnet.ibm.com&gt;
IBM Linux Technology Center

On 01/03/2013 15:58, Adam Litke wrote:
>  On Thu, 2013-02-28 at 12:51 -0500, Doron Fediuck wrote:
> > ----- Original Message -----
> >> From: "Itamar Heim" <iheim(a)redhat.com&gt;
> >> To: vdsm-devel(a)lists.fedorahosted.org
> >> Sent: Wednesday, February 20, 2013 5:39:21 PM
> >> Subject: [vdsm] VDSM - top 10 with patches with no activity for more than
> 30	days
> >>
> >> thoughts on how to trim these?
> >> (in openstack gerrit they auto-abandon patches with no activity for a
> >> couple of weeks - author can revive them back when they are relevant)
> >>
> >>    preferred_email             | count
> >>    ----------------------------+------
> >>    fsimonce(a)redhat.com         | 34
> >>    smizrahi(a)redhat.com         | 23
> >>    lvroyce(a)linux.vnet.ibm.com  | 13
> >>    ewarszaw(a)redhat.com         | 12
> >>    wudxw(a)linux.vnet.ibm.com    | 12
> >>    xuhj(a)linux.vnet.ibm.com     | 11
> >>    shaohef(a)linux.vnet.ibm.com  | 6
> >>    lilei(a)linux.vnet.ibm.com    | 6
> >>    zhshzhou(a)linux.vnet.ibm.com | 6
> >>    shuming(a)linux.vnet.ibm.com  | 5
> >> _______________________________________________
> >
> > Review day? Anyone thinks a monthly review day will
> > help?
>
>  We've discussed this in the past and part of the reason for the backlog
>  is that folks like Saggi and Federico like to use gerrit to store
>  work-in-progress patches that don't need review.  They may not be
>  working on those patches at the moment but want them in gerrit for them
>  to come back to.  If we want to allow this use of gerrit then we will
>  always have some stale patches lying around.
>  
I think a private github is a better place for soemthing which gets no 
attention for so long.
personally i'd like to make sure patches do get attention, and having 
old ones is cluttering.
I like openstack auto-abandon approach. i think a patch with no activity 
for more than 30 days should be auto abandoned. its trivial to get it 
back to life after that (openstack are much harsher than this actually)

On Fri, Mar 01, 2013 at 04:27:33PM +0200, Itamar Heim wrote:
>  On 01/03/2013 15:58, Adam Litke wrote:
>  >On Thu, 2013-02-28 at 12:51 -0500, Doron Fediuck wrote:
>  >>----- Original Message -----
>  >>>From: "Itamar Heim" <iheim(a)redhat.com&gt;
>  >>>To: vdsm-devel(a)lists.fedorahosted.org
>  >>>Sent: Wednesday, February 20, 2013 5:39:21 PM
>  >>>Subject: [vdsm] VDSM - top 10 with patches with no activity for more than
> 30	days
>  >>>
>  >>>thoughts on how to trim these?
>  >>>(in openstack gerrit they auto-abandon patches with no activity for a
>  >>>couple of weeks - author can revive them back when they are relevant)
>  >>>
>  >>>   preferred_email             | count
>  >>>   ----------------------------+------
>  >>>   fsimonce(a)redhat.com         | 34
>  >>>   smizrahi(a)redhat.com         | 23
>  >>>   lvroyce(a)linux.vnet.ibm.com  | 13
>  >>>   ewarszaw(a)redhat.com         | 12
>  >>>   wudxw(a)linux.vnet.ibm.com    | 12
>  >>>   xuhj(a)linux.vnet.ibm.com     | 11
>  >>>   shaohef(a)linux.vnet.ibm.com  | 6
>  >>>   lilei(a)linux.vnet.ibm.com    | 6
>  >>>   zhshzhou(a)linux.vnet.ibm.com | 6
>  >>>   shuming(a)linux.vnet.ibm.com  | 5
>  >>>_______________________________________________
>  >>
>  >>Review day? Anyone thinks a monthly review day will
>  >>help?
>  >
>  >We've discussed this in the past and part of the reason for the backlog
>  >is that folks like Saggi and Federico like to use gerrit to store
>  >work-in-progress patches that don't need review.  They may not be
>  >working on those patches at the moment but want them in gerrit for them
>  >to come back to.  If we want to allow this use of gerrit then we will
>  >always have some stale patches lying around.
>  >
>
>  I think a private github is a better place for soemthing which gets
>  no attention for so long.
>  personally i'd like to make sure patches do get attention, and
>  having old ones is cluttering. 
Don't gerrit drafts suffice for personal WIP branches?

