| 20 Apr 2025 |
 Andrew | does it track with a delay or something? | 16:10:39 |
 K900 | No | 16:10:41 |
| K900 | When you disable a swap device, the pages stored on it don't go poof | 16:11:05 |
| K900 | They get moved to RAM | 16:11:07 |
| K900 | These are likely those pages | 16:11:12 |
| Andrew | uhh, but how can I make them go poof? | 16:11:38 |
| K900 | You can't | 16:11:52 |
| K900 | Because something will use those pages | 16:11:58 |
| Andrew | Redacted or Malformed Event | 16:12:11 |
| Andrew | sadge | 16:12:20 |
| K900 | You can reboot your system | 16:12:26 |
| K900 | Then you won't have any pages to be evacuated from swap space | 16:12:40 |
| K900 | Or at least a lot less of them | 16:12:52 |
| K900 | Ideally you would disable swap in your config, and then reboot | 16:13:04 |
| K900 | So there's nowhere to swap things out to | 16:13:09 |
| K900 | If you want to see the behavior of the system without swap | 16:13:22 |
| K900 | Not that it'll be any different to the behavior of the system with swap | 16:13:35 |
| Andrew | what? | 16:13:51 |
| Andrew | it should accumulate less RAM by the time it hits 7 days, no? without swap | 16:14:30 |
| K900 | No? | 16:14:37 |
| K900 | Why would it? | 16:14:39 |
| Andrew | well, comparatively to what I have now | 16:15:05 |
| Andrew | I have a shit ton of swap pages in my ram, correct? | 16:15:33 |
| K900 | Yes, but those pages don't have magic swap pixie dust in them | 16:15:51 |
| K900 | They contain data that was swapped out from RAM | 16:15:58 |
| K900 | That data is still used by something | 16:16:03 |
| K900 | /proc/allocinfo can't tell what the data actually is, it only tracks where the memory was allocated | 16:16:19 |
| K900 | And this memory was allocated in "reading from swap" | 16:16:42 |
| Andrew | so you are saying that I will still have the same 12.4 GiB of RAM regardless if I have or not have swