9:43 a.m.
Hi all,
This morning, during the automation workshop, I had the occasion of
speaking about this with Pingou and Threebean.
Thanks to Pingou hints, I've created a query to get pending ACL from
pkgdb.
What I'd like to share with you all is the list of users that can
approve/deny ACL requests (older than 1 month) but have not done it yet
(the number refers to the number of ACL pending).
I think that people should take care of the pending ACL they can
approve/deny and actually approve or deny them ASAP.
263 cwickert
169 spot
123 bkabrda
114 patches
92 anishpatil
89 devrim
85 chkr
84 salimma
70 laxathom
63 mtasaka
53 sharkcz
50 mmorsi
43 timn
42 huzaifas
41 goldmann
37 ausil
36 willb
33 awjb
33 mmahut
31 dwmw2
31 hvad
30 mebrown
30 sxw
29 dcallagh
27 mclasen
27 dsommers
27 lystor
25 rmeggins
25 nhosoi
25 amigadave
25 pwouters
24 fnasser
24 pbrobinson
24 adev
24 ralph
24 dsd
24 corsepiu
23 steve
23 nkinder
22 brouhaha
22 jcapik
21 mbarnes
21 yyang
21 rishi
21 vcrhonek
21 mfojtik
21 thias
20 nb
20 jsteffan
19 tieugene
19 mharmsen
19 jamielinux
18 geoff
18 gecko-maint
18 ingvar
18 dwalsh
18 apevec
18 ajax
17 sdz
17 dvratil
17 gomix
17 thomasvs
16 jskarvad
16 tomprince
16 otaylor
16 dwalluck
16 smilner
15 caillon
15 kanarip
15 walters
15 nucleo
15 cjb
15 praveenp
14 sgallagh
14 s4504kr
14 flaper87
14 mlichvar
14 gholms
14 jstanley
13 lucilanga
13 rlandmann
13 tagoh
13 stingray
13 jvcelak
13 xavierb
13 spike
12 weli
12 chitlesh
12 melmorabity
12 sophiekovalevsky
12 alcapcom
12 susmit
12 robmv
12 elwell
12 bpepple
12 aalam
12 stransky
12 npmccallum
12 nushio
12 pvoborni
12 elad
12 dcbw
12 kylev
12 whot
12 ssp
11 fkocina
11 tuxbrewr
11 richardfearn
11 madko
11 oron
10 imcleod
10 tdabasin
10 potty
10 zaitcev
10 jamielennox
10 steved
10 fkluknav
10 clalance
10 lmacken
10 giallu
10 dmach
9 jklimes
9 larsks
9 pmatilai
9 elsupergomez
9 uraeus
9 rclark
9 shreyankg
9 tomspur
9 msivak
9 hubbitus
9 mgrepl
9 gemi
9 erikos
9 sundaram
9 stevetraylen
8 petersen
8 jeckersb
8 echevemaster
8 sseago
8 radez
8 puiterwijk
8 averi
8 jaruga
7 timj
7 terminalmage
7 mchehab
7 abo
7 djuran
7 lsm5
7 uwog
7 abbot
6 rmyers
6 rspanton
6 kdudka
6 ewan
6 rcritten
6 lennart
6 anttix
6 rohara
6 mstuchli
6 phatina
6 zeenix
6 bkearney
6 scox
6 matthicksj
6 tonet666p
6 alvesadrian
6 cleech
6 drsmith2
6 wolfy
6 hpejakle
6 radford
6 skottler
6 adrian
6 lzap
6 izhar
6 karsten
6 bentiss
6 ovasik
6 elanthis
6 mjw
6 sadmac
6 renich
6 jgarzik
6 hadess
6 scop
6 twoerner
6 pmachata
6 pfrankli
6 jpo
6 ianweller
6 lberk
6 pravins
6 dbhole
6 peter
6 fche
6 matriux
6 rvokal
6 mhlavink
6 sgrubb
6 ellert
6 nathans
6 myoung
6 jstanek
6 wcohen
6 cerberus
6 tomeu
6 john2583
6 jforbes
6 jistone
6 ivazquez
6 zironid
6 firnsy
5 praiskup
5 kushal
5 strobert
5 nsantos
5 dougsland
5 filiperosset
5 pavlix
5 jcp
5 sayanchowdhury
5 adalloz
5 sejeff
5 alikins
5 somlo
5 mikeb
5 jjh
5 delete
5 ixs
5 ssalevan
5 ndipanov
5 tbreeds
5 van
5 sross
4 rcurtin
4 fabbione
4 rajalakshmi
4 aviso
4 mdomsch
4 than
4 nbecker
4 besser82
4 denisarnaud
4 rakesh
4 kevin
4 nacho
4 aviram
4 sdake
4 davidz
4 toshio
4 herrold
4 daveisfera
4 akurtakov
3 james
3 magcius
3 perex
3 mmuzila
3 zpavlas
3 pjp
3 jbernard
3 johnp
3 dwayne
3 galt
3 nalin
3 meyering
3 rkennke
3 pjones
3 fabiand
3 hhorak
3 alexl
3 pbrady
3 rjones
3 cbb
3 nkumar
3 svashisht
3 marx
3 jcm
3 rharwood
3 jlaska
3 packaging-team
3 shenson
3 yselkowitz
3 clime
3 ffesti
3 davidx
3 acme
3 itamarjp
3 lsedlar
3 mcrha
3 guidograzioli
3 nim
3 vicodan
3 tk009
3 ke4qqq
3 mbooth
3 jwakely
3 jchaloup
3 martinkg
3 jkastner
3 ankursinha
3 mintojoseph
3 jdennis
3 leonn
3 jwboyer
3 mathstuf
3 oliver
3 hicham
3 williamjmorenor
3 xhorak
3 neugens
3 rrelyea
3 krege
3 amluto
3 omajid
3 jamesni
3 alexises
3 fonts-sig
3 athimm
3 jsynacek
3 jussilehtola
3 xinwu
3 olea
