1 |
On 01/09/2015 02:12, covici@××××××××××.com wrote: |
2 |
> Alan McKinnon <alan.mckinnon@×××××.com> wrote: |
3 |
> |
4 |
>> On 31/08/2015 18:54, covici@××××××××××.com wrote: |
5 |
>>>> The words make sense, the meaning doesn't :-) |
6 |
>>>>> |
7 |
>>>>> It looks like fail2ban wants systemd without python support, but the |
8 |
>>>>> true reason is still hidden. The fail2ban ebuild has this: |
9 |
>>>>> |
10 |
>>>>> RDEPEND=" |
11 |
>>>>> ... |
12 |
>>>>> systemd? ( $(python_gen_cond_dep '|| ( |
13 |
>>>>> dev-python/python-systemd[${PYTHON_USEDEP}] |
14 |
>>>>> sys-apps/systemd[python(-),${PYTHON_USEDEP}] |
15 |
>>>>> |
16 |
>>>>> |
17 |
>>>>> I'm thinking maybe you have a specific portage entry that's getting in |
18 |
>>>>> the way. What are your results for: |
19 |
>>>>> |
20 |
>>>>> emerge --info |
21 |
>>>>> grep -r python /etc/portage |
22 |
>>>>> grep -r systemd /etc/portage |
23 |
>>> Just to let you know, most of the python entries were mandated by |
24 |
>>> portage, certainly the systemd one. |
25 |
>> |
26 |
>> |
27 |
>> I'm having a hard time figuring out what is making portage do this. |
28 |
>> I also figure you're OK with a downgraded systemd meanwhile, but just |
29 |
>> for kicks, lets test my theory: If you run this, does portage offer to |
30 |
>> upgrade systemd? |
31 |
>> |
32 |
>> |
33 |
>> USE="-python" emerge -pv systemd |
34 |
> |
35 |
> Well, here is what I got |
36 |
> [ebuild U ] sys-apps/systemd-225:0/2::gentoo [219_p112:0/2::gentoo] |
37 |
> USE="acl kdbus* kmod lz4 pam policykit seccomp ssl -apparmor -audit |
38 |
> -cryptsetup -curl -elfutils -gcrypt -gnuefi% -http -idn -importd -lzma |
39 |
> -nat -qrcode (-selinux) -sysv-utils {-test} -vanilla -xkb (-doc%*) |
40 |
> (-gudev%) (-introspection%*) (-python%*) |
41 |
> (-terminal%)" ABI_X86="32 (64) (-x32)" |
42 |
> PYTHON_SINGLE_TARGET="(-python2_7%*) (-python3_3%) (-python3_4%)" |
43 |
> PYTHON_TARGETS="(-python2_7%*) (-python3_3%) (-python3_4%*)" 3,788 KiB |
44 |
> |
45 |
> Total: 1 package (1 upgrade), Size of downloads: 3,788 KiB |
46 |
> |
47 |
> !!! Multiple package instances within a single package slot have been |
48 |
> pulled |
49 |
> !!! into the dependency graph, resulting in a slot conflict: |
50 |
> |
51 |
> sys-apps/systemd:0 |
52 |
> |
53 |
> (sys-apps/systemd-225:0/2::gentoo, ebuild scheduled for merge) pulled |
54 |
> in by |
55 |
> sys-apps/systemd (Argument) |
56 |
> |
57 |
> (sys-apps/systemd-219_p112:0/2::gentoo, installed) pulled in by |
58 |
> sys-apps/systemd[python(-),python_targets_python2_7(-)?,python_single_target_python2_7(+)?,python_targets_python3_3(-)?,python_single_target_python3_3(+)?,python_targets_python3_4(-)?,python_single_target_python3_4(+)?] |
59 |
> required by (net-analyzer/fail2ban-0.9.3:0/0::gentoo, installed) |
60 |
|
61 |
|
62 |
|
63 |
Got it, finally :-) |
64 |
|
65 |
fail2ban wants sys-apps/systemd[python(-)], and systemd-219_p112 is the |
66 |
highest version with an explicit python USE flag. All later versions do |
67 |
not have the flag at all. |
68 |
|
69 |
Your choices are either to have fail2ban fixed to deal with recent |
70 |
systemd USE, and tolerate the systemd downgrade meanwhile; or to replace |
71 |
fail2ban with something equivalent |
72 |
|
73 |
|
74 |
-- |
75 |
Alan McKinnon |
76 |
alan.mckinnon@×××××.com |