1 |
On May 23, 2012 1:55 PM, "Johannes Huber" <johu@g.o> wrote: |
2 |
> |
3 |
> Am Mittwoch 23 Mai 2012, 14:42:37 schrieb Michael Weber: |
4 |
> |
5 |
> > -----BEGIN PGP SIGNED MESSAGE----- |
6 |
> |
7 |
> > Hash: SHA256 |
8 |
> |
9 |
> > |
10 |
> |
11 |
> > Hi, |
12 |
> |
13 |
> > |
14 |
> |
15 |
> > i've looked at the blockers of "[TRACKER] portage migration to git" |
16 |
> |
17 |
> > [1] and want to discuss "testing git-cvsserver" [2]. |
18 |
> |
19 |
> > |
20 |
> |
21 |
> > There are two proposed scenarios how to migrate the developers write |
22 |
> |
23 |
> > access to the portage tree. |
24 |
> |
25 |
> > |
26 |
> |
27 |
> > "Clean cut" turns of cvs access on a given and announced timestamp, |
28 |
> |
29 |
> > rsync-generation/updates is suspended (no input -> no changes), some |
30 |
> |
31 |
> > magic scripts prepare the git repo (according to [3], some hours |
32 |
> |
33 |
> > duration) and we all checkout the tree (might be some funny massive |
34 |
load). |
35 |
> |
36 |
> > |
37 |
> |
38 |
> > "testing git-cvsserver" proses "Clean cut" with the additional ability |
39 |
> |
40 |
> > to continue using cvs update/commit, - in best case - on the old |
41 |
> |
42 |
> > checkout w/o alteration on the developers side. |
43 |
> |
44 |
> > |
45 |
> |
46 |
> > "Clean cut" forces us to clean up out dirty checkouts (I have some |
47 |
> |
48 |
> > added directories, added ebuilds i hesitated to `repoman commit`). |
49 |
> |
50 |
> > Plus we have to alter all our hot-wired portage mangling scripts from |
51 |
> |
52 |
> > cvs'ish to git'ish (I use my read/write checkout as portage tree (cvs |
53 |
> |
54 |
> > checkout + egencache for checkout) and have an automated google-chrome |
55 |
> |
56 |
> > bump script). But this can be accomplished on a per developer basis, |
57 |
> |
58 |
> > and slackers don't stall the process. |
59 |
> |
60 |
> > |
61 |
> |
62 |
> > "testing git-cvsserver" forces us all to test these cvs'ish scripts |
63 |
> |
64 |
> > and behaviours against a git-cvsserver and report. |
65 |
> |
66 |
> > We all know that this test-runs will never happen, stalling this bug |
67 |
> |
68 |
> > till infinity. |
69 |
> |
70 |
> > Plus infra/"subset of devs marshalling the migration" get stuck |
71 |
> |
72 |
> > between fixing git issues and git-cvsserver. |
73 |
> |
74 |
> > |
75 |
> |
76 |
> > *if you still read this* *wow* |
77 |
> |
78 |
> > |
79 |
> |
80 |
> > Please discuss my arguments and come to the conclusions to |
81 |
> |
82 |
> > RESO/WONT-FIX "testing git-cvsserver", make a "clean cut" and remove |
83 |
> |
84 |
> > this bug from the blockers of "[TRACKER] portage migration to git". |
85 |
> |
86 |
> > |
87 |
> |
88 |
> > My lengthy 2 cents. |
89 |
> |
90 |
> > |
91 |
> |
92 |
> > [1] https://bugs.gentoo.org/333531 |
93 |
> |
94 |
> > [2] https://bugs.gentoo.org/333699 |
95 |
> |
96 |
> > [3] https://bugs.gentoo.org/333705#c2 |
97 |
> |
98 |
> > - -- |
99 |
> |
100 |
> > Gentoo Dev |
101 |
> |
102 |
> > http://xmw.de/ |
103 |
> |
104 |
> > -----BEGIN PGP SIGNATURE----- |
105 |
> |
106 |
> > Version: GnuPG v2.0.17 (GNU/Linux) |
107 |
> |
108 |
> > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ |
109 |
> |
110 |
> > |
111 |
> |
112 |
> > iF4EAREIAAYFAk+82z0ACgkQknrdDGLu8JBUWAD/dmuqyES/mYDrMam+/txnHmgd |
113 |
> |
114 |
> > VaQaqwHMlwzzqQwbpY4A/0h+5Vp8sLbOE78k4SCaGE2dCQtmeOz0jd1YxkDzP+YW |
115 |
> |
116 |
> > =jXLQ |
117 |
> |
118 |
> > -----END PGP SIGNATURE----- |
119 |
> |
120 |
> |
121 |
> |
122 |
> I support RESOLUTION WONTFIX, if nobody cares about the bug since it was |
123 |
opened it is obvious out of interest. There is no reason to support |
124 |
jurassic software. |
125 |
> |
126 |
> |
127 |
> |
128 |
> Clean cut++ |
129 |
> |
130 |
> |
131 |
> |
132 |
> Cheers |
133 |
> |
134 |
> -- |
135 |
> |
136 |
> Johannes Huber (johu) |
137 |
> |
138 |
> Gentoo Linux Developer / KDE Team |
139 |
> |
140 |
> GPG Key ID F3CFD2BD |
141 |
> |
142 |
> |
143 |
|
144 |
Another vote for clean cut from me. |