3 linuxdonald
3 katzj
3 kjaleel
3 bioinfornatics
3 vbatts
3 veillard
3 beckerde
3 mso
3 jimi
3 abompard
3 gerd
3 dborkmann
3 jbowes
3 airlied
3 junghans
3 dwrobel
3 jerboaa
3 zoglesby
2 harald
2 swagiaal
2 sheltren
2 smizrahi
2 jpopelka
2 cfu
2 janfrode
2 fschwarz
2 kay
2 frafra
2 glommer
2 aalvarez
2 hardaker
2 pmuldoon
2 jjohnstn
2 athmane
2 mjg59
2 micklweiss
2 epienbro
2 tstclair
2 gbailey
2 diegobz
2 pingou
2 stahnma
2 dtimms
2 thozza
2 berrange
2 xaeth
2 ensc
2 bronhaim
2 robert
2 cicku
1 zultron
1 bphinz
1 axelvvd
1 beekhof
1 zaneb
1 dknox
1 timlau
1 astokes
1 fsimonce
1 mizdebsk
1 spstarr
1 jkaluza
1 lnykryn
1 moceap
1 rrati
1 mrunge
1 lfarkas
1 mmraka
1 coolsvap
1 danken
1 miminar
1 russellb
1 rebus
1 hguemar
1 dros
1 imntreal
1 mystro256
1 hobbes1069
1 mycae
1 sgros
1 pertusus
1 kzak
1 gwei3
1 domcleal
1 ejratl
1 fab
I hope people will actually take care of those numbers and I'll
generate similar reports over time to keep track of the situation.
Cheers,
Fale
--
Fabio Alessandro Locati
Red Hat - Senior Consultant
PGP Fingerprint: E815 3C49 2A8D FD8B 1CBD BC85 FDB3 DF20 B2DC 9C1B
2:41 p.m.
On Thu, Aug 04, 2016 at 04:43:40PM +0200, Fabio Alessandro Locati wrote:
Hi all,
This morning, during the automation workshop, I had the occasion of
speaking about this with Pingou and Threebean.
Thanks to Pingou hints, I've created a query to get pending ACL from
pkgdb.
What I'd like to share with you all is the list of users that can
approve/deny ACL requests (older than 1 month) but have not done it yet
(the number refers to the number of ACL pending).
I think that people should take care of the pending ACL they can
approve/deny and actually approve or deny them ASAP.
Your email needs a "call to action" link, otherwise no one will know
what they are supposed to do about it. In this case it's probably:
https://admin.fedoraproject.org/pkgdb/acl/pending/
However I visited the above URL, logged in, and it says:
Pending ACLs
No pending ACLs for you
So I guess this is wrong or perhaps refers to something else:
...
3 rjones
...
Rich.
--
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
virt-top is 'top' for virtual machines. Tiny program with many
powerful monitoring features, net stats, disk stats, logging, etc.
http://people.redhat.com/~rjones/virt-top
4:14 a.m.
On Thu, 4 Aug 2016 20:41:56 +0100
"Richard W.M. Jones" <rjones(a)redhat.com> wrote:
On Thu, Aug 04, 2016 at 04:43:40PM +0200, Fabio Alessandro Locati
wrote:
> Hi all,
>
> This morning, during the automation workshop, I had the occasion of
> speaking about this with Pingou and Threebean.
> Thanks to Pingou hints, I've created a query to get pending ACL from
> pkgdb.
> What I'd like to share with you all is the list of users that can
> approve/deny ACL requests (older than 1 month) but have not done it
> yet (the number refers to the number of ACL pending).
>
> I think that people should take care of the pending ACL they can
> approve/deny and actually approve or deny them ASAP.
Your email needs a "call to action" link, otherwise no one will know
what they are supposed to do about it. In this case it's probably:
https://admin.fedoraproject.org/pkgdb/acl/pending/
However I visited the above URL, logged in, and it says:
Pending ACLs
No pending ACLs for you
also there should be no action required from the owner for "watchcommit"
or "watchbugzilla" requests, looks as a bug in the conversion when
deploying the recent pkgdb
Dan
8:07 a.m.
On Fri, Aug 05, 2016 at 11:14:02AM +0200, Dan Horák wrote:
On Thu, 4 Aug 2016 20:41:56 +0100
"Richard W.M. Jones" <rjones(a)redhat.com> wrote:
> On Thu, Aug 04, 2016 at 04:43:40PM +0200, Fabio Alessandro Locati
> wrote:
> > Hi all,
> >
> > This morning, during the automation workshop, I had the occasion of
> > speaking about this with Pingou and Threebean.
