1 |
-----BEGIN PGP SIGNED MESSAGE----- |
2 |
Hash: SHA256 |
3 |
|
4 |
On 07/08/15 11:30 AM, William Hubbs wrote: |
5 |
> On Thu, Aug 06, 2015 at 08:07:44PM -0400, Ian Stakenvicius wrote: |
6 |
>> |
7 |
>> Can we get "nofail" immediately in the mount -a variants of |
8 |
>> localmount/netmount and expand that in netmount to make the |
9 |
>> nfsclient dep be a "use" or a "need" depending on if it's set or |
10 |
>> not?? That would imo kill the existing bug that started all of |
11 |
>> this too. |
12 |
> |
13 |
> Sure, I can get the nofail support in pretty quick, and I think |
14 |
> that is a feature we should have. |
15 |
> |
16 |
> Right now, netmount is using the use dependency to make sure |
17 |
> network file system utilities are started before us. Because of |
18 |
> the all-or-nothing nature of netmount, we can't switch those |
19 |
> dependencies to need. It would cause netmount to fail if one of |
20 |
> those utilities fails to start. The use dependency is the best one |
21 |
> we can use at this time, and a migration path was specifically laid |
22 |
> out in the news item. |
23 |
> |
24 |
|
25 |
My thinking here is that, unless network mounts in fstab are listed as |
26 |
'nofail', that netmount failing due to the dependent services not |
27 |
being able to start would be a valid case. Sysadmins that don't want |
28 |
netmount to fail no matter what would be able to use 'nofail' to |
29 |
ensure that happens. |
30 |
|
31 |
This is of course predicated on (1) it being a good idea, and (2) |
32 |
fstabinfo or whatever the check currently is that would add nfsclient |
33 |
to depend() could easily swap 'use' for 'need' based on the (lack of) |
34 |
existence of the nofail attribute in fstab. |
35 |
|
36 |
|
37 |
-----BEGIN PGP SIGNATURE----- |
38 |
Version: GnuPG v2 |
39 |
|
40 |
iF4EAREIAAYFAlXE2JAACgkQAJxUfCtlWe2UDQEAhuuOs8OCR3X1NI97MSNt652V |
41 |
V7wmM9Fjy2GXGOEXFaUA/3vjrKTV8R+jKw4pxeHRYvhSiRw4SPEb71oGOdRKccE+ |
42 |
=Tr3F |
43 |
-----END PGP SIGNATURE----- |