You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@zookeeper.apache.org by Michael Han <ha...@cloudera.com> on 2017/04/18 17:23:16 UTC
FYI - Forced push on branch-3.4
I did a forced push on branch-3.4 this morning to remove a bad commit that
breaks the build of branch-3.4. I wasn't aware that forced push should be
avoided until Pat Hunt reminds me.
After force push I noticed some inconsistency state of top ref for
branch-3.4, and I am following up with INFRA on this [1].
Apologize for any inconvenience caused. Please let me know if you observe
other problems caused by the change of the state for branch-3.4.
[1] https://issues.apache.org/jira/browse/INFRA-13916
--
Cheers
Michael.
Re: FYI - Forced push on branch-3.4
Posted by Patrick Hunt <ph...@apache.org>.
Sounds reasonable to me, +1, thanks Michael.
Patrick
On Wed, Apr 19, 2017 at 6:35 PM, Michael Han <ha...@cloudera.com> wrote:
> INFRA-13916 <https://issues.apache.org/jira/browse/INFRA-13916> is
> resolved
> now.
>
> I am planning to file another INFRA JIRA to apply protection scheme on all
> branches of ZK git repo, because the force push should be an exception
> rather than default. If we do need force push we can always ask INFRA to
> temporarily open the gate. Please shout out if you have concerns on
> disabling force push on git repo - otherwise if I hear nothing by end of
> this week I'll let INFRA disable force push on ZK git repo.
>
>
> On Tue, Apr 18, 2017 at 1:47 PM, Michael Han <ha...@cloudera.com> wrote:
>
> > Another more recent issue related to force push and Hadoop:
> > https://issues.apache.org/jira/browse/INFRA-13902
> >
> > I think ZooKeeper git project should be configured to disable force push
> > by default for all branches. I believe the force push was not currently
> > configured as disabled otherwise I would not succeed on force pushing
> which
> > leads to an observable inconsistent state for branch-3.4. I'll follow up
> > with INFRA once we restore the linearizability of branch-3.4.
> >
> >
> >
> > On Tue, Apr 18, 2017 at 12:40 PM, Patrick Hunt <ph...@apache.org> wrote:
> >
> >> fwiw some background on what other teams have done:
> >> https://issues.apache.org/jira/browse/INFRA-11236
> >>
> >> Patrick
> >>
> >> On Tue, Apr 18, 2017 at 10:23 AM, Michael Han <ha...@cloudera.com>
> wrote:
> >>
> >> > I did a forced push on branch-3.4 this morning to remove a bad commit
> >> that
> >> > breaks the build of branch-3.4. I wasn't aware that forced push should
> >> be
> >> > avoided until Pat Hunt reminds me.
> >> > After force push I noticed some inconsistency state of top ref for
> >> > branch-3.4, and I am following up with INFRA on this [1].
> >> >
> >> > Apologize for any inconvenience caused. Please let me know if you
> >> observe
> >> > other problems caused by the change of the state for branch-3.4.
> >> >
> >> > [1] https://issues.apache.org/jira/browse/INFRA-13916
> >> > --
> >> > Cheers
> >> > Michael.
> >> >
> >>
> >
> >
> >
> > --
> > Cheers
> > Michael.
> >
>
>
>
> --
> Cheers
> Michael.
>
Re: FYI - Forced push on branch-3.4
Posted by Michael Han <ha...@cloudera.com>.
INFRA-13916 <https://issues.apache.org/jira/browse/INFRA-13916> is resolved
now.
I am planning to file another INFRA JIRA to apply protection scheme on all
branches of ZK git repo, because the force push should be an exception
rather than default. If we do need force push we can always ask INFRA to
temporarily open the gate. Please shout out if you have concerns on
disabling force push on git repo - otherwise if I hear nothing by end of
this week I'll let INFRA disable force push on ZK git repo.