> > Thanks to Pingou hints, I've created a query to get pending ACL from
> > pkgdb.
> > What I'd like to share with you all is the list of users that can
> > approve/deny ACL requests (older than 1 month) but have not done it
> > yet (the number refers to the number of ACL pending).
> >
> > I think that people should take care of the pending ACL they can
> > approve/deny and actually approve or deny them ASAP.
>
> Your email needs a "call to action" link, otherwise no one will know
> what they are supposed to do about it. In this case it's probably:
>
> https://admin.fedoraproject.org/pkgdb/acl/pending/
>
> However I visited the above URL, logged in, and it says:
>
> Pending ACLs
> No pending ACLs for you
also there should be no action required from the owner for "watchcommit"
or "watchbugzilla" requests, looks as a bug in the conversion when
deploying the recent pkgdb
Well, the recent pkgdb is already quite a bit old and it should definitively
auto-accept the watch* ACLs.
Do you have a link so I could look at the package/history?
Thanks,
Pierre
8:34 a.m.
I have a strong opinion over this
All the ACL's should be accepted, doesn't matter the level.
And why i think of this ?
Two simple reasons:
- The packager abandoned the package, because several reasons, and then is
far away from Fedora systems for some time
- The packager is actively using Fedora, but seen not care to even properly
take care of his package, not in the minimal sense to deny the ACL, which
would be acceptable.
In both cases, the package became hostage to someone that for sure aren't
caring much for the distro, unless prove me wrong.
On Fri, Aug 5, 2016 at 3:07 PM, Pierre-Yves Chibon <pingou(a)pingoured.fr>
wrote:
On Fri, Aug 05, 2016 at 11:14:02AM +0200, Dan Horák wrote:
> On Thu, 4 Aug 2016 20:41:56 +0100
> "Richard W.M. Jones" <rjones(a)redhat.com> wrote:
>
> > On Thu, Aug 04, 2016 at 04:43:40PM +0200, Fabio Alessandro Locati
> > wrote:
> > > Hi all,
> > >
> > > This morning, during the automation workshop, I had the occasion of
> > > speaking about this with Pingou and Threebean.
> > > Thanks to Pingou hints, I've created a query to get pending ACL from
> > > pkgdb.
> > > What I'd like to share with you all is the list of users that can
> > > approve/deny ACL requests (older than 1 month) but have not done it
> > > yet (the number refers to the number of ACL pending).
> > >
> > > I think that people should take care of the pending ACL they can
> > > approve/deny and actually approve or deny them ASAP.
> >
> > Your email needs a "call to action" link, otherwise no one will know
> > what they are supposed to do about it. In this case it's probably:
> >
> > https://admin.fedoraproject.org/pkgdb/acl/pending/
> >
> > However I visited the above URL, logged in, and it says:
> >
> > Pending ACLs
> > No pending ACLs for you
>
> also there should be no action required from the owner for "watchcommit"
> or "watchbugzilla" requests, looks as a bug in the conversion when
> deploying the recent pkgdb
Well, the recent pkgdb is already quite a bit old and it should
definitively
auto-accept the watch* ACLs.
Do you have a link so I could look at the package/history?
Thanks,
Pierre
--
devel mailing list
devel(a)lists.fedoraproject.org
https://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org
9:24 a.m.
On Fri, Aug 05, 2016 at 03:34:58PM +0200, Helio Chissini de Castro wrote:
I have a strong opinion over this
All the ACL's should be accepted, doesn't matter the level.
And why i think of this ?
Two simple reasons:
- The packager abandoned the package, because several reasons, and then is
far away from Fedora systems for some time
- The packager is actively using Fedora, but seen not care to even properly
take care of his package, not in the minimal sense to deny the ACL, which
would be acceptable.
In both cases, the package became hostage to someone that for sure aren't
caring much for the distro, unless prove me wrong.
Hi,
On one hand I agree with you, on the other hand I do think that
sometimes pkgdb notifications are not so visible and therefore someone
could inadvertly loose track of some notifications and with your
proposal, this would trigger an auto-approval.
Best,
Fale
--
Fabio Alessandro Locati
Red Hat - Senior Consultant
PGP Fingerprint: E815 3C49 2A8D FD8B 1CBD BC85 FDB3 DF20 B2DC 9C1B
9:26 a.m.
On Sat, Aug 6, 2016 at 10:24 AM, Fabio Alessandro Locati
<fale(a)redhat.com> wrote:
On Fri, Aug 05, 2016 at 03:34:58PM +0200, Helio Chissini de Castro
wrote:
> I have a strong opinion over this
>
> All the ACL's should be accepted, doesn't matter the level.
> And why i think of this ?
>
> Two simple reasons:
> - The packager abandoned the package, because several reasons, and then is
> far away from Fedora systems for some time
> - The packager is actively using Fedora, but seen not care to even properly
> take care of his package, not in the minimal sense to deny the ACL, which
> would be acceptable.
>
> In both cases, the package became hostage to someone that for sure aren't
> caring much for the distro, unless prove me wrong.
Hi,
On one hand I agree with you, on the other hand I do think that
sometimes pkgdb notifications are not so visible and therefore someone
could inadvertly loose track of some notifications and with your
proposal, this would trigger an auto-approval.
The last time someone requested ACLs from one of my packages, I
received an email about it. So if they're not reading email about
their package, then that's a problem.
