1 |
On Sat, Aug 08, 2020 at 11:38:36PM +0200, Jaco Kroon wrote: |
2 |
> |
3 |
> -----BEGIN PGP SIGNED MESSAGE----- |
4 |
> Hash: SHA256 |
5 |
> |
6 |
> Hi, |
7 |
> |
8 |
> On 2020/08/08 22:57, William Hubbs wrote: |
9 |
> > On Sat, Aug 08, 2020 at 09:17:20PM +0100, Roy Bamford wrote: |
10 |
> >> On 2020.08.08 19:51, William Hubbs wrote: |
11 |
> >>> All, |
12 |
> >>> |
13 |
> >>> I would like to propose that we switch the default udev provider on |
14 |
> >>> new |
15 |
> >>> systems from eudev to udev. |
16 |
> >>> |
17 |
> >>> This is not a lastrites, and it will not affect current systems since |
18 |
> >>> they have to migrate manually. Also, this change can be overridden at |
19 |
> >>> the profile level if some profile needs eudev (the last time I |
20 |
> >>> checked, |
21 |
> >>> this applies to non-glibc configurations). |
22 |
> >>> |
23 |
> >>> What do people think? |
24 |
> >>> |
25 |
> >>> Thanks, |
26 |
> >>> |
27 |
> >>> William |
28 |
> >>> |
29 |
> >>> |
30 |
> >> |
31 |
> >> William, |
32 |
> >> |
33 |
> >> With the declared aim from upstream of making udev inseparable from |
34 |
> >> systemd, its not something to be done lightly. |
35 |
> >> That's the entire reason that eudev was necessary. |
36 |
> > |
37 |
> > Eudev never became necessary unless you are using a non-glibc system, |
38 |
> > and as I said, this can be handled in the profiles. |
39 |
> > Udev runs completely fine without systemd, so I fail to see how eudev |
40 |
> > is necessary for most of Gentoo. |
41 |
> |
42 |
> It actually works is enough reason for me. Was forced to migrate a |
43 |
> bunch of systems not six months back from systemd-udev to eudev because |
44 |
> systemd-udev is absolutely terrible w.r.t. race conditions resulting in |
45 |
> lockups that kept forcing us into manual intervention situations. |
46 |
> Mostly on systems with LVM. |
47 |
|
48 |
I don't exactly know what your situation is, but as I said, this |
49 |
proposal wouldn't affect your systems. I'm not talking about lastrites |
50 |
for eudev, just making it the default for new installs. |
51 |
|
52 |
> I'm completely against the proposal. |
53 |
> |
54 |
> >> I would want some convincing that it was not another step on the road |
55 |
> >> to Gentoo being assimilated by systemd. |
56 |
> >> |
57 |
> >> We had this discussion several years ago when the default became |
58 |
> >> eudev. What's changed? |
59 |
> > |
60 |
> > If systemd folks do make udev inseparable from systemd, then we would |
61 |
> > need eudev to be the default, but as I see it right now, there is not |
62 |
> > a case for it being the default. |
63 |
> |
64 |
> Other than that it works and the systemd version does not. Might be |
65 |
> configuration dependent, but I don't expect a default udev |
66 |
> configuration/system side to not cause LVM breakages when running |
67 |
> commands as simple as "lvs". eudev in coparison just works. |
68 |
|
69 |
I don't know what is going on with your systems, but I suspect possible |
70 |
configuration dependence. |
71 |
|
72 |
When are the breakages happening-- just at random or during bootup? |
73 |
|
74 |
William |
75 |
|
76 |
> > |
77 |
> > Another thing to consider is bus factor (eudev is maintained by one |
78 |
> > person primarily, so I have doubts about its viability as the default. |
79 |
> |
80 |
> Yes, this is a problem. |
81 |
> |
82 |
> Kind Regards, |
83 |
> Jaco |
84 |
> |
85 |
> -----BEGIN PGP SIGNATURE----- |
86 |
> |
87 |
> iQEzBAEBCAAdFiEEyyCUcKjG7P5BDam8CC3Esa/37p4FAl8vG1AACgkQCC3Esa/3 |
88 |
> 7p7Yvgf6Apoi1oCUKSyLEvH8fAEgbMIODULJAZx5+/C1dbROdjkWEzTTp3pNjWiQ |
89 |
> u8S2qz3xmh9QmKBwTAxB38U/gqXVRpF+xYfSF7K/CDUVcfAg5ViTL3W7YeJMPFNa |
90 |
> Jk8BgrarAc1Ln8OXCJ37Gf0eeuyBTsQQQ5qqubzNjdLBhrZegWY57gElrItE0Ywb |
91 |
> IjVBUO4QX3PSoOpZ5UlIo8Ioh+8ANXc/ADg7wASVQkd3dciyewZasZho/q6cNn6W |
92 |
> c44aMNFRTeiUfcK4+bpGMslq70y7D7JITkjkP+9e68e8wkh93L8fVs4BszBYEtUY |
93 |
> G6IXc4QtJ5Jf3bQRbyCnGcFYXrSLgg== |
94 |
> =rF5/ |
95 |
> -----END PGP SIGNATURE----- |
96 |
> |
97 |
> |