New (probably unsafe) binary digest #1457
Replies: 3 comments 2 replies
|
because sidestore doesn't want to be distributor of apple libs...we will be removing these too maybe in future and will ask user to unpack and prepare the libs for which ss team can provide a script or something... but basically these features are in testing for now. by next release it should have been settled. also there are lots of things going on, once we have slowed down and stabilized we can better take these things up. |
|
asking users to follow new procedure rn, when things are in flux is gonna be a long shot to educate all, so stay tuned for updates...once we are near ready, we will be cleaning this up. |
|
Thank you for the answer! However, keep in mind that I find that to be a shaky argument against safety of a lot of devices :)
By the way, who operates zzz.haus? Is it you? Do you trust them? I suggest that the app stores the SHA-256 of each .so file as a constant. The app must stop with an error when a file does not match. A hash of an Apple file does not make SideStore a distributor of that file. The same check also works for the future flow, where users prepare the libraries themselves. |
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
@mahee96
I noticed this was introduced in commit 02b41c2, with
defaultODAMetadataURLpointing tohttps://zzz.haus/oda.json, which distributeslibCoreADI.soandlibstoreservicescore.so.I checked the bundled binaries: libCoreADI.so is byte identical to the copy in the Apple signed Apple Music 6.5.2 APK, and libstoreservicescore.so also matches known Apple Music copies.
Could you clarify why these authentication related executable libraries are served from that domain instead of an Apple controlled source? It would also seem safer to verify them against a pinned hash or other trusted signature before use.
All reactions