--
真実はいつも一つ！/ Always, there's only one truth!
11:25 a.m.
On Sat, Aug 06, 2016 at 10:26:36AM -0400, Neal Gompa wrote:
On Sat, Aug 6, 2016 at 10:24 AM, Fabio Alessandro Locati
<fale(a)redhat.com> wrote:
> On Fri, Aug 05, 2016 at 03:34:58PM +0200, Helio Chissini de Castro wrote:
>> I have a strong opinion over this
>>
>> All the ACL's should be accepted, doesn't matter the level.
>> And why i think of this ?
>>
>> Two simple reasons:
>> - The packager abandoned the package, because several reasons, and then is
>> far away from Fedora systems for some time
>> - The packager is actively using Fedora, but seen not care to even properly
>> take care of his package, not in the minimal sense to deny the ACL, which
>> would be acceptable.
>>
>> In both cases, the package became hostage to someone that for sure aren't
>> caring much for the distro, unless prove me wrong.
>
> Hi,
>
> On one hand I agree with you, on the other hand I do think that
> sometimes pkgdb notifications are not so visible and therefore someone
> could inadvertly loose track of some notifications and with your
> proposal, this would trigger an auto-approval.
The last time someone requested ACLs from one of my packages, I
received an email about it. So if they're not reading email about
their package, then that's a problem.
Sometimes an email could get missplaced. I think that at least an
(additional) weekly recap email should be sent (only to people with
pending ACL approval).
Best,
Fale
--
Fabio Alessandro Locati
Red Hat - Senior Consultant
PGP Fingerprint: E815 3C49 2A8D FD8B 1CBD BC85 FDB3 DF20 B2DC 9C1B
12:37 p.m.
Sometimes a maintainer doesn't want to approve ACLs for "reasons" but
doesn't want to reject the request for various reasons including the requestor
re-request of denied requests.
Dennis
On August 5, 2016 3:34:58 PM GMT+02:00, Helio Chissini de Castro <helio(a)kde.org>
wrote:
I have a strong opinion over this
All the ACL's should be accepted, doesn't matter the level.
And why i think of this ?
Two simple reasons:
- The packager abandoned the package, because several reasons, and then
is
far away from Fedora systems for some time
- The packager is actively using Fedora, but seen not care to even
properly
take care of his package, not in the minimal sense to deny the ACL,
which
would be acceptable.
In both cases, the package became hostage to someone that for sure
aren't
caring much for the distro, unless prove me wrong.
On Fri, Aug 5, 2016 at 3:07 PM, Pierre-Yves Chibon
<pingou(a)pingoured.fr>
wrote:
> On Fri, Aug 05, 2016 at 11:14:02AM +0200, Dan Horák wrote:
> > On Thu, 4 Aug 2016 20:41:56 +0100
> > "Richard W.M. Jones" <rjones(a)redhat.com> wrote:
> >
> > > On Thu, Aug 04, 2016 at 04:43:40PM +0200, Fabio Alessandro Locati
> > > wrote:
> > > > Hi all,
> > > >
> > > > This morning, during the automation workshop, I had the
occasion of
> > > > speaking about this with Pingou and Threebean.
> > > > Thanks to Pingou hints, I've created a query to get pending ACL
from
> > > > pkgdb.
> > > > What I'd like to share with you all is the list of users that
can
> > > > approve/deny ACL requests (older than 1 month) but have not
done it
> > > > yet (the number refers to the number of ACL pending).
> > > >
> > > > I think that people should take care of the pending ACL they
can
> > > > approve/deny and actually approve or deny them ASAP.
> > >
> > > Your email needs a "call to action" link, otherwise no one will
know
> > > what they are supposed to do about it. In this case it's
probably:
> > >
> > > https://admin.fedoraproject.org/pkgdb/acl/pending/
> > >
> > > However I visited the above URL, logged in, and it says:
> > >
> > > Pending ACLs
> > > No pending ACLs for you
> >
> > also there should be no action required from the owner for
"watchcommit"
> > or "watchbugzilla" requests, looks as a bug in the conversion when
> > deploying the recent pkgdb
>
> Well, the recent pkgdb is already quite a bit old and it should
> definitively
> auto-accept the watch* ACLs.
> Do you have a link so I could look at the package/history?
>
> Thanks,
> Pierre
> --
> devel mailing list
> devel(a)lists.fedoraproject.org
>
https://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org
>
------------------------------------------------------------------------
--
devel mailing list
devel(a)lists.fedoraproject.org
https://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org
--
Sent from my Android device with K-9 Mail. Please excuse my brevity.
11:15 p.m.
Hi,
On Sat, Aug 6, 2016 at 11:07 PM, <dennis(a)ausil.us> wrote:
Sometimes a maintainer doesn't want to approve ACLs for
"reasons" but
doesn't want to reject the request for various reasons including the
requestor re-request of denied requests.
To add few things here,
1) Why do we need such "Pending ACLs" weekly emails? The package owner
already gets notifications when someone requests ACLs. Now if you say
people forgets then why not implement something in pkgdb only which
will keep sending weekly notification again to package owners that
someone has requested ACLs on their packages? What benefit will others
get by reading these emails that someone requested ACLs on some
package and its owner still not approving?
