| 12 Apr 2024 |
 andi- | restarted synapse.. it has been running for some time and looked rather leaky | 16:33:48 |
| andi- | also running a deployment... but I'll only be back in a couple hours to check on it | 16:34:04 |
| andi- | I hope you aren't trying to join matrix HQ as that would probably be something that creates a ton of headache | 16:35:32 |
 @rhendric:nixos.dev | Ha ha no, I only use this account for nixos.org stuff and private rooms with NixOS people. | 17:35:26 |
| andi- | finally managed to deploy a new nixos closure.. for some reason ssh-ng is transmitting only a few bytes per roundtrip... Probably some old Nix version incompatibility... Hoping the server does catch up quick and will be more responsive going forward | 21:39:40 |
 Andreas Rammhold | alright also upped the instance to 4 dedicated cores and 16GiB RAM from 2 shared cores with 4GiB of RAM. Much more responsive. | 22:01:01 |
| @rhendric:nixos.dev | Massively helpful, thank you! | 22:07:16 |
| andi- | Added a federation worker as well... Hope that keeps that python process away from 100% CPU for extended periods of time :D | 22:21:28 |
| 23 Apr 2024 |
| Andreas Rammhold | Doing a bit of DB maintenance.. we are up to 200GiB of storage required... vacuuming a few tables should reduce that slightly but cause a bit of a slowdown | 08:34:09 |
| Andreas Rammhold | * Doing a bit of DB maintenance.. we are up to 200GiB of storage required... vacuuming a few tables should reduce that slightly but cause a bit of a slowdown (while it is running) | 08:35:20 |
| Andreas Rammhold | https://discourse.nixos.org/t/matrix-account-hosting-for-nix-os-hackers/14036/7 | 08:56:25 |
| Andreas Rammhold | garr, reverted back to Postgresql12 after >10h of f*cking around with shit | 19:33:00 |
| 24 Apr 2024 |
|  @lourkeur:nixos.dev left the room. | 14:26:48 |
| 25 Apr 2024 |
|  David Mell (zraexy) joined the room. | 23:00:22 |
| David Mell (zraexy) changed their display name from David Mell to David Mell (zraexy). | 23:52:01 |
| 26 Apr 2024 |
|  @squalus:nixos.dev left the room. | 17:54:43 |
| 17 May 2024 |
|  @shadowrz:halogen.place joined the room. | 00:02:58 |
| @shadowrz:halogen.place | Ping @andi:kack.it Looks like medias on nixos.dev broke, and also there's some strange HTTP errors when syncing. | 00:04:38 |
| andi- | I can't see anything wrong. I've recently (>1 month ago) started purging 14d old remote content. If it was old pictures that might have been the issue. I'm currently keeping thumbnails of old images for >30d as those require a lot less space. | 10:30:41 |
| 18 May 2024 |
 Yorusaka Miyabi | In reply to @andi:kack.it
I can't see anything wrong. I've recently (>1 month ago) started purging 14d old remote content. If it was old pictures that might have been the issue. I'm currently keeping thumbnails of old images for >30d as those require a lot less space. OK, Understand. | 01:51:08 |
| Yorusaka Miyabi changed their profile picture. | 03:33:19 |
| Yorusaka Miyabi changed their profile picture. | 03:51:40 |
| 19 Jun 2024 |
| @shadowrz:halogen.place | Ping @andi- Looks like the server broke again 🫠| 00:00:15 |
|  hand-unhand-locative-stony joined the room. | 01:41:44 |
| andi- | In reply to @shadowrz:halogen.place
Ping @andi- Looks like the server broke again 🫠Fixe, sorry :-) | 02:11:30 |
| andi- | In reply to @shadowrz:halogen.place
Ping @andi- Looks like the server broke again 🫠* Fixed, sorry :-) | 02:11:34 |
| Yorusaka Miyabi | In reply to @andi:kack.it
Fixed, sorry :-) BTW, did you set up local media retention and auto compressor? | 04:09:04 |
| andi- | In reply to @shadowrz:nixos.dev
BTW, did you set up local media retention and auto compressor? No retention configured for local media. I am running the compressor daily. Sadly the database continues to grow unbounded. At 250GB now. I’ll have to give the Postgres migration another stab. A few months back I tried version 15 but synapse was not happy with that. It never caught back up. | 06:44:44 |
| Yorusaka Miyabi | In reply to @andi:kack.it
No retention configured for local media. I am running the compressor daily. Sadly the database continues to grow unbounded. At 250GB now. I’ll have to give the Postgres migration another stab. A few months back I tried version 15 but synapse was not happy with that. It never caught back up. Do you plan to have media retention configured? | 13:07:39 |
| andi- | In reply to @shadowrz:nixos.dev
Do you plan to have media retention configured? Maybe. If it becomes a bottleneck. So far it is not a concern so no. | 15:24:51 |
