| 5 Jul 2021 |
Zhaofeng Li | That's weird, given that port numbers are much more precious than MAC and IPv4 addresses | 19:40:55 |
@hexa:lossy.network | uh, I think the important ports are allocated, lots of them have useless allocations | 19:42:14 |
@hexa:lossy.network | and everyone else reuses ports despite their official allocation | 19:42:27 |
@hexa:lossy.network | so what does it matter | 19:42:30 |
@hexa:lossy.network | https://www.iana.org/form/ports-services | 19:43:14 |
@hexa:lossy.network | here you go | 19:43:17 |
@hexa:lossy.network | make it good! | 19:43:24 |
| spacesbot - keeps a log of public NixOS channels | 19:49:35 |
| 6 Jul 2021 |
| nurelin joined the room. | 11:44:03 |
@hexa:lossy.network | dotlambda: let's start with an archive.org pinned version and add a TODO marker | 13:48:53 |
@hexa:lossy.network | can't be bothered to get this right until tomorrow | 13:49:34 |
@hexa:lossy.network | In reply to @hexa:lossy.network
_________________ ERROR collecting tests/test_itemsequence.py __________________
ImportError while importing test module '/build/python-crfsuite-0.9.7/tests/test_itemsequence.py'.
Hint: make sure your test modules/packages have valid Python names.
Traceback:
/nix/store/2nhfiak8a30vw67mxksc2kdb69np2jcw-python3-3.8.9/lib/python3.8/importlib/__init__.py:127: in import_module
return _bootstrap._gcd_import(name[level:], package, level)
tests/test_itemsequence.py:5: in <module>
import pycrfsuite
pycrfsuite/__init__.py:2: in <module>
from ._pycrfsuite import *
E ModuleNotFoundError: No module named 'pycrfsuite._pycrfsuite'
it imported the source module instead of the built one, and that went nowhere. rm'ing the src module was the solution here. | 13:51:43 |
| iclanzan set a profile picture. | 15:57:45 |
@hexa:lossy.network | dotlambda: looks like we have more time. https://github.com/home-assistant/core/pull/52573 | 18:32:20 |
| spacesbot - keeps a log of public NixOS channels changed their display name from spacesbot to spacesbot - keeps a log of public NixOS channels. | 22:11:46 |
| 8 Jul 2021 |
iclanzan | Iโve updated my NixOS channel, and now some things in home-assistant are broken. I think itโs from the hardening changes that were made to the systemd service but I am having a hard time figuring out how to fix these issues.
First, hass can no longer read images I was using with a camera entity. I have added the relevant directoy to allowlist_external_dirs and made sure the directory and files are readable by the hass user but no dice. I am still getting permission errors in hass.
Secondly, I am getting permission errors from the z-wave integration when trying to read the z-wave controller under /dev/ttyUSB0.
I would appreciate any pointers in the right direction!
| 03:19:57 |
@andi:kack.it | Perhaps start by stating which channel that is on? :) | 08:04:39 |
@hexa:lossy.network | and whether you are using zwave or zwave_js | 11:24:56 |
@hexa:lossy.network | for the serial access for zwave I'll provide the fix with tdoays update | 11:35:05 |
@hexa:lossy.network | * for the serial access for zwave I'll provide the fix with todays update | 11:35:13 |
@hexa:lossy.network | the latest update of home-assistant has tied multicast membership requests into a central fixture ._. | 12:19:42 |
@hexa:lossy.network | basically the hass fixture (a toy home-assistant instance) tries to to ssdp | 12:21:47 |
@etu:semi.social | ugh, zwave_js, I don't feel like migrating my zwave ๐ | 12:23:13 |
@etu:semi.social | And why mix in js? | 12:23:25 |
@etu:semi.social | It's already messy enough | 12:23:32 |
@hexa:lossy.network | I kinda wish we had a dummy netdev in the stdenv, so tests could find interfaces, subscribe to multicast groups | 12:23:40 |
@hexa:lossy.network | and have that interface lead nowhere | 12:23:43 |
@hexa:lossy.network | etu: it is a reimplementation of zwave-js in pure python | 12:23:53 |
@etu:semi.social | ah | 12:24:00 |
@etu:semi.social | I still have hopes for another future than hass ๐ | 12:24:14 |