Please never force push to any non-feature branch. I thought we had protections to stop force pushes on master and have filed INFRA-12602 to get them in place.
To fix erroneous commit messages, please revert the offending commits and then reapply them with a correct commit message. You can see a recent example of this in master when Ted corrected himself on missing the contributor for HBASE-16491. On Tue, Sep 13, 2016 at 12:53 AM, Gary Helmling <[email protected]> wrote: > Yes this is one commit message where the JIRA number was left off. I only > noticed it was missing after I pushed to master. Mea culpa. > > I've rebased to amend the commit message and force pushed. However, there > were two intervening commits in the meantime, which also got included in > the rebase: > > commit 77b327320a72ca01b35f655c42f8c13f659dff31 > Author: anoopsamjohn <[email protected]> > Date: Tue Sep 13 11:43:26 2016 +0530 > > HBASE-16229 Cleaning up size and heapSize calculation. > > commit 2566cfeb60de644f287ac192d360f3fc15376c8f > Author: chenheng <[email protected]> > Date: Tue Sep 13 10:07:45 2016 +0800 > > HBASE-16592 Unify Delete request with AP > > > I like a clear commit history as much as anyone, but let's consider the > consequences of "fixing" this. Unfortunately anyone who pulled master > between the original commit and the fix may now need to re-checkout master > to avoid divergent commits. In hindsight, I think the fix may create more > pain than the original problem. My apologies if you have to re-checkout > master as a result. > > Anoop and Heng, you will both likely fall into this camp. Again, my > apologies. > > This reaffirms my prejudice against force pushing to shared branches. > > > On Mon, Sep 12, 2016 at 6:24 PM Duo Zhang <[email protected]> wrote: > >> >> https://github.com/apache/hbase/commit/1cdc5acfd456688f106287c19ae9af62fd03a3af >> >> I can not find the related issue of this commit... >> >> Thanks. >> -- busbey
