Still Invalid @openjdk.java.net mail addresses, and unassociated commits at GH

[Thomas Stüfe]

Thanks a lot for your help, Magnus. I appreciate your and Tims efforts.

Thankfully, adding @openjdk.org did the trick for me too, now the GitHub
user is linked. That, and knowing that mails sent to @openjdk.java.net mail
addresses are not lost, solves the issues for me.

Again, thanks, I'm happy now.

Cheers, Thomas

On Fri, Feb 19, 2021 at 9:27 AM Magnus Ihse Bursie <
magnus.ihse.bursie at oracle.com> wrote:

>
>
> On 2021-02-19 08:19, Thomas Stüfe wrote:
>
> Hi Magnus,
>
> Thanks a lot for helping! Answers inline below.
>
> On Thu, Feb 18, 2021 at 5:31 PM Magnus Ihse Bursie <
> magnus.ihse.bursie at oracle.com> wrote:
>
>> On 2021-02-18 15:38, Thomas Stüfe wrote:
>> > Hi guys,
>> >
>> > It has been six months now.
>> >
>> >
>> https://mail.openjdk.java.net/pipermail/skara-dev/2020-October/003617.html
>> > - we still have invalid mail addresses like stuefe at openjdk.java.net,
>> which
>> > cannot be replied to, read or verified in GitHub. They clutter the
>> mailing
>> > list archive with invalid mail addresses too.
>>
>> I do believe this works now. At least I get mails to
>> ihse at openjdk.java.net delivered to my Oracle mail address. Maybe the
>> mapping for non-Oracle mail address is not finished yet..?
>>
>> (I also get mails to ihse at openjdk.org, but that seems to work only from
>> Oracle's SMTP servers, not from outside servers)
>>
>
> Yes, seems this started working for us too. At least for
> stuefe at openjdk.java.net. stuefe at openjdk.net and both org variants don't
> get through. Remaining nit is that Github still marks
> stuefe at openjdk.java.net as undeliverable, but I can live with that.
>
>
>
>
>> >
>> https://mail.openjdk.java.net/pipermail/skara-dev/2020-October/003752.html
>> > - we also still have this issue, which may linked to the invalid mail
>> > addresses, that a large number of the developers has no real link
>> between
>> > their GitHub user and their commits to openjdk are not linked to their
>> user
>> > name. The proposed workaround of adding the unverified openjdk mail
>> address
>> > to GH did not work, at least not in my case.
>> >
>> > Is there any hope that these issues will be resolved? Should I open
>> bugs to
>> > track these?
>> Once again, it works for me. See e.g.
>>
>> https://github.com/openjdk/jdk/commit/d5a4d2266b45107cf8d7c0d0137c77b797d836b6
>>
>> I have added "ihse at openjdk.org" as a mail address to my Github account.
>> I think the trick is to use @openjdk.org, not @openjdk.java.net -- maybe
>> that is what you have done?
>>
>>
> I tried, in the past, @openjdk.java.net and openjdk.java.org, without
> success. I'll retry with your suggestion (leaving out the java part).
>
>
> There are only two domains that are of interest here -- openjdk.java.net
> (which we have traditionally been using) and openjdk.org (which we would
> like to transition to, but things are moving slowly for this kind of
> changes within Oracle). There is no openjdk.net nor java.org, so you
> don't need to test those.
>
> Skara is creating commits for OpenJDK projects with the form <openjdk-id>@
> openjdk.org, so that is what you need to add to your github profile to
> get commits connected to your profile.
>
> The PR comment mail bridge bot sends mail with the sender specified as
> <openjdk-id>@openjdk.java.net.
>
> The reason for this disconnect is, I assume, the fact that commits will be
> Forever<tm>, and there we want it to look right from the start. But the
> mails are sent right now, and for practical reasons it was simpler (if not
> "simple") to get openjdk.java.net mail working, than for openjdk.org.
>
> I assume the long-term goal here is to make the bots send mail using
> <openjdk-id>@openjdk.org as well, and to get that MTA working. I believe
> Tim is working as well as he can to push this forward, but this is an
> unusual request within Oracle and that unfortunately makes it painful and
> slow to go through with. :(
>
> /Magnus
>
>
>
>> Unverified
>>
>> >
>> > Thanks & Cheers, Thomas
>>
>>
> Thanks again,
>
> Thomas
>
>
>
>