hive-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Lefty Leverenz <leftylever...@gmail.com>
Subject Re: Please double-check JIRA number on commit
Date Mon, 08 Jun 2015 02:32:11 GMT
Here are the ones I know about:

   - 5a576b6fbf1680ab4dd8f275cad484a2614ef2c1  HIVE-10391
   <https://issues.apache.org/jira/browse/HIVE-10391>
   - 582f4e1bc39b9605d11f762480b29561a44688ae  HIVE-10217
   <https://issues.apache.org/jira/browse/HIVE-10217>
   - 8981f365bf0cf921bc0ac2ff8914df44ca2f7de7  HIVE-10500
   <https://issues.apache.org/jira/browse/HIVE-10500>
   - 09100831adff7589ee48e735a4beac6ebb25cb3e
   & f3ab5fda6af57afff31c29ad048d906fd095d5fb  HIVE-10885
   <https://issues.apache.org/jira/browse/HIVE-10885>

Those have JIRA comments listing the commit IDs.  Should we list commit IDs
on all issues?

-- Lefty

On Sun, Jun 7, 2015 at 2:16 PM, Lefty Leverenz <leftyleverenz@gmail.com>
wrote:

> Good idea!  I'll compile a list.
>
> -- Lefty
>
> On Sat, Jun 6, 2015 at 9:00 AM, Owen O'Malley <omalley@apache.org> wrote:
>
>> Editing commits creates a lot of problems. May I suggest that we start a
>> file that lists the exceptions that you've noticed and check that into our
>> source tree? That will give everyone access to fix things when they
>> recognize it. It would look like:
>>
>> 75b679c54c9088434678da2a4d89122620903d72 hive-1234
>> e8520424879d07e50ea07c28a671251e95c382f6 hive-1235
>>
>> Of course there may be multiple commits for the same jira, especially on
>> different branches.
>>
>> .. Owen
>>
>> .. Owen
>>
>> On Sat, Jun 6, 2015 at 2:30 AM, Lefty Leverenz <leftyleverenz@gmail.com>
>> wrote:
>>
>> > >
>> > > ... we should put the commit log in the corresponding
>> > > JIRA comment as a cross-reference.
>> > >
>> >
>> > I've only noticed one more slip-up since the end of April, but still
>> think
>> > this is a good idea.
>> >
>> > Any more suggestions?
>> >
>> >
>> > -- Lefty
>> >
>> > On Thu, Apr 30, 2015 at 4:53 PM, Xuefu Zhang <xzhang@cloudera.com>
>> wrote:
>> >
>> > > Yes, for that, I think we should put the commit log in the
>> corresponding
>> > > JIRA comment as a cross-reference. I have to admit that I had to fix
>> > commit
>> > > message for a few of my svn commits in the past.
>> > >
>> > > On Thu, Apr 30, 2015 at 3:53 PM, Lefty Leverenz <
>> leftyleverenz@gmail.com
>> > >
>> > > wrote:
>> > >
>> > > > Found a negative answer to #2 in some ASF instructions
>> > > > <http://www.apache.org/security/committers.html> for security
>> issues
>> > > > (*underline
>> > > > added*):
>> > > >
>> > > > 15.  The log for the svn commit that applied the fix is updated to
>> > > include
>> > > > > the CVE number. Projects that use git as their primary source
code
>> > > > control
>> > > > > system should not do this as *editing a pushed commit causes
all
>> > sorts
>> > > of
>> > > > > problems*.
>> > > >
>> > > >
>> > > > All the more reason to be careful about the JIRA number when you
>> > commit a
>> > > > patch.
>> > > >
>> > > >
>> > > > -- Lefty
>> > > >
>> > > > On Thu, Apr 30, 2015 at 2:57 AM, Lefty Leverenz <
>> > leftyleverenz@gmail.com
>> > > >
>> > > > wrote:
>> > > >
>> > > > > Since switching to git, three commits have had the wrong JIRA
>> number.
>> > > > Two
>> > > > > were typos and the third one gave the umbrella issue's number.
>> > > > Fortunately
>> > > > > the correct issue could be found by searching for the summary
>> text.
>> > > > >
>> > > > >    1. How can we reduce this failure rate?
>> > > > >    2. Can JIRA numbers be fixed after-the-fact in commit records?
>> > > > >    3. Should JIRA issues track commit IDs?
>> > > > >
>> > > > > -- Lefty
>> > > > >
>> > > >
>> > >
>> >
>>
>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message