jdk/jdk fails hg verify

[coleen.phillimore at oracle.com]

Adding ops.

On 7/24/19 9:48 AM, coleen.phillimore at oracle.com wrote:
>
> Is somebody working on this?  Do we need to not update our 
> repositories and check things in?
>
> thanks,
> Coleen
>
> On 7/24/19 4:33 AM, Alan Bateman wrote:
>> On 24/07/2019 09:00, Aleksey Shipilev wrote:
>>> Hi,
>>>
>>> My CI jobs are doing "hg verify" after every pull to validate that 
>>> the repository state is sane. It
>>> started to fail yesterday on jdk/jdk, on multiple machines and 
>>> multiple checkouts, like this:
>>>
>>> $ hg verify
>>> checking changesets
>>> checking manifests
>>> crosschecking files in changesets and manifests
>>> checking files
>>> test/langtools/jdk/javadoc/doclet/testLinkOption/TestOptionOrder.java at 55761: 
>>> f7fb5d9b7490 not in
>>> manifests
>>> checked 55787 changesets with 479505 changes to 184101 files
>>> 1 integrity errors encountered!
>>> (first damaged changeset appears to be 55761)
>>>
>>> $ hg log -r 55761
>>> changeset:   55761:afe8584ac8d9
>>> user:        jjg
>>> date:        Mon Jul 22 16:10:25 2019 -0700
>>> summary:     8222791: Order of evaluation of -link params in Javadoc 
>>> tool reversed -> regression
>>> with split packages
>>>
>>> Two fresh jdk/jdk clones on two different machines yield the same 
>>> verification trouble.
>>>
>>> Something smells fishy here: it looks either like remote repo 
>>> corruption, or repo history was edited?
>>>
>> I'm seeing the same thing after sync'ing up today. It might be best 
>> for everyone to hold off pushing to jdk/jdk while this is investigated.
>>
>> -Alan
>