2) Sometime back I don't remember if I filed issue against pkgdb but I
discussed on IRC why not add some text entry box while requesting
ACLs. Not everyone knows everyother one here. Let the package owner
know why the ACL requestor need other package access. That will help
package owners to decide to approve or deny quickly. There can be some
people who want to have access/co-maintainer for some packages where
they really not needed to be.
3) I am not sure but I think there is some automation for something in
pkgdb which automatically grants package access to requstor if package
owner do not take action. Correct me if I am wrong here.
Regards,
Parag.
4:39 a.m.
On Sun, Aug 07, 2016 at 09:45:57AM +0530, Parag Nemade wrote:
Hi,
On Sat, Aug 6, 2016 at 11:07 PM, <dennis(a)ausil.us> wrote:
> Sometimes a maintainer doesn't want to approve ACLs for "reasons" but
> doesn't want to reject the request for various reasons including the
> requestor re-request of denied requests.
>
2) Sometime back I don't remember if I filed issue against pkgdb but I
discussed on IRC why not add some text entry box while requesting
ACLs. Not everyone knows everyother one here. Let the package owner
know why the ACL requestor need other package access. That will help
package owners to decide to approve or deny quickly. There can be some
people who want to have access/co-maintainer for some packages where
they really not needed to be.
There is a ticket: https://github.com/fedora-infra/pkgdb2/issues/307
3) I am not sure but I think there is some automation for something
in
pkgdb which automatically grants package access to requstor if package
owner do not take action. Correct me if I am wrong here.
Not for ACLs, this is only when someone requests a new branch for a package that
you have commit on while they do not.
This way, if the current maintainers do not either approve or block the request,
it will automatically be sent to rel-eng that will check if it is possible to
add the branch and do it if it is.
Pierre
3:36 a.m.
Not works.
You need then talk with the requester and then decide to deny or not.
And if he tries again, do the process again.
What you can' t do is letting anyone waiting or making the package hostage.
About the notifications, yes, everyone get notifications if set your email
properly or use @fedoraproject, and is working well as delivering.
So increase the amount of notifications or do it in public lists will not
fix the issue that is really is, the package
owner. Would even piss of more the maintainers or other users ( in case of
public )
If he decide to ignore or just not reading, then all efforts to " increase"
or " improve" notifications system is useless.
Packages should have at least two lead owners and one group that can take
equal decisions over it.
On Sat, Aug 6, 2016 at 7:37 PM, <dennis(a)ausil.us> wrote:
Sometimes a maintainer doesn't want to approve ACLs for
"reasons" but
doesn't want to reject the request for various reasons including the
requestor re-request of denied requests.
Dennis
On August 5, 2016 3:34:58 PM GMT+02:00, Helio Chissini de Castro <
helio(a)kde.org> wrote:
>
> I have a strong opinion over this
>
> All the ACL's should be accepted, doesn't matter the level.
> And why i think of this ?
>
> Two simple reasons:
> - The packager abandoned the package, because several reasons, and then
> is far away from Fedora systems for some time
> - The packager is actively using Fedora, but seen not care to even
> properly take care of his package, not in the minimal sense to deny the
> ACL, which would be acceptable.
>
> In both cases, the package became hostage to someone that for sure aren't
> caring much for the distro, unless prove me wrong.
>
>
>
> On Fri, Aug 5, 2016 at 3:07 PM, Pierre-Yves Chibon <pingou(a)pingoured.fr>
> wrote:
>
>> On Fri, Aug 05, 2016 at 11:14:02AM +0200, Dan Horák wrote:
>> > On Thu, 4 Aug 2016 20:41:56 +0100
>> > "Richard W.M. Jones" <rjones(a)redhat.com> wrote:
>> >
>> > > On Thu, Aug 04, 2016 at 04:43:40PM +0200, Fabio Alessandro Locati
>> > > wrote:
>> > > > Hi all,
>> > > >
>> > > > This morning, during the automation workshop, I had the occasion
of
>> > > > speaking about this with Pingou and Threebean.
>> > > > Thanks to Pingou hints, I've created a query to get pending
ACL
>> from
>> > > > pkgdb.
>> > > > What I'd like to share with you all is the list of users that
can
>> > > > approve/deny ACL requests (older than 1 month) but have not done
it
>> > > > yet (the number refers to the number of ACL pending).
>> > > >
>> > > > I think that people should take care of the pending ACL they can
>> > > > approve/deny and actually approve or deny them ASAP.
>> > >
>> > > Your email needs a "call to action" link, otherwise no one
will know
>> > > what they are supposed to do about it. In this case it's
probably:
>> > >
>> > > https://admin.fedoraproject.org/pkgdb/acl/pending/
>> > >
>> > > However I visited the above URL, logged in, and it says:
>> > >
>> > > Pending ACLs
>> > > No pending ACLs for you
>> >
>> > also there should be no action required from the owner for
>> "watchcommit"
>> > or "watchbugzilla" requests, looks as a bug in the conversion
when
>> > deploying the recent pkgdb
>>
>> Well, the recent pkgdb is already quite a bit old and it should
>> definitively
>> auto-accept the watch* ACLs.
>> Do you have a link so I could look at the package/history?
