| 22 Oct 2024 |
 bumperboat | why is --disk-usage say theres only 16M in use when there is 4.4G | 02:29:47 |
| bumperboat | and i have to specify --directory /var/log/journal to get the right answer | 02:30:14 |
| bumperboat | i think its some goofy msft sytemd thing | 02:33:01 |
| bumperboat | interested in if yours does the same tho | 02:33:39 |
 Arson Copperflame | I have the same behavior
NixOS nzbr@pulsar
ZS ~> journalctl --disk-usage
Archived and active journals take up 16.0M in the file system.
NixOS nzbr@pulsar
ZS ~> journalctl --disk-usage --directory /var/log/journal
Archived and active journals take up 3.8G in the file system.
| 02:37:19 |
| Arson Copperflame | Seems to work as intended on Ubuntu | 02:38:57 |
| Arson Copperflame | I don't see anything related to journald in mount though | 02:39:30 |
| bumperboat | if i do journalctl --firectory /var/log/journal --vacuum-size 3G it goes through the files that it wants to delete or rewrite, but it doesnt do anything to them | 02:57:46 |
| bumperboat | * if i do journalctl --directory /var/log/journal --vacuum-size 3G it goes through the files that it wants to delete or rewrite, but it doesnt do anything to them | 02:58:00 |
| bumperboat | and my ubuntu wsl does have the right behaviour as well | 02:59:26 |
| Arson Copperflame | I looked through the systemd derivation and the patches it applies, but I didn't find anything that would explain this | 03:06:34 |
| bumperboat | why does /etc/machine-id change on every boot? | 03:46:02 |
| bumperboat | all the dirs in my /var/log/journal are from the various machine-id's ever since i created it | 03:47:25 |
| bumperboat | and if i do journalctl --list-boots it will only show the one boot because the machine-id changed | 03:48:02 |
| bumperboat | oh its tmpfs thats fun | 03:53:43 |
| bumperboat | yeah so journalctl only looks at directories in /var/log/journal/$(cat /etc/machine-id) so it is disregarding the logs from all the previous boots when vacuuming, or anything really | 04:09:44 |
| bumperboat | * yeah so journalctl only looks at files in /var/log/journal/$(cat /etc/machine-id) so it is disregarding the logs from all the previous boots when vacuuming, or anything really | 04:11:11 |
 K900 | Hmm | 05:58:40 |
| K900 | I wonder if that's an us problem | 05:58:47 |
| K900 | Or a WSL problem | 05:58:54 |
| bumperboat | is a static file on my ubuntu wsl | 06:10:52 |
| bumperboat | from reading machine-id (5) systemd will create a random id when /etc is RO for live systems | 06:12:04 |
| bumperboat | so thats my hunch | 06:15:29 |
| bumperboat | running systemd-machine-id-setup --commit manually resolves. but this should be ran by systemd-machine-id-commit.service (8) once /etc is writable - dunno why it isnt starting | 08:10:33 |
 common16 | Hey folks, I'm jumping back into this and I'm noticing my SSH agent is timing out when trying to push / pull from GH. I'm not quite sure how this should be configured but my goal is to completely remove everything from the Windows side and lean into doing everything in WSL | 17:10:28 |
| common16 | Here's what my flake currently looks like: http://pastie.org/p/3N2t6Kw1PGAZ6QXtemPcXB | 17:10:41 |
| common16 | Every time I start WSL I'm using ssh-add to manually add my GH creds but I'm not sure if it's being added to the wrong(?) ssh-agent or what's going on | 17:11:16 |
| common16 | I also removed git completely from the Windows side of my machine so I'm not even sure if the .gitconfig is still valid to "inject" into my WSL(?) | 17:12:16 |
| bumperboat | without seeing an error - my guess since I do not see any configuration to start and configure a user ssh-agent within your user environment that the ssh-add will fail | 22:08:38 |
| bumperboat | i would suggest look at using home-manager for your user it can be configured to start the agent for you and manage all the things (tm) | 22:10:22 |
