Conversation
811d4a1 to
f99527f
Compare
f99527f to
4f7b850
Compare
4f7b850 to
393e901
Compare
This comment was marked as outdated.
This comment was marked as outdated.
|
ugh |
|
I think its an issue with how nix.conf was setup by nixpkgs-review-gha. Let me see if I can fix that. I don't know about darwin and linux-builder on darwin etc. I will try to fix the aarch64-linux tests. |
|
doronbehar
left a comment
There was a problem hiding this comment.
Changes look good to me. Let's try to figure out what's up with the nixpkgs-review potentially false failures.
6e55dbf to
cd51773
Compare
|
Added the |
cd51773 to
a939c13
Compare
SigmaSquadron
left a comment
There was a problem hiding this comment.
Looks good to me! Let's hope that the nixpkgs-review failures are flukes and this deprecated encryption system isn't somehow load-bearing
|
|
Oh no :< |
There was a problem hiding this comment.
also lgtm, aarch64-linux nixostest failure seems weird, test timed out
You have to add EXTRA_NIX_CONFIG over at your repo's https://github.com/phanirithvij/nixpkgs-review-gha/settings/variables/actions with a value of system-features = nixos-test benchmark big-parallel kvm uid-range to fix the aarch64-linux issues
|
I'm the author of |
So this works now after you added it? |
|
Yes, see #479934 (comment) |
Also, the PR adds evaluation errors that should warn them prior to that state of unmountable homes. |
|
@Sigmanificient please write a message here: https://discourse.nixos.org/t/breaking-changes-announcement-for-unstable/17574/116 |
|
@Sigmanificient please add a link to this PR in the thread above 🙏 . |
|
This pull request has been mentioned on NixOS Discourse. There might be relevant details there: https://discourse.nixos.org/t/breaking-changes-announcement-for-unstable/17574/117 |
|
I'll include my "quick migration to fscrypt guide" here in case it helps someone:
Help the next person by commenting back here with any tweaks that you needed. |
|
ecryptfs seems to be receiving active updates. There's some discussion about possibly beginning a slow deprecation procedure (starting with removing write support) in the future. fscrypt isn't really a viable migration path for people using ecryptfs on unsupported or uncontrolled filesystems (e.g. remote cloud backups, which was always a primary target for ecryptfs). Maybe CryFS or gocryptfs, though performance seems to be worse and (since they aren't maintained in the kernel) I'd consider stability guarantees to be a little less reliable, which is always a bit scary for this sort of thing — nobody wants to find their backups unrecoverable in a few years. |
Just followed this procedure for my migration. Two notes:
Apart from that, the procedure worked fine. |
While looking into python2 remaining bits, i come across this package, and it seems appropriate to remove it.
ecryptfshasn't been updated sine May 2016, and seems to have questionable maintenance, which is a huge security concern for a filesystem.ecryptfs-helperwas also dropped a while ago.Things done
passthru.tests.nixpkgs-reviewon this PR. See nixpkgs-review usage../result/bin/.Add a 👍 reaction to pull requests you find important.