>>
>> Thanks,
>> Pierre
>> --
>> devel mailing list
>> devel(a)lists.fedoraproject.org
>> https://lists.fedoraproject.org/admin/lists/devel@lists.fedo
>> raproject.org
>>
>
> --
> devel mailing list
> devel(a)lists.fedoraproject.org
> https://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org
>
>
--
Sent from my Android device with K-9 Mail. Please excuse my brevity.
10:34 a.m.
On Monday, August 8, 2016 10:36:11 AM CDT Helio Chissini de Castro wrote:
Not works.
I do not understand what does not work
You need then talk with the requester and then decide to deny or
not.
And if he tries again, do the process again.
What you can' t do is letting anyone waiting or making the package hostage.
Sometimes talking does not get anywhere and you reach an impase and you should
just leave the acls be pending. The package is not held hostage in anyway
shape or form. I honestly think you are attacking this from the wrong angle.
About the notifications, yes, everyone get notifications if set your
email
properly or use @fedoraproject, and is working well as delivering.
So increase the amount of notifications or do it in public lists will not
fix the issue that is really is, the package
owner. Would even piss of more the maintainers or other users ( in case of
public )
not sure what this is dirrected at
If he decide to ignore or just not reading, then all efforts to " increase"
or " improve" notifications system is useless.
Packages should have at least two lead owners and one group that can take
equal decisions over it.
why? please give some reasoning
On Sat, Aug 6, 2016 at 7:37 PM, <dennis(a)ausil.us> wrote:
> Sometimes a maintainer doesn't want to approve ACLs for "reasons" but
> doesn't want to reject the request for various reasons including the
> requestor re-request of denied requests.
>
> Dennis
>
> On August 5, 2016 3:34:58 PM GMT+02:00, Helio Chissini de Castro <
>
> helio(a)kde.org> wrote:
>> I have a strong opinion over this
>>
>> All the ACL's should be accepted, doesn't matter the level.
>> And why i think of this ?
>>
>> Two simple reasons:
>> - The packager abandoned the package, because several reasons, and then
>> is far away from Fedora systems for some time
>> - The packager is actively using Fedora, but seen not care to even
>> properly take care of his package, not in the minimal sense to deny the
>> ACL, which would be acceptable.
>>
>> In both cases, the package became hostage to someone that for sure aren't
>> caring much for the distro, unless prove me wrong.
>>
>>
>>
>> On Fri, Aug 5, 2016 at 3:07 PM, Pierre-Yves Chibon <pingou(a)pingoured.fr>
>>
>> wrote:
>>> On Fri, Aug 05, 2016 at 11:14:02AM +0200, Dan Horák wrote:
>>> > On Thu, 4 Aug 2016 20:41:56 +0100
>>> >
>>> > "Richard W.M. Jones" <rjones(a)redhat.com> wrote:
>>> > > On Thu, Aug 04, 2016 at 04:43:40PM +0200, Fabio Alessandro Locati
>>> > >
>>> > > wrote:
>>> > > > Hi all,
>>> > > >
>>> > > > This morning, during the automation workshop, I had the
occasion
>>> > > > of
>>> > > > speaking about this with Pingou and Threebean.
>>> > > > Thanks to Pingou hints, I've created a query to get
pending ACL
>>>
>>> from
>>>
>>> > > > pkgdb.
>>> > > > What I'd like to share with you all is the list of users
that can
>>> > > > approve/deny ACL requests (older than 1 month) but have not
done
>>> > > > it
>>> > > > yet (the number refers to the number of ACL pending).
>>> > > >
>>> > > > I think that people should take care of the pending ACL they
can
>>> > > > approve/deny and actually approve or deny them ASAP.
>>> > >
>>> > > Your email needs a "call to action" link, otherwise no
one will know
>>> > >
>>> > > what they are supposed to do about it. In this case it's
probably:
>>> > > https://admin.fedoraproject.org/pkgdb/acl/pending/
>>> > >
>>> > > However I visited the above URL, logged in, and it says:
>>> > > Pending ACLs
>>> > > No pending ACLs for you
>>> >
>>> > also there should be no action required from the owner for
>>>
>>> "watchcommit"
>>>
>>> > or "watchbugzilla" requests, looks as a bug in the conversion
when
>>> > deploying the recent pkgdb
>>>
>>> Well, the recent pkgdb is already quite a bit old and it should
>>> definitively
>>> auto-accept the watch* ACLs.
>>> Do you have a link so I could look at the package/history?
>>>
>>> Thanks,
>>> Pierre
>>> --
>>> devel mailing list
>>> devel(a)lists.fedoraproject.org
>>> https://lists.fedoraproject.org/admin/lists/devel@lists.fedo
>>> raproject.org
>>
>> --
>> devel mailing list
>> devel(a)lists.fedoraproject.org
>> https://lists.fedoraproject.org/admin/lists/devel@lists.fedoraproject.org
>
> --
> Sent from my Android device with K-9 Mail. Please excuse my brevity.
6:33 a.m.
First of all sorry if i looks like rude on my answers, was not the
intention.
On Tue, Aug 9, 2016 at 5:34 PM, Dennis Gilmore <dennis(a)ausil.us> wrote:
On Monday, August 8, 2016 10:36:11 AM CDT Helio Chissini de Castro
wrote:
> Not works.
I do not understand what does not work
> You need then talk with the requester and then decide to deny or not.
> And if he tries again, do the process again.
> What you can' t do is letting anyone waiting or making the package
hostage.
