Unexpected rebase of libsolv to 0.7.1 in F29, F28; please report any issues to bugzilla

[Jaroslav Mracek]

Hello everyone,

There was an announcement of release libsolv-0.7.0 ([HEADS UP] libsolv 0.7)
into rawhide, but the rebase also ended up in stable branches of Fedora 28
and 29. This release could affect not only libsolv users, but also libdnf,
PackageKit, microdnf, or dnf related applications.
I would like to ask everyone for intensive testing and reporting any issues
concerning the rebase.

Thanks a lot for your help

Jaroslav
on behalf of DNF team

[Peter Robinson]

On Mon, Nov 12, 2018 at 3:35 PM Jaroslav Mracek <jmracek(a)redhat.com&gt; wrote:
>  Hello everyone,
>
>  There was an announcement of release libsolv-0.7.0 ([HEADS UP] libsolv 0.7) into rawhide,
> but the rebase also ended up in stable branches of Fedora 28 and 29. This release could
> affect not only libsolv users, but also libdnf, PackageKit, microdnf, or dnf related
> applications.
>  I would like to ask everyone for intensive testing and reporting any issues concerning
> the rebase. 
How did this this happen? It's kind of strange that people weren't
aware this was happening, what is some auto "git merge master"
mistake. It's a fairly big problem to "accidentally" rebase to a major
new release and not realise it was happening, especially on something
so core as core updates infrastructure. What sort of things are you
going to put in place to ensure random rebases don't just happen
again?

Peter

[Neal Gompa]

On Tue, Nov 13, 2018 at 1:42 PM Peter Robinson <pbrobinson(a)gmail.com&gt; wrote:
>  On Mon, Nov 12, 2018 at 3:35 PM Jaroslav Mracek <jmracek(a)redhat.com&gt; wrote:
>  >
>  > Hello everyone,
>  >
>  > There was an announcement of release libsolv-0.7.0 ([HEADS UP] libsolv 0.7) into
> rawhide, but the rebase also ended up in stable branches of Fedora 28 and 29. This release
> could affect not only libsolv users, but also libdnf, PackageKit, microdnf, or dnf related
> applications.
>  > I would like to ask everyone for intensive testing and reporting any issues
> concerning the rebase.
>
>  How did this this happen? It's kind of strange that people weren't
>  aware this was happening, what is some auto "git merge master"
>  mistake. It's a fairly big problem to "accidentally" rebase to a major
>  new release and not realise it was happening, especially on something
>  so core as core updates infrastructure. What sort of things are you
>  going to put in place to ensure random rebases don't just happen
>  again? 
It wasn't a random rebase. A FESCo ticket was submitted and
approved[1]. However, there was a miscommunication that led to the DNF
team not being aware it happened.

[1]: https://pagure.io/fesco/issue/2009



-- 
真実はいつも一つ！/ Always, there's only one truth!

[Randy Barlow]

On Tue, 2018-11-13 at 13:43 -0500, Neal Gompa wrote:
>  It wasn't a random rebase. A FESCo ticket was submitted and
>  approved[1]. However, there was a miscommunication that led to the
>  DNF
>  team not being aware it happened.
>  
>  [1]: https://pagure.io/fesco/issue/2009 
This was not approved - there was a -1 vote and so it was planned to be
discussed in the next meeting.

[Igor Gnatenko]

On Tue, Nov 13, 2018 at 8:49 PM Randy Barlow
<bowlofeggs(a)fedoraproject.org&gt; wrote:
>  On Tue, 2018-11-13 at 13:43 -0500, Neal Gompa wrote:
>  > It wasn't a random rebase. A FESCo ticket was submitted and
>  > approved[1]. However, there was a miscommunication that led to the
>  > DNF
>  > team not being aware it happened.
>  >
>  > [1]: https://pagure.io/fesco/issue/2009
>
>  This was not approved - there was a -1 vote and so it was planned to be
>  discussed in the next meeting. 
I commented the ticket, but I will copy my response here: there was no
single -1 within a week after opening a ticket so to my knowledge the
ticket was approved.

