| 19 Apr 2026 |
 Atemu | Redacted or Malformed Event | 21:43:22 |
| Atemu | Redacted or Malformed Event | 21:43:51 |
| 20 Apr 2026 |
 oak 🏳️🌈♥️ | Seriously though, what's the governance model of this open source project? And has there been discussions or decisions where to refer to wrt. use of "agentic coding"? I think NixOS project itself has been adopting it quite rapidly | 01:08:41 |
 pentane | -> #offtopic:nixos.org | 05:46:24 |
| pentane | oh and btw, just to reiterate what I said earlier, I won't merge any LLM-generated or LLM-assisted contributions, ever. | 06:34:12 |
| pentane | looked into this just now btw and it seems like LOS did us the favour of documenting the "API" the update server needs to have: https://github.com/LineageOS/android_packages_apps_Updater/blob/lineage-23.2/README.md | 06:35:14 |
| pentane | having dug into this a bit deeper - the JSON response format dorsn't seem to have changed at all, but we rather need to change the resource URL we set via the product RRO in modules/apps/updater.nix | 07:26:38 |
| pentane | ok nvm. the problem solely seems to be that the overlay doesn't get applied correctly | 22:07:51 |
| pentane | namespace collision with some stock LineageOS overlay maybe? | 22:08:06 |
| 23 Apr 2026 |
| pentane | robotnix is not an official NixOS project, and has no formal governance model except for a de-factor meritocracy à la "whoever steps up to maintain it gets to call the shots" | 18:07:24 |
 magic_rb | And the current BDFL has a more sane take than the nixos project thankfully | 18:09:24 |
| pentane | but still </topic> please | 18:11:13 |
| pentane | * but still, </topic> please | 18:11:19 |
| pentane | * robotnix is not an official NixOS project, and has no formal governance model except for a de-facto meritocracy à la "whoever steps up to maintain it gets to call the shots" | 18:12:19 |
| pentane | Oh lol, this room got matrixed again as it seems | 18:37:26 |
| pentane | Okay so I think I've narrowed down the problem to a namespace collision w.r.t. Updater__lineage_FP4__auto_generated_rro_product | 21:45:08 |
| pentane | I am however unsure whether the two auto-generated overlays (the stock lineage one and the robotnix one) should be merged, or whether they should be two different APKs with distinct names | 21:46:22 |
| 27 Apr 2026 |
|  tom joined the room. | 13:24:10 |
|  bigfreakintaco joined the room. | 15:35:26 |
| bigfreakintaco | Hello Everyone :) | 15:37:37 |
| bigfreakintaco | I just want to say thank you to all the devs/contributors to this project! | 15:56:05 |
 chrillefkr | Hey, anyone gotten this issue before?
$ nix run .\#robotnixConfigurations.dailydriver.releaseScript -- keys
Signing target files
....
signing: com.android.adbd.apex container (/dev/shm/robotnix_keys.XXXXGNTOWw/com.android.adbd)
: com.android.adbd.apex payload (/dev/shm/robotnix_keys.XXXXGNTOWw/com.android.adbd.pem)
signing: com.android.adservices.apex container (/dev/shm/robotnix_keys.XXXXGNTOWw/com.android.adservices)
: com.android.adservices.apex payload (/dev/shm/robotnix_keys.XXXXGNTOWw/com.android.adservices.pem)
NOT signing: com.android.apex.cts.shim.apex
(skipped due to special cert string)
signing: com.android.appsearch.apex container (/dev/shm/robotnix_keys.XXXXGNTOWw/com.android.appsearch)
: com.android.appsearch.apex payload (/dev/shm/robotnix_keys.XXXXGNTOWw/com.android.appsearch.pem)
Traceback (most recent call last):
File "<frozen runpy>", line 198, in _run_module_as_main
File "<frozen runpy>", line 88, in _run_code
File "/nix/store/09k3hq5374awmdc5x43lqlbin4fwk6xn-ota-tools/bin/sign_target_files_apks/__main__.py", line 52, in <module>
File "/nix/store/09k3hq5374awmdc5x43lqlbin4fwk6xn-ota-tools/bin/sign_target_files_apks/__main__.py", line 46, in _soong_main
File "<frozen runpy>", line 198, in _run_module_as_main
File "<frozen runpy>", line 88, in _run_code
File "sign_target_files_apks.py", line 1958, in <module>
File "sign_target_files_apks.py", line 1933, in main
File "sign_target_files_apks.py", line 828, in ProcessTargetFiles
File "apex_utils.py", line 518, in SignApex
File "apex_utils.py", line 443, in SignCompressedApex
File "apex_utils.py", line 358, in SignUncompressedApex
File "apex_utils.py", line 95, in ProcessApexFile
apex_utils.ApexSigningError: Failed to find signing keys for apk file /tmp/tmptg_lzm5z/priv-app/com.android.appsearch.apk@BP4A.251205.006/com.android.appsearch.apk.apk in apex /tmp/original-apex-jkgopzve.apex. Use "-e <apkname>=" to specify a key
| 19:15:53 |
| pentane | are you trying to sign a lineageos build perchance? | 20:50:48 |
| chrillefkr | Yeahh, for Fairphone 4 | 21:06:08 |
| pentane | hm, are you creating an entirely new build, or updating an existing one? | 21:10:51 |
| pentane | Because if you're creating a new one, I'd recommend setting stateVersion = "3" | 21:11:19 |
| pentane | (signing is known to be somewhat brittle with the earlier state versions) | 21:11:34 |
| chrillefkr | Updating existing one, from back in february. I've set stateVersion to 1, as it seemed to fit the instructions given, after I had updated flake inputs. What is stateVersion really? | 21:12:35 |
| pentane | Ah okay, then you should keep your stateVersion fixed to "1" | 21:13:10 |
| pentane | I'm a bit surprised though that the signing seems to have worked back then, I thought LOS signing was pretty much consistently broken until ~march | 21:13:43 |