Sometimes talking does not get anywhere and you reach an impase and you
should
just leave the acls be pending. The package is not held hostage in anyway
shape or form. I honestly think you are attacking this from the wrong
angle.
I know, but still, i strongly believe that if this deadlock situation
happens, then i think
is ok to deny it with a polite reasoning that we're not ready to agree on
rights over the package.
We can change our mind later, or even the requester decides not going
further, but it's life.
We just can't let things hanging.
> About the notifications, yes, everyone get notifications if set
your
email
> properly or use @fedoraproject, and is working well as delivering.
> So increase the amount of notifications or do it in public lists will not
> fix the issue that is really is, the package
> owner. Would even piss of more the maintainers or other users ( in case
of
> public )
not sure what this is dirrected at
Is just because people complaining that we need improve the package acls
notification, but more notifications
would not solve the issue, would just be "more"
I can be wrong on that and something really was not efficient, but i
personally would not like to
have more and more regular notifications beyond the first one.
>
> If he decide to ignore or just not reading, then all efforts to "
increase"
> or " improve" notifications system is useless.
>
> Packages should have at least two lead owners and one group that can take
> equal decisions over it.
why? please give some reasoning
The reasoning on have two maintainers is, unless we have someone paid for
maintaining the packages, the reality is that we're
community bounded. And sometime the so called "life" will take our free
time for do this things, or any other reasoning that could
prevent us to get active more often.
With two persons, would be easy to avoid a package became hostage in
general and then have all the unresponsive maintainers requests
we've been seen often.
But then, the two can disappear as well, so that's comes from the groups.
The group is the more important thing, since give powers over a package
with some trusted people, not specific one, so it can be moving from time
to
time the crew, and never been a hostage package for a single point of
contact.
KDE and Qt packages are this way, they are not my or any other
responsibility alone, but to @kdesig group, and any guys that enters in
this group
are effectively a new admin to the packages, granted as is in the group and
we trusted to be working with us.
8:20 a.m.
On Fri, Aug 05, 2016 at 11:14:02AM +0200, Dan Horák wrote:
On Thu, 4 Aug 2016 20:41:56 +0100
"Richard W.M. Jones" <rjones(a)redhat.com> wrote:
> Your email needs a "call to action" link, otherwise no one will know
> what they are supposed to do about it. In this case it's probably:
>
> https://admin.fedoraproject.org/pkgdb/acl/pending/
>
> However I visited the above URL, logged in, and it says:
Thanks, I'll add a call to action next time.
also there should be no action required from the owner for
"watchcommit"
or "watchbugzilla" requests, looks as a bug in the conversion when
deploying the recent pkgdb
This is true. About 25% of the ACL are in this situation. I'll update my
database and in a couple of days run it again removing those.
Cheers,
Fale
--
Fabio Alessandro Locati
Red Hat - Senior Consultant
PGP Fingerprint: E815 3C49 2A8D FD8B 1CBD BC85 FDB3 DF20 B2DC 9C1B
9:48 a.m.
On Thu, Aug 04, 2016 at 08:41:56PM +0100, Richard W.M. Jones wrote:
Your email needs a "call to action" link, otherwise no one
will know
what they are supposed to do about it. In this case it's probably:
https://admin.fedoraproject.org/pkgdb/acl/pending/
However I visited the above URL, logged in, and it says:
Pending ACLs
No pending ACLs for you
So I guess this is wrong or perhaps refers to something else:
...
> 3 rjones
...
Hi Richard,
It seems like pkgdb is hiding those ACL requests on the page you linked.
It seems like the following requests could/should be approved by you:
requester | req_acl | package | distro | version | approver
-----------+-------------+-----------------+--------+---------+----------
epienbro | commit | mingw32-gtk-vnc | Fedora | devel | rjones
epienbro | approveacls | mingw32-gtk-vnc | Fedora | devel | rjones
ktietz | approveacls | mingw32-openssl | Fedora | devel | rjones
Best,
Fale
--
Fabio Alessandro Locati
Red Hat - Senior Consultant
PGP Fingerprint: E815 3C49 2A8D FD8B 1CBD BC85 FDB3 DF20 B2DC 9C1B
10:36 a.m.
On Sat, Aug 06, 2016 at 03:48:36PM +0100, Fabio Alessandro Locati wrote:
On Thu, Aug 04, 2016 at 08:41:56PM +0100, Richard W.M. Jones wrote:
> Your email needs a "call to action" link, otherwise no one will know
> what they are supposed to do about it. In this case it's probably:
>
> https://admin.fedoraproject.org/pkgdb/acl/pending/
>
> However I visited the above URL, logged in, and it says:
>
> Pending ACLs
> No pending ACLs for you
>
> So I guess this is wrong or perhaps refers to something else:
>
> ...
> > 3 rjones
> ...
Hi Richard,
It seems like pkgdb is hiding those ACL requests on the page you linked.
It seems like the following requests could/should be approved by you:
requester | req_acl | package | distro | version | approver
-----------+-------------+-----------------+--------+---------+----------
epienbro | commit | mingw32-gtk-vnc | Fedora | devel | rjones
epienbro | approveacls | mingw32-gtk-vnc | Fedora | devel | rjones
ktietz | approveacls | mingw32-openssl | Fedora | devel | rjones
I've checked again just now, and I still don't see those ACLs.
