| 1 |
-----BEGIN PGP SIGNED MESSAGE----- |
| 2 |
Hash: SHA256 |
| 3 |
|
| 4 |
On 24/08/12 07:48 PM, William Hubbs wrote: |
| 5 |
> On Sat, Aug 25, 2012 at 07:40:43AM +0900, heroxbd@g.o |
| 6 |
> wrote: |
| 7 |
>> Besides, IMHO, we should avoid changing OpenRC's default |
| 8 |
>> dependency too often. The solution for one user can be received |
| 9 |
>> as a regression to others. |
| 10 |
>> |
| 11 |
>> People file bugs saying "it worked for OpenRC-0.9 but not 0.10". |
| 12 |
>> For devs, we know we just changed default value of something |
| 13 |
>> perfectly configurable. But for that user, it is quite |
| 14 |
>> discouraging to feel "something in OpenRC is still unstable". |
| 15 |
> |
| 16 |
> Another thing to consider is, do the services we say "need net" |
| 17 |
> really _NEED_ net? |
| 18 |
> |
| 19 |
> If a service is a listener like sshd that isn't bound to a |
| 20 |
> specific address by default and can deal with interfaces going up |
| 21 |
> and down, I would guess that it doesn't. I am thinking that we can |
| 22 |
> set up the depends for sshd as follows: |
| 23 |
> |
| 24 |
> use net after net |
| 25 |
> |
| 26 |
> That would make sure a net provider runs before sshd if one is in |
| 27 |
> the runlevel, but allow sshd to start if one is not as well. |
| 28 |
> |
| 29 |
> This would have to be tested on a case-by-case basis. |
| 30 |
> |
| 31 |
> About changing default values, I think that if we have a good |
| 32 |
> reason to change them, we should inform users and change them. |
| 33 |
> |
| 34 |
> I think the advantage here is that it makes openrc run out of the |
| 35 |
> box in more situations. |
| 36 |
> |
| 37 |
> William |
| 38 |
> |
| 39 |
|
| 40 |
I think this may again come down to the meaning of "net" -- in the |
| 41 |
case where rc_depend_strict="no" then "net" just means that the |
| 42 |
network interface infrastructure is up and running (ie net.lo); this |
| 43 |
should be true and imo is required for something like ssh. When "net" |
| 44 |
goes beyond that and includes other interfaces (ie, |
| 45 |
rc_depend_strict="yes") then the 'need net' might be a bit strict; on |
| 46 |
the other hand if a user has things set up that way then it may very |
| 47 |
well be for a reason (for instance, I tend to prefer that sshd is |
| 48 |
started after my hotplugged iface is up and likewise goes down when |
| 49 |
that iface disappears. I don't see that happening with a "use net" |
| 50 |
case when compared against a "need net". |
| 51 |
|
| 52 |
-----BEGIN PGP SIGNATURE----- |
| 53 |
Version: GnuPG v2.0.19 (GNU/Linux) |
| 54 |
|
| 55 |
iF4EAREIAAYFAlA4KMcACgkQ2ugaI38ACPCdxQD8DWD+LOJq1V5722MUkC2tvp0i |
| 56 |
skFHngOAJNGFyW4q3gMBAJnXZ2TQE77MUjcbbWGbfXr71EBLVBcoy9vzcxHOK/Oj |
| 57 |
=yO7o |
| 58 |
-----END PGP SIGNATURE----- |
Archives