> _______________________________________________
> devel mailing list -- devel(a)lists.fedoraproject.org
> To unsubscribe send an email to devel-leave(a)lists.fedoraproject.org
> Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
> List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
> List Archives:
https://lists.fedoraproject.org/archives/list/[e-mail]

[Panu Matilainen]

On 11/13/18 10:24 PM, Igor Gnatenko wrote:
>  On Tue, Nov 13, 2018 at 8:49 PM Randy Barlow
>  <bowlofeggs(a)fedoraproject.org&gt; wrote:
> >
> > On Tue, 2018-11-13 at 13:43 -0500, Neal Gompa wrote:
> >> It wasn't a random rebase. A FESCo ticket was submitted and
> >> approved[1]. However, there was a miscommunication that led to the
> >> DNF
> >> team not being aware it happened.
> >>
> >> [1]: https://pagure.io/fesco/issue/2009
> >
> > This was not approved - there was a -1 vote and so it was planned to be
> > discussed in the next meeting.
>  
>  I commented the ticket, but I will copy my response here: there was no
>  single -1 within a week after opening a ticket so to my knowledge the
>  ticket was approved. 
Which is why you need to wait until the actual decision has been stated 
in the ticket.

	- Panu -

[Igor Gnatenko]

On Wed, Nov 14, 2018 at 10:14 AM Panu Matilainen <pmatilai(a)redhat.com&gt; wrote:
>  On 11/13/18 10:24 PM, Igor Gnatenko wrote:
>  > On Tue, Nov 13, 2018 at 8:49 PM Randy Barlow
>  > <bowlofeggs(a)fedoraproject.org&gt; wrote:
>  >>
>  >> On Tue, 2018-11-13 at 13:43 -0500, Neal Gompa wrote:
>  >>> It wasn't a random rebase. A FESCo ticket was submitted and
>  >>> approved[1]. However, there was a miscommunication that led to the
>  >>> DNF
>  >>> team not being aware it happened.
>  >>>
>  >>> [1]: https://pagure.io/fesco/issue/2009
>  >>
>  >> This was not approved - there was a -1 vote and so it was planned to be
>  >> discussed in the next meeting.
>  >
>  > I commented the ticket, but I will copy my response here: there was no
>  > single -1 within a week after opening a ticket so to my knowledge the
>  > ticket was approved.
>
>  Which is why you need to wait until the actual decision has been stated
>  in the ticket. 
https://docs.fedoraproject.org/en-US/fesco/#_ticket_policy says
"""
Once a ticket has a formal proposal offered, FESCo members have one
week to either vote for or against it or else propose the ticket for
the next weekly meeting agenda. At the end of that one week, if the
proposal has gained at least three "for" votes and no "against"
votes,
it is approved. Any "against" votes mean that it goes onto the next
meeting agenda. If the week passes and the required number of votes
have not been met, the proposal is extended by one further week and
the minimum requirement becomes a single positive "for" vote. This is
intended to ensure that proposals do not languish.
"""

* The ticket had formal proposal offered (means that FESCo members can vote)
* 6 FESCo members voted +1 within a week (which means that there were
at least three "for" votes and no "against" votes)

Considering 2 points above -- I can read that it is approved. Waiting
for anyone to put note that it is approved is nice, but not must. If
the policy is somehow different from what I have read, please update
docs.fp.o and announce it.

>
>         - Panu -
> _______________________________________________
> devel mailing list -- devel(a)lists.fedoraproject.org
> To unsubscribe send an email to devel-leave(a)lists.fedoraproject.org
> Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
> List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
> List Archives:
https://lists.fedoraproject.org/archives/list/[e-mail]

[Panu Matilainen]