On Tue, Apr 18, 2017 at 1:47 PM, Michael Han <ha...@cloudera.com> wrote:
> Another more recent issue related to force push and Hadoop:
> https://issues.apache.org/jira/browse/INFRA-13902
>
> I think ZooKeeper git project should be configured to disable force push
> by default for all branches. I believe the force push was not currently
> configured as disabled otherwise I would not succeed on force pushing which
> leads to an observable inconsistent state for branch-3.4. I'll follow up
> with INFRA once we restore the linearizability of branch-3.4.
>
>
>
> On Tue, Apr 18, 2017 at 12:40 PM, Patrick Hunt <ph...@apache.org> wrote:
>
>> fwiw some background on what other teams have done:
>> https://issues.apache.org/jira/browse/INFRA-11236
>>
>> Patrick
>>
>> On Tue, Apr 18, 2017 at 10:23 AM, Michael Han <ha...@cloudera.com> wrote:
>>
>> > I did a forced push on branch-3.4 this morning to remove a bad commit
>> that
>> > breaks the build of branch-3.4. I wasn't aware that forced push should
>> be
>> > avoided until Pat Hunt reminds me.
>> > After force push I noticed some inconsistency state of top ref for
>> > branch-3.4, and I am following up with INFRA on this [1].
>> >
>> > Apologize for any inconvenience caused. Please let me know if you
>> observe
>> > other problems caused by the change of the state for branch-3.4.
>> >
>> > [1] https://issues.apache.org/jira/browse/INFRA-13916
>> > --
>> > Cheers
>> > Michael.
>> >
>>
>
>
>
> --
> Cheers
> Michael.
>
--
Cheers
Michael.
Re: FYI - Forced push on branch-3.4
Posted by Michael Han <ha...@cloudera.com>.
Another more recent issue related to force push and Hadoop:
https://issues.apache.org/jira/browse/INFRA-13902
I think ZooKeeper git project should be configured to disable force push by
default for all branches. I believe the force push was not currently
configured as disabled otherwise I would not succeed on force pushing which
leads to an observable inconsistent state for branch-3.4. I'll follow up
with INFRA once we restore the linearizability of branch-3.4.
On Tue, Apr 18, 2017 at 12:40 PM, Patrick Hunt <ph...@apache.org> wrote:
> fwiw some background on what other teams have done:
> https://issues.apache.org/jira/browse/INFRA-11236
>
> Patrick
>
> On Tue, Apr 18, 2017 at 10:23 AM, Michael Han <ha...@cloudera.com> wrote:
>
> > I did a forced push on branch-3.4 this morning to remove a bad commit
> that
> > breaks the build of branch-3.4. I wasn't aware that forced push should be
> > avoided until Pat Hunt reminds me.
> > After force push I noticed some inconsistency state of top ref for
> > branch-3.4, and I am following up with INFRA on this [1].
> >
> > Apologize for any inconvenience caused. Please let me know if you observe
> > other problems caused by the change of the state for branch-3.4.
> >
> > [1] https://issues.apache.org/jira/browse/INFRA-13916
> > --
> > Cheers
> > Michael.
> >
>
--
Cheers
Michael.
Re: FYI - Forced push on branch-3.4
Posted by Patrick Hunt <ph...@apache.org>.
fwiw some background on what other teams have done:
https://issues.apache.org/jira/browse/INFRA-11236
Patrick
On Tue, Apr 18, 2017 at 10:23 AM, Michael Han <ha...@cloudera.com> wrote:
> I did a forced push on branch-3.4 this morning to remove a bad commit that
> breaks the build of branch-3.4. I wasn't aware that forced push should be
> avoided until Pat Hunt reminds me.
> After force push I noticed some inconsistency state of top ref for
> branch-3.4, and I am following up with INFRA on this [1].
>
> Apologize for any inconvenience caused. Please let me know if you observe
> other problems caused by the change of the state for branch-3.4.
>
> [1] https://issues.apache.org/jira/browse/INFRA-13916
> --
> Cheers
> Michael.
>