Bugzilla Bug 56211

email: webmaster@securesystem.info
age:   20
loc:   Stockton, CA USA
real name: Chris White
IRC:   ChrisWhite
mentor: me
role:  media-{sound,video} maintainer

2 week mentorship period ends July 20

he's all setup now =)

Ok, so his latest round of screwups include:

* committing a tarball (net-analyzer/admlogger_ebuild.tar.bz2) into gentoo-x86
* branching the whole of gentoo-x86 with a new 'chris' branch
* committing a bogus top-level (perl-gcpan/)
* committing bogus files to top-level (en.txt, ja.txt)

Reopening since this isn't exactly the first time he's utterly h0rked the tree.

To explain the situation, I was testing kde program (cervisa) by doing a cvs 
checkout of the tree.  Unfortunately this mess you see here arose.  The 
following then occured:

Halyc0n notified me of the gcpan directory, and a few other ebuilds that were 
problematic.  I fixed the mentioned ones, and looked at package.gentoo.org to 
make sure I hadn't missed anything.

ferringb notified me of the branch issue shortly afterwards, to which I was 
called to #gentoo-infra and discussed the issue with Ramareth, ferringb, and 
Halyc0n.  The branch fixing looked fairly easy, and there didn't seem to be any 
outstanding occuring that would cause me to believe the problem was that severe.

I kept an eye on #gentoo-bugs till about 4AM or so fixing any bugs that appeared 
to be something that was caused by the situation.

Notification by Mr_Bones_ and Anarchy on other packages I missed.  Those were 
fixed as I noticed them.

tomk notified me to cleanout my entire local cvs directory, I did so.  I was 
notified about the ja/en.txt files by Flameeyes, but wanted to ask a cvs admin 
for removal, as they exist in the toplevel cvs, and I didn't feel comfortable 
commiting something at that level (same with the perl-gcpan directory), and 
would rather take the better approach at that.  The admlogger tarball I didn't 
realize about and it looks to be missing (if I did miss it, feel free to open a 
seperate bug and I'll fix it).

Since I will most likely be asked a "Why should we keep you" question (I assume 
judging from the tone of the last comment it's just that), I'll present the 
following (these are for the time period before this event):

- Before the event, I do not recall a time that a QA member has approached me 
with a repoman issue.
- I helped to increase the number of x86 markings and security turnarounds for 
the x86 herd (Feel free to ask Halyc0n)
- I helped to collaborate with mcummings and address the need for ebuild bumps
- The actions during the event were not intentional (I have the feeling some 
people believe they were?). No, I didn't feel good at all about what happened.
- Tested the entire of KDE, every single app and provided patches in some 
instances
- Helped gustavoz with narrowing down on a missing patch that was causing touch 
to create files with invalid dates, important enough to have a new kernel rolled 
out

That is all, if supenssion is the end result, I propose 60 days.  Appoligies to 
all that were effected (users and devs), and expecially to infra for the work 
they had to do because of it.

> * branching the whole of gentoo-x86 with a new 'chris' branch

Our friendly infra monkeys have now dealt with this issue or so I'm told.

(In reply to comment #3)
> To explain the situation, I was testing kde program (cervisa) by doing a cvs 
> checkout of the tree.  Unfortunately this mess you see here arose.  The 
> following then occured:
First off, you shouldn't have tested naything new like that against our tree(s)
[snip]
> That is all, if supenssion is the end result, I propose 60 days.  Appoligies to 
> all that were effected (users and devs), and expecially to infra for the work 
> they had to do because of it.
A suspension is just plain silly.  You fucked up(rather badly), when people
pointed it out to you, you did what you could to correct it, and then
apoligized.  You should review the developer handbook for the proper way to
commit to the tree.

If this were the first time that Chris had done something silly and ended up
breaking things through sheer carelessness, I wouldn't've reopened the bug.

Yes, he has been doing a lot to help out, but I was unaware he was only
committing most of KDE-3.4.3 to stable.  He said he was going to handle this the
other day, but our users have been having fun with packages upgrading and
downgrading because of packages still being ~x86 (this is bug #115357 for
reference).  He also missed one package in the monolithic KDE packages that
completely broke the dep tree for users.

Also, there is still a tag, "start", in the tree that he created.

should be devrel

any reason this is closed ?  i dont think there is ...

closing again

Hi, 
As ChrisWhite is the #gentoo ops liasion for user-relations I would like him
added to the user-relations alias please.

Added a while ago, closing.

Chris recently announced his resignation.

Chris, good luck in the future.

Closing as Chris decided to stay after all.

MIA since October 2006, reopening.

Well, I reassigned the bug m-needed since they were really really stale; Chris
is still CCed in case he gets less busy w/ pole dancing. *g*

Infra, chriswhite should be ready for retirement.

Retired by infra on 2007/July/10.
Pending planet and forums.

infra stuff completed.

Forums done.

Updated userinfo.xml; Removed from #gentoo-dev access; Removed from herds.xml
(perl and pda); metadata.xml entries updated.

project pages cleaned up.

removed from planet

Our bug again.

removed #-dev access, removed cloak, updated userinfo.xml

No metadata.xml entries present any longer, went away w/
http://article.gmane.org/gmane.linux.gentoo.devel/50444.

Retirement done, closing bug.