It still says:
Pending ACLs
No pending ACLs for you
Copyright © 2013-2016 Red Hat pkgdb2 -- 2.4.3 -- Documentation -- API
Rich.
--
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
Fedora Windows cross-compiler. Compile Windows programs, test, and
build Windows installers. Over 100 libraries supported.
http://fedoraproject.org/wiki/MinGW
11:23 a.m.
On Sat, Aug 06, 2016 at 04:36:34PM +0100, Richard W.M. Jones wrote:
I've checked again just now, and I still don't see those
ACLs.
My guess is that it's a PkgDB bug and for this reason I'm trying to get
in touch with Pingou.
Cheers,
Fale
--
Fabio Alessandro Locati
Red Hat - Senior Consultant
PGP Fingerprint: E815 3C49 2A8D FD8B 1CBD BC85 FDB3 DF20 B2DC 9C1B
12:09 p.m.
On Sat, 6 Aug 2016 17:23:31 +0100
Fabio Alessandro Locati <fale(a)redhat.com> wrote:
On Sat, Aug 06, 2016 at 04:36:34PM +0100, Richard W.M. Jones wrote:
> I've checked again just now, and I still don't see those ACLs.
My guess is that it's a PkgDB bug and for this reason I'm trying to
get in touch with Pingou.
Note that he is traveling back from flock now, so I'm sure he will get
in contact once he is back home. :)
kevin
12:35 p.m.
On 08/06/2016 05:36 PM, Richard W.M. Jones wrote:
On Sat, Aug 06, 2016 at 03:48:36PM +0100, Fabio Alessandro Locati
wrote:
> On Thu, Aug 04, 2016 at 08:41:56PM +0100, Richard W.M. Jones wrote:
>> Your email needs a "call to action" link, otherwise no one will know
>> what they are supposed to do about it. In this case it's probably:
>>
>> https://admin.fedoraproject.org/pkgdb/acl/pending/
>>
>> However I visited the above URL, logged in, and it says:
>>
>> Pending ACLs
>> No pending ACLs for you
>>
>> So I guess this is wrong or perhaps refers to something else:
>>
>> ...
>>> 3 rjones
>> ...
>
> Hi Richard,
>
> It seems like pkgdb is hiding those ACL requests on the page you linked.
> It seems like the following requests could/should be approved by you:
>
> requester | req_acl | package | distro | version | approver
> -----------+-------------+-----------------+--------+---------+----------
> epienbro | commit | mingw32-gtk-vnc | Fedora | devel | rjones
> epienbro | approveacls | mingw32-gtk-vnc | Fedora | devel | rjones
> ktietz | approveacls | mingw32-openssl | Fedora | devel | rjones
I've checked again just now, and I still don't see those ACLs.
It still says:
Pending ACLs
No pending ACLs for you
We retired all mingw32- packages a few years ago and renamed them to
mingw-, so I think pkgdb is correct here to not show these to you.
--
Kalev
4:47 p.m.
On Sat, Aug 06, 2016 at 07:35:04PM +0200, Kalev Lember wrote:
On 08/06/2016 05:36 PM, Richard W.M. Jones wrote:
> On Sat, Aug 06, 2016 at 03:48:36PM +0100, Fabio Alessandro Locati wrote:
>> On Thu, Aug 04, 2016 at 08:41:56PM +0100, Richard W.M. Jones wrote:
>>> Your email needs a "call to action" link, otherwise no one will
know
>>> what they are supposed to do about it. In this case it's probably:
>>>
>>> https://admin.fedoraproject.org/pkgdb/acl/pending/
>>>
>>> However I visited the above URL, logged in, and it says:
>>>
>>> Pending ACLs
>>> No pending ACLs for you
>>>
>>> So I guess this is wrong or perhaps refers to something else:
>>>
>>> ...
>>>> 3 rjones
>>> ...
>>
>> Hi Richard,
>>
>> It seems like pkgdb is hiding those ACL requests on the page you linked.
>> It seems like the following requests could/should be approved by you:
>>
>> requester | req_acl | package | distro | version | approver
>> -----------+-------------+-----------------+--------+---------+----------
>> epienbro | commit | mingw32-gtk-vnc | Fedora | devel | rjones
>> epienbro | approveacls | mingw32-gtk-vnc | Fedora | devel | rjones
>> ktietz | approveacls | mingw32-openssl | Fedora | devel | rjones
>
> I've checked again just now, and I still don't see those ACLs.
> It still says:
>
> Pending ACLs
>
> No pending ACLs for you
We retired all mingw32- packages a few years ago and renamed them to
mingw-, so I think pkgdb is correct here to not show these to you.
Ah thanks Kalev, that makes sense.
Perhaps pkgdb should reap these retired packages (although as it is
hiding them, that could be correct behaviour already).
Rich.
--
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
virt-df lists disk usage of guests without needing to install any
software inside the virtual machine. Supports Linux and Windows.
http://people.redhat.com/~rjones/virt-df/
2860
days inactive
2866
days old
20 comments
10 participants
participants (10)
-
Dan Horák -
Dennis Gilmore -
Fabio Alessandro Locati -
Helio Chissini de Castro -
Kalev Lember -
Kevin Fenzi -
Neal Gompa -
Parag Nemade -
Pierre-Yves Chibon -
Richard W.M. Jones