On 11/14/18 12:56 PM, Igor Gnatenko wrote:
>  On Wed, Nov 14, 2018 at 10:14 AM Panu Matilainen
> <pmatilai(a)redhat.com&gt; wrote:
> >
> > On 11/13/18 10:24 PM, Igor Gnatenko wrote:
> >> On Tue, Nov 13, 2018 at 8:49 PM Randy Barlow
> >> <bowlofeggs(a)fedoraproject.org&gt; wrote:
> >>>
> >>> On Tue, 2018-11-13 at 13:43 -0500, Neal Gompa wrote:
> >>>> It wasn't a random rebase. A FESCo ticket was submitted and
> >>>> approved[1]. However, there was a miscommunication that led to the
> >>>> DNF
> >>>> team not being aware it happened.
> >>>>
> >>>> [1]: https://pagure.io/fesco/issue/2009
> >>>
> >>> This was not approved - there was a -1 vote and so it was planned to be
> >>> discussed in the next meeting.
> >>
> >> I commented the ticket, but I will copy my response here: there was no
> >> single -1 within a week after opening a ticket so to my knowledge the
> >> ticket was approved.
> >
> > Which is why you need to wait until the actual decision has been stated
> > in the ticket.
>  
>  https://docs.fedoraproject.org/en-US/fesco/#_ticket_policy says
>  """
>  Once a ticket has a formal proposal offered, FESCo members have one
>  week to either vote for or against it or else propose the ticket for
>  the next weekly meeting agenda. At the end of that one week, if the
>  proposal has gained at least three "for" votes and no "against"
> votes,
>  it is approved. Any "against" votes mean that it goes onto the next
>  meeting agenda. If the week passes and the required number of votes
>  have not been met, the proposal is extended by one further week and
>  the minimum requirement becomes a single positive "for" vote. This is
>  intended to ensure that proposals do not languish.
>  """
>  
>  * The ticket had formal proposal offered (means that FESCo members can vote)
>  * 6 FESCo members voted +1 within a week (which means that there were
>  at least three "for" votes and no "against" votes)
>  
>  Considering 2 points above -- I can read that it is approved. Waiting
>  for anyone to put note that it is approved is nice, but not must. If
>  the policy is somehow different from what I have read, please update
>  docs.fp.o and announce it. 
I stand corrected then.

	- Panu -

[Randy Barlow]

On Wed, 2018-11-14 at 11:56 +0100, Igor Gnatenko wrote:
>  > > > This was not approved - there was a -1 vote and so it
> was
>  > > > planned to be
>  > > > discussed in the next meeting.
>  > > 
>  > > I commented the ticket, but I will copy my response here: there
>  > > was no
>  > > single -1 within a week after opening a ticket so to my knowledge
>  > > the
>  > > ticket was approved.
>  
>  * The ticket had formal proposal offered (means that FESCo members
>  can vote)
>  * 6 FESCo members voted +1 within a week (which means that there were
>  at least three "for" votes and no "against" votes)
>  
>  Considering 2 points above -- I can read that it is approved. Waiting
>  for anyone to put note that it is approved is nice, but not must. If
>  the policy is somehow different from what I have read, please update
>  docs.fp.o and announce it. 
I suppose that's a fair interpretation of the wording in the policy,
though it wasn't my personal interpretation based off of memory. I
think I would generally prefer that people wait for the ticket to
officially say it's approved, but you are right that the policy doesn't
explicitly state that.

In any case, fair enough, I retract my claim that it wasn't approved ☺

[Stephen Gallagher]

On Wed, Nov 14, 2018 at 11:44 AM Randy Barlow <bowlofeggs(a)fedoraproject.org&gt;
wrote:

>  On Wed, 2018-11-14 at 11:56 +0100, Igor Gnatenko wrote:
>  > > > > This was not approved - there was a -1 vote and so it was
>  > > > > planned to be
>  > > > > discussed in the next meeting.
>  > > >
>  > > > I commented the ticket, but I will copy my response here: there
>  > > > was no
>  > > > single -1 within a week after opening a ticket so to my knowledge
>  > > > the
>  > > > ticket was approved.
>  >
>  > * The ticket had formal proposal offered (means that FESCo members
>  > can vote)
>  > * 6 FESCo members voted +1 within a week (which means that there were
>  > at least three "for" votes and no "against" votes)
>  >
>  > Considering 2 points above -- I can read that it is approved. Waiting
>  > for anyone to put note that it is approved is nice, but not must. If
>  > the policy is somehow different from what I have read, please update
>  > docs.fp.o and announce it.
>
>  I suppose that's a fair interpretation of the wording in the policy,
>  though it wasn't my personal interpretation based off of memory. I
>  think I would generally prefer that people wait for the ticket to
>  officially say it's approved, but you are right that the policy doesn't
>  explicitly state that.
>
>  In any case, fair enough, I retract my claim that it wasn't approved ☺
>  
For the record, your memory is faulty. When we reworked the policy, we made
it explicit that once it passed that week, if it had at least +3 and no -1
votes, it was approved. This was to avoid FESCo's classic problem of taking
forever to reach a decision on things. After that point, it's pending an
*announcement*, but the ruling is official.

