[Gluster-devel] Changing Submit Type for glusterfs

[Nigel Babu]

On Thu, Sep 01, 2016 at 11:33:55PM -0400, Vijay Bellur wrote:
> On Thu, Sep 1, 2016 at 3:33 AM, Kaushal M <kshlmster at gmail.com> wrote:
> > On Thu, Sep 1, 2016 at 12:36 PM, Nigel Babu <nigelb at redhat.com> wrote:
> >> On Mon, Aug 29, 2016 at 10:55 AM, Poornima Gurusiddaiah
> >> <pgurusid at redhat.com> wrote:
> >>>
> >>> Hi,
> >>>
> >>> Regarding the enforcement of the dependencies while merging, i see that
> >>> the dependent patches on any patch is mentioned in the "Related Changes"
> >>> column [1]. It still doesn't enforce, in the cherry-pick way of
> >>> submitting changes, by default it ignores the lineage [2]. But there are
> >>> ways to enforce this. Will let the gluster infra maintainers to comment
> >>> on the same.
> >>>
> >>> [1]
> >>> https://gerrit-review.googlesource.com/Documentation/user-review-ui.html#related-changes
> >>> [2]
> >>> https://gerrit-review.googlesource.com/Documentation/project-configuration.html#project_options
> >>>
> >>> Regards,
> >>> Poornima
> >>
> >>
> >> Thank you Poornima for pointing this out. You're right, it's worth changing
> >> our submission type to either "Rebase If Necessary" or "Merge If Necessary"
> >> to enforce this. I'm not sure what the implications are, so I'll report back
> >> after I've setup a test on staging so we can experiment and see what works.
> >
> > The reason cherry-pick was chosen was to keep the branch linear and
> > avoid merge-commits as (I'm guessing here) this makes the tree hard to
> > follow.
> > Merge-if-necessary will not keep the branch linear. I'm not sure how
> > rebase-if-necessary works though.
> >
> > Vijay, can you provide anymore background for the choice of
> > cherry-pick and you opinion on the change?
> >
>
> Unfortunately I do not recollect the reason for cherry-pick being the
> current choice. FWIW, I think dependencies were being enforced a while
> back in the previous version(s) of gerrit. Not sure if something has
> changed in the more recent gerrit versions.
>

According to the documentation, the behavior was intended to be like how it is
currently. If it worked in the past, it may have been a bug. Let me setup
a test with Rebase-If-Necessary. Then we can make an informed decision on which
way to go about it.

--
nigelb