| 1 |
On 2021-01-04 10:23, Michał Górny wrote: |
| 2 |
> Not modifying an existing user is a horrible default that has already |
| 3 |
> bricked one system (by removing /dev/null). So, over my dead commit |
| 4 |
> access. |
| 5 |
|
| 6 |
Have you seen how many user were hit caused by the recent rebuilt on |
| 7 |
2020-12-28 and are already complaining/asking for help through various |
| 8 |
channels? |
| 9 |
|
| 10 |
It's like asking for service auto-restart support in PMS as requested as |
| 11 |
part of current OpenSSH upgrade because if you move from <8.3_p1 to |
| 12 |
>=8.3_p1 and don't restart OpenSSH in time, you can get locked out. |
| 13 |
|
| 14 |
However, an easily looking solution like |
| 15 |
|
| 16 |
> Just add something like |
| 17 |
> |
| 18 |
> if [[ -d /run/systemd/system ]]; then |
| 19 |
> systemctl try-restart sshd |
| 20 |
> else |
| 21 |
> rc-service -q --ifstarted sshd restart |
| 22 |
> fi |
| 23 |
> |
| 24 |
> to pkg_postinst |
| 25 |
|
| 26 |
is wrong because even if it works for *some* users it won't work for all |
| 27 |
users but has the potential to cause major problems. |
| 28 |
|
| 29 |
That's why we have elog and newitem system. However, 8.3 is in |
| 30 |
repository for while and multiple people forgot about the newitem and |
| 31 |
didn't pay attention to elog messages. While I agree that it's a problem |
| 32 |
when you lose access to a remote box you don't have physical access to, |
| 33 |
this reached a level where I have to say, |
| 34 |
|
| 35 |
> We cannot rescue/protect everyone. |
| 36 |
|
| 37 |
Back to topic, acct-* stuff: |
| 38 |
|
| 39 |
Like already said in February 2020 when I joined a thread created by a |
| 40 |
user posting same concerns: |
| 41 |
|
| 42 |
There is a reason why *no* distribution on this planet is trying to mess |
| 43 |
with existing data/configurations: Every attempt trying to analyze given |
| 44 |
setup to apply required changes to fix/migrate something automatically |
| 45 |
has been prone to fail the long run. |
| 46 |
|
| 47 |
Please get some experience from real world. Preferable from running |
| 48 |
headless systems not just for yourself and where you are not the only |
| 49 |
person touching the system. |
| 50 |
|
| 51 |
When I worked on bug 605008 long time ago for example, I also ended up |
| 52 |
over-engineering. There is stuff you cannot fix. I am still thinking |
| 53 |
about creating everything the way it should look like in $D and report |
| 54 |
any difference like changed file permissions to user on merge to allow |
| 55 |
them to notice (an improvement, now user only have to pay attention and |
| 56 |
you need to solve the additional problem that the more information you |
| 57 |
present all the time, the more information will be ignored). But |
| 58 |
sometimes users are making changes we wouldn't do, not recommend or just |
| 59 |
don't understand at first. That all doesn't matter: We have to keep in |
| 60 |
mind that these aren't our systems and we have to respect whatever the |
| 61 |
user did on their system. |
| 62 |
|
| 63 |
|
| 64 |
-- |
| 65 |
Regards, |
| 66 |
Thomas Deutschmann / Gentoo Linux Developer |
| 67 |
fpr: C4DD 695F A713 8F24 2AA1 5638 5849 7EE5 1D5D 74A5 |
Archives