[Igor Gnatenko]

On Tue, Nov 13, 2018 at 7:49 PM Peter Robinson <pbrobinson(a)gmail.com&gt; wrote:
>  On Mon, Nov 12, 2018 at 3:35 PM Jaroslav Mracek <jmracek(a)redhat.com&gt; wrote:
>  >
>  > Hello everyone,
>  >
>  > There was an announcement of release libsolv-0.7.0 ([HEADS UP] libsolv 0.7) into
> rawhide, but the rebase also ended up in stable branches of Fedora 28 and 29. This release
> could affect not only libsolv users, but also libdnf, PackageKit, microdnf, or dnf related
> applications.
>  > I would like to ask everyone for intensive testing and reporting any issues
> concerning the rebase.
>
>  How did this this happen? It's kind of strange that people weren't
>  aware this was happening, what is some auto "git merge master"
>  mistake. It's a fairly big problem to "accidentally" rebase to a major
>  new release and not realise it was happening, especially on something
>  so core as core updates infrastructure. What sort of things are you
>  going to put in place to ensure random rebases don't just happen
>  again? 
It is not random rebase, there was even FESCo ticket.

> Peter
> _______________________________________________
> devel mailing list -- devel(a)lists.fedoraproject.org
> To unsubscribe send an email to devel-leave(a)lists.fedoraproject.org
> Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
> List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
> List Archives:
https://lists.fedoraproject.org/archives/list/[e-mail]

[Igor Gnatenko]

On Mon, Nov 12, 2018 at 4:45 PM Jaroslav Mracek <jmracek(a)redhat.com&gt; wrote:
>  Hello everyone,
>
>  There was an announcement of release libsolv-0.7.0 ([HEADS UP] libsolv 0.7) into rawhide,
> but the rebase also ended up in stable branches of Fedora 28 and 29. This release could
> affect not only libsolv users, but also libdnf, PackageKit, microdnf, or dnf related
> applications. 
libdnf, PK, microdnf and dnf **are** libsolv users.

>  I would like to ask everyone for intensive testing and reporting any
> issues concerning the rebase. 
There is nothing in Fedora which is using any of functionality which
was changed in incompatible way (except for zypper which we handled
carefully with Neal Gompa in the same update).

If there won't be SONAME change, it would be released as 0.6.36 and no
one would notice any changes after rebase.

> Thanks a lot for your help
>
> Jaroslav
> on behalf of DNF team
> _______________________________________________
> devel mailing list -- devel(a)lists.fedoraproject.org
> To unsubscribe send an email to devel-leave(a)lists.fedoraproject.org
> Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
> List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
> List Archives:
https://lists.fedoraproject.org/archives/list/[e-mail]

[Jonathan Underwood]

On Mon, 12 Nov 2018 at 15:35, Jaroslav Mracek <jmracek(a)redhat.com&gt; wrote:
>  Hello everyone,
>
>  There was an announcement of release libsolv-0.7.0 ([HEADS UP] libsolv 0.7) into rawhide,
> but the rebase also ended up in stable branches of Fedora 28 and 29. This release could
> affect not only libsolv users, but also libdnf, PackageKit, microdnf, or dnf related
> applications.
>  I would like to ask everyone for intensive testing and reporting any issues concerning
> the rebase.
>
>  Thanks a lot for your help 
What risk does this present for dnf upgrades from F28 to F29?

[Igor Gnatenko]

Absolutely no risk. It would be released as usual 0.6.37 (or number like
that) if not some ABI breakage due to removal of very old cruft.

On Wed, Nov 14, 2018, 18:50 Jonathan Underwood <jonathan.underwood(a)gmail.com
wrote:

>  On Mon, 12 Nov 2018 at 15:35, Jaroslav Mracek
> <jmracek(a)redhat.com&gt; wrote:
>  >
>  > Hello everyone,
>  >
>  > There was an announcement of release libsolv-0.7.0 ([HEADS UP] libsolv
>  0.7) into rawhide, but the rebase also ended up in stable branches of
>  Fedora 28 and 29. This release could affect not only libsolv users, but
>  also libdnf, PackageKit, microdnf, or dnf related applications.
>  > I would like to ask everyone for intensive testing and reporting any
>  issues concerning the rebase.
>  >
>  > Thanks a lot for your help
>
>  What risk does this present for dnf upgrades from F28 to F29?
>  _______________________________________________
>  devel mailing list -- devel(a)lists.fedoraproject.org
>  To unsubscribe send an email to devel-leave(a)lists.fedoraproject.org
>  Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
>  List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
>  List Archives:
>  https://lists.fedoraproject.org/archives/list/[e-mail]
>