1 |
On 1/20/20 5:08 PM, Alec Warner wrote: |
2 |
> |
3 |
> So I can describe in detail one example, but its not running Gentoo; so |
4 |
> I'm not sure if you care in practice. |
5 |
|
6 |
Yes, I'm happy to see a real example. |
7 |
|
8 |
|
9 |
> At work we had sec=krb5 NFS v3 mounted home directories. They were |
10 |
> mounted in /home (via the automounter.) So if these machines ran Gentoo |
11 |
> and you went to do something like "create /home/amavisd" it would fail |
12 |
> because the root user doesn't have the ability to make home directories |
13 |
> in /home (uid=0 is mapped to nobody, who doesn't have +w on /home.) All |
14 |
> home directories were created by a business application and there were |
15 |
> specific hosts where root was not squashed (and we used sec=sys instead |
16 |
> of krb5) and so root on the admin host would have +w on /home and not be |
17 |
> squashed to nobody.) |
18 |
> |
19 |
> In practice in that enterprise environment, if we needed something like |
20 |
> /home/web/ (which I think did exist at one point) we would create a role |
21 |
> account in LDAP (www-data is a common user for example), assign it a |
22 |
> uid, create the homedirectory (/home/web) and it would be owned by |
23 |
> www-data:www-data. Then we would configure the web front ends to use |
24 |
> www-data instead of the normal user (apache or nginx or whatever.) |
25 |
|
26 |
That's all relatively normal. As I've said, a human uses the "amavis" |
27 |
account. Yes, the install of acct-user/amavis would crash because it |
28 |
can't create the home directory, but I contend that crashing is the best |
29 |
thing to do. |
30 |
|
31 |
When the acct-user ebuild crashes, you get to ask yourself if you want |
32 |
his home directory to be shared among the people with authority to |
33 |
release spam from the quarantine. I'm betting you would, and that you |
34 |
would therefore add the account to LDAP and start over. Same deal as |
35 |
apache/web, and you don't have to involve an overlay to do the right |
36 |
thing. In this case, the fact that we used /home was a boon, because it |
37 |
helped you accomplish what you were trying to accomplish by sharing |
38 |
/home in the first place. |
39 |
|
40 |
If you don't want to share the home directory... well, no harm done. |
41 |
You'll have to override the ebuild to tell it what location to use as an |
42 |
alternative. But I think this case is somewhat less likely, and the base |
43 |
rate was already single digits. |
44 |
|
45 |
If only good exceptions are made (with home directories that people |
46 |
would actually want to share under /home), this approach does a little |
47 |
good and no bad. |
48 |
|
49 |
|
50 |
|
51 |
> (2) I don't think most people running Gentoo are running these |
52 |
> environments, which is why you don't see many practical objections on |
53 |
> the list. I think it's reasonable to avoid service account homedirs in |
54 |
> /home not because of fancy examples like above (that maybe 10 companies |
55 |
> in the world run) and instead just focus on this idea that "system stuff |
56 |
> doesn't go in /home." Its somewhat arbitrary as mgorny points out |
57 |
> earlier in the thread. |
58 |
|
59 |
I was never discounting these sorts of environments. On the contrary, |
60 |
the point I'm trying to make above appeared somewhere in the discussion |
61 |
with rich0, but it's hard to articulate without details. |
62 |
|
63 |
If it's arbitrary and we admit that, I'm fine with it. I'm moving on |
64 |
with my life. QA can choose what kind of sauce users get on their turd |
65 |
sandwich =P |