1 |
On Wed, 1 Dec 2010 12:13:03 -0800 |
2 |
Alec Warner <antarus@g.o> wrote: |
3 |
|
4 |
> On Tue, Nov 30, 2010 at 8:02 PM, Jorge Manuel B. S. Vicetto |
5 |
> <jmbsvicetto@g.o> wrote: |
6 |
> > -----BEGIN PGP SIGNED MESSAGE----- |
7 |
> > Hash: SHA1 |
8 |
> > |
9 |
> > On 29-11-2010 10:34, Sebastian Pipping wrote: |
10 |
> >> On 11/29/10 09:35, Arfrever Frehtes Taifersar Arahesis wrote: |
11 |
> >>> There will probably be no active version of Python set. |
12 |
> >> |
13 |
> >> You had two weeks to come up with this. |
14 |
> >> |
15 |
> >> Please find my on IRC to team up on an agreed fix. |
16 |
> > |
17 |
> > As Arfrever noted, this is likely the cause of the broken |
18 |
> > automated weekly stages for this past week. By not having a |
19 |
> > python symlink / wrapper, stages generation failed on stage2 run. |
20 |
> > I'd like to take this chance to recall this is the 2nd time on |
21 |
> > the last few months where stage generation was broken by python |
22 |
> > changes. Also, we've been unable to create hardened stages for |
23 |
> > over 8 weeks because of a sandbox issue. |
24 |
> > The weekly stages generation depends on the quality and stability |
25 |
> > of the "stable" tree. Therefore, the RelEng team kindly asks all |
26 |
> > maintainers to pay attention to the stable ebuilds in the system |
27 |
> > set and to please fix any failures asap as they may / can prevent |
28 |
> > stage generation. Be sure to think carefully about changes that |
29 |
> > can impact the stage generation, in particular when they involve |
30 |
> > python. |
31 |
> |
32 |
> Two issues: |
33 |
> |
34 |
> proj/en/releng is old as hell and doesn't even mention stage |
35 |
> generation. |
36 |
> |
37 |
> How does a developer know when the stage generation is broken? Is |
38 |
> there a dashboard? At work we have a guy who is basically a build |
39 |
> cop and checks our build dashboard once a day or so and if it is |
40 |
> broken he goes and finds the guy who broke it and punches him in |
41 |
> the face until he fixes it. I imagine we do not have staff for |
42 |
> this (and no one has invented punching over the internet.) |
43 |
|
44 |
Catalyst sends automated emails to releng@g.o from the |
45 |
various build boxes: dolphin, poseidon, other dev.g.o machines. |
46 |
|
47 |
> I am curious how often stage builds fail (how long can they be |
48 |
> broken until we actually care?) |
49 |
|
50 |
Fairly often, especially in the last couple of months or so. There |
51 |
were some arches that, last I checked, hadn't had |
52 |
any new media in several months. Python is the usual cause. |
53 |
Remember the last huge Python debacle that resulted in suspension? |
54 |
Yeah, that was one of the reasons for continually broken media. |
55 |
|
56 |
Python issues are pretty much the only reason why general stage builds |
57 |
fail (hardened is its own set of problems.) |
58 |
|
59 |
Here's part of a typical message from one of the boxes, minus a whole |
60 |
bunch of "bad interpreter" errors: |
61 |
|
62 |
--------------------------------------------------------------------- |
63 |
[[ (1/3) Configuring environment ]] |
64 |
/usr/portage/scripts/bootstrap.sh: line 307: python: command not found |
65 |
--------------------------------------------------------------------- |
66 |
[[ (2/3) Updating portage ]] |
67 |
env: emerge: No such file or directory |
68 |
|
69 |
!!! catalyst: run script failed. |
70 |
|
71 |
Traceback (most recent call last): |
72 |
File "modules/generic_stage_target.py", line 1207, in run_local |
73 |
"run script failed.",env=self.env) |
74 |
File "/usr/lib64/catalyst/modules/catalyst_support.py", line 542, |
75 |
in cmd |
76 |
raise CatalystError,myexc |
77 |
CatalystError |
78 |
None |
79 |
|
80 |
I see messages like this pretty much every day. Releng is |
81 |
understaffed on a few arches, which is why no one has time to track |
82 |
down the errors, fix them, and get the builds completed. |