You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hbase.apache.org by Josh Elser <jo...@gmail.com> on 2017/10/23 15:45:09 UTC
PreCommit whitespace failures on JVM crash dump files
Hiya,
Noticed a PreCommit failure from over the weekend on the whitespace
(EOL, tabs) checks [1][2]:
e.g.
```
./hbase-server/hs_err_pid30492.log:637:731000000-734f00000 rw-p 00000000
00:00 0
./hbase-server/hs_err_pid30492.log:638:734f00000-751000000 rw-p 00000000
00:00 0
```
Any thoughts on the right way to fix this one?
The whitespace check ran before the tests did, so I can only assume
these were laying around on the filesystem before the PreCommit job
itself ran. Do we need to be a bit more aggressive in the clean steps?
I do see in the Jenkinsfile, we are taking advantage of the options in
Yetus to ignore certain files for whitespace. That would be another
option to "fix" this.
- Josh
[1]
https://builds.apache.org/job/PreCommit-HBASE-Build/9309/artifact/patchprocess/whitespace-eol.txt
[2]
https://builds.apache.org/job/PreCommit-HBASE-Build/9309/artifact/patchprocess/whitespace-tabs.txt
Re: PreCommit whitespace failures on JVM crash dump files
Posted by Josh Elser <jo...@gmail.com>.
Thanks, Sean & Mike! Glad I'm not the only one who has seen this :)
Re .gitignore: not that I can tell (touch'ing an
hbase-server/hs_err_pid30492.log shows it up as untracked in my local
checkout).
On 10/23/17 12:12 PM, Sean Busbey wrote:
> we use git clean I think to do the post-checkout cleanup. are the dump
> files in our git ignore?
>
> On Mon, Oct 23, 2017 at 10:48 AM, Mike Drob <md...@apache.org> wrote:
>> I've seen this too, but I don't understand how the dumps are present from
>> what should notionally be a clean checkout.
>>
>> On Mon, Oct 23, 2017 at 10:45 AM, Josh Elser <jo...@gmail.com> wrote:
>>
>>> Hiya,
>>>
>>> Noticed a PreCommit failure from over the weekend on the whitespace (EOL,
>>> tabs) checks [1][2]:
>>>
>>> e.g.
>>>
>>> ```
>>> ./hbase-server/hs_err_pid30492.log:637:731000000-734f00000 rw-p 00000000
>>> 00:00 0
>>> ./hbase-server/hs_err_pid30492.log:638:734f00000-751000000 rw-p 00000000
>>> 00:00 0
>>> ```
>>>
>>> Any thoughts on the right way to fix this one?
>>>
>>> The whitespace check ran before the tests did, so I can only assume these
>>> were laying around on the filesystem before the PreCommit job itself ran.
>>> Do we need to be a bit more aggressive in the clean steps?
>>>
>>> I do see in the Jenkinsfile, we are taking advantage of the options in
>>> Yetus to ignore certain files for whitespace. That would be another option
>>> to "fix" this.
>>>
>>> - Josh
>>>
>>> [1] https://builds.apache.org/job/PreCommit-HBASE-Build/9309/art
>>> ifact/patchprocess/whitespace-eol.txt
>>> [2] https://builds.apache.org/job/PreCommit-HBASE-Build/9309/art
>>> ifact/patchprocess/whitespace-tabs.txt
>>>
Re: PreCommit whitespace failures on JVM crash dump files
Posted by Sean Busbey <bu...@apache.org>.
we use git clean I think to do the post-checkout cleanup. are the dump
files in our git ignore?
On Mon, Oct 23, 2017 at 10:48 AM, Mike Drob <md...@apache.org> wrote:
> I've seen this too, but I don't understand how the dumps are present from
> what should notionally be a clean checkout.
>
> On Mon, Oct 23, 2017 at 10:45 AM, Josh Elser <jo...@gmail.com> wrote:
>
>> Hiya,
>>
>> Noticed a PreCommit failure from over the weekend on the whitespace (EOL,
>> tabs) checks [1][2]:
>>
>> e.g.
>>
>> ```
>> ./hbase-server/hs_err_pid30492.log:637:731000000-734f00000 rw-p 00000000
>> 00:00 0
>> ./hbase-server/hs_err_pid30492.log:638:734f00000-751000000 rw-p 00000000
>> 00:00 0
>> ```
>>
>> Any thoughts on the right way to fix this one?
>>
>> The whitespace check ran before the tests did, so I can only assume these
>> were laying around on the filesystem before the PreCommit job itself ran.
>> Do we need to be a bit more aggressive in the clean steps?
>>
>> I do see in the Jenkinsfile, we are taking advantage of the options in
>> Yetus to ignore certain files for whitespace. That would be another option
>> to "fix" this.
>>
>> - Josh
>>
>> [1] https://builds.apache.org/job/PreCommit-HBASE-Build/9309/art
>> ifact/patchprocess/whitespace-eol.txt
>> [2] https://builds.apache.org/job/PreCommit-HBASE-Build/9309/art
>> ifact/patchprocess/whitespace-tabs.txt
>>
Re: PreCommit whitespace failures on JVM crash dump files
Posted by Mike Drob <md...@apache.org>.
I've seen this too, but I don't understand how the dumps are present from
what should notionally be a clean checkout.
On Mon, Oct 23, 2017 at 10:45 AM, Josh Elser <jo...@gmail.com> wrote:
> Hiya,
>
> Noticed a PreCommit failure from over the weekend on the whitespace (EOL,
> tabs) checks [1][2]:
>
> e.g.
>
> ```
> ./hbase-server/hs_err_pid30492.log:637:731000000-734f00000 rw-p 00000000
> 00:00 0
> ./hbase-server/hs_err_pid30492.log:638:734f00000-751000000 rw-p 00000000
> 00:00 0
> ```
>
> Any thoughts on the right way to fix this one?
>
> The whitespace check ran before the tests did, so I can only assume these
> were laying around on the filesystem before the PreCommit job itself ran.
> Do we need to be a bit more aggressive in the clean steps?
>
> I do see in the Jenkinsfile, we are taking advantage of the options in
> Yetus to ignore certain files for whitespace. That would be another option
> to "fix" this.
>
> - Josh
>
> [1] https://builds.apache.org/job/PreCommit-HBASE-Build/9309/art
> ifact/patchprocess/whitespace-eol.txt
> [2] https://builds.apache.org/job/PreCommit-HBASE-Build/9309/art
> ifact/patchprocess/whitespace-tabs.txt
>