>  I like openstack auto-abandon approach. i think a patch with no
>  activity for more than 30 days should be auto abandoned. its trivial
>  to get it back to life after that (openstack are much harsher than
>  this actually) 
As long as the user is notified I agree. A 30 day old patch is likely to
no longer cleanly apply anyway so a rebase may be needed anyway.

On 01/03/2013 16:32, Ewoud Kohl van Wijngaarden wrote:
>  On Fri, Mar 01, 2013 at 04:27:33PM +0200, Itamar Heim wrote:
> > On 01/03/2013 15:58, Adam Litke wrote:
> >> On Thu, 2013-02-28 at 12:51 -0500, Doron Fediuck wrote:
> >>> ----- Original Message -----
> >>>> From: "Itamar Heim" <iheim(a)redhat.com&gt;
> >>>> To: vdsm-devel(a)lists.fedorahosted.org
> >>>> Sent: Wednesday, February 20, 2013 5:39:21 PM
> >>>> Subject: [vdsm] VDSM - top 10 with patches with no activity for more than
> 30	days
> >>>>
> >>>> thoughts on how to trim these?
> >>>> (in openstack gerrit they auto-abandon patches with no activity for a
> >>>> couple of weeks - author can revive them back when they are relevant)
> >>>>
> >>>>    preferred_email             | count
> >>>>    ----------------------------+------
> >>>>    fsimonce(a)redhat.com         | 34
> >>>>    smizrahi(a)redhat.com         | 23
> >>>>    lvroyce(a)linux.vnet.ibm.com  | 13
> >>>>    ewarszaw(a)redhat.com         | 12
> >>>>    wudxw(a)linux.vnet.ibm.com    | 12
> >>>>    xuhj(a)linux.vnet.ibm.com     | 11
> >>>>    shaohef(a)linux.vnet.ibm.com  | 6
> >>>>    lilei(a)linux.vnet.ibm.com    | 6
> >>>>    zhshzhou(a)linux.vnet.ibm.com | 6
> >>>>    shuming(a)linux.vnet.ibm.com  | 5
> >>>> _______________________________________________
> >>>
> >>> Review day? Anyone thinks a monthly review day will
> >>> help?
> >>
> >> We've discussed this in the past and part of the reason for the backlog
> >> is that folks like Saggi and Federico like to use gerrit to store
> >> work-in-progress patches that don't need review.  They may not be
> >> working on those patches at the moment but want them in gerrit for them
> >> to come back to.  If we want to allow this use of gerrit then we will
> >> always have some stale patches lying around.
> >>
> >
> > I think a private github is a better place for soemthing which gets
> > no attention for so long.
> > personally i'd like to make sure patches do get attention, and
> > having old ones is cluttering.
>
>  Don't gerrit drafts suffice for personal WIP branches?
>
> > I like openstack auto-abandon approach. i think a patch with no
> > activity for more than 30 days should be auto abandoned. its trivial
> > to get it back to life after that (openstack are much harsher than
> > this actually)
>
>  As long as the user is notified I agree. A 30 day old patch is likely to
>  no longer cleanly apply anyway so a rebase may be needed anyway. 
user gets an email on any change to their patch, including abandon.
we can send a heads up email a few days earlier as well

[Dave Neary]

Hi,

On 02/28/2013 06:51 PM, Doron Fediuck wrote:
> > thoughts on how to trim these?
> > (in openstack gerrit they auto-abandon patches with no activity for a
> > couple of weeks - author can revive them back when they are relevant)
>
>  Review day? Anyone thinks a monthly review day will
>  help? 
Yes, definitely a review day (announced beforehand) would help - another 
thing which would help in general is a page where you can see the oldest 
unreviewed patches.

Also, I would love to see us have some social way that people can start 
reviewing patches when they join the project - there's no better way to 
understand the project. The best way to lower the number of unreviewed 
patches is to have more reviewers.

Also - as someone not that familiar with Gerrit, what's involved in 
picking up someone's patch and revising it for them? A pattern I see 
over & over again is:

Jane submits patch
Alex reviews: -1 with suggestions for improvements
Jane submits patch rev #2
Barry reviews: -1 (whitespace issues)
Jane submits patch #3
Hoda reviews: -1 with suggestions for another approach
Jane loses interest and patch remains *almost* ready forever until it no 
longer applies or gets dropped.

Is there a way we can promote "adopt a patch" where you take someone 
else's patch and push it through review?

Cheers,
Dave.


-- 
Dave Neary - Community Action and Impact
Open Source and Standards, Red Hat - http://community.redhat.com
Ph: [phone] / Cell: [phone]

----- Original Message -----
>  From: "Dave Neary" <dneary(a)redhat.com&gt;
>  To: "Doron Fediuck" <dfediuck(a)redhat.com&gt;
>  Cc: "Itamar Heim" <iheim(a)redhat.com&gt;, vdsm-devel(a)lists.fedorahosted.org
>  Sent: Wednesday, March 6, 2013 6:54:56 PM
>  Subject: Re: [vdsm] VDSM - top 10 with patches with no activity for more than 30	days
>  
>  Hi,
>  
>  On 02/28/2013 06:51 PM, Doron Fediuck wrote:
>  >> thoughts on how to trim these?
>  >> (in openstack gerrit they auto-abandon patches with no activity
>  >> for a
>  >> couple of weeks - author can revive them back when they are
>  >> relevant)
>  >
>  > Review day? Anyone thinks a monthly review day will
>  > help?
>  
>  Yes, definitely a review day (announced beforehand) would help -
>  another
>  thing which would help in general is a page where you can see the
>  oldest
>  unreviewed patches.
>  
>  Also, I would love to see us have some social way that people can
>  start
>  reviewing patches when they join the project - there's no better way
>  to
>  understand the project. The best way to lower the number of
>  unreviewed
>  patches is to have more reviewers.
>  
>  Also - as someone not that familiar with Gerrit, what's involved in
>  picking up someone's patch and revising it for them? A pattern I see
>  over & over again is:
>  
>  Jane submits patch
>  Alex reviews: -1 with suggestions for improvements
>  Jane submits patch rev #2
>  Barry reviews: -1 (whitespace issues)
>  Jane submits patch #3
>  Hoda reviews: -1 with suggestions for another approach
>  Jane loses interest and patch remains *almost* ready forever until it
>  no
>  longer applies or gets dropped.
>  
>  Is there a way we can promote "adopt a patch" where you take someone
>  else's patch and push it through review?
>  
>  Cheers,
>  Dave.
>  
>  
>  --
>  Dave Neary - Community Action and Impact
>  Open Source and Standards, Red Hat - http://community.redhat.com
>  Ph: [phone] / Cell: [phone]
>   
Dave,
I agree, but this is also shared by engine dev's, so maybe
have a discussion on it in arch list?

On 03/06/2013 06:54 PM, Dave Neary wrote:
>  Also - as someone not that familiar with Gerrit, what's involved
> in
>  picking up someone's patch and revising it for them? A pattern I see
>  over & over again is:
>
>  Jane submits patch
>  Alex reviews: -1 with suggestions for improvements
>  Jane submits patch rev #2
>  Barry reviews: -1 (whitespace issues)
>  Jane submits patch #3
>  Hoda reviews: -1 with suggestions for another approach
>  Jane loses interest and patch remains *almost* ready forever until it no
>  longer applies or gets dropped.
>
>  Is there a way we can promote "adopt a patch" where you take someone
>  else's patch and push it through review? 
shouldn't be an issue for someone else to fetch the patch and submit an 
update to someone else's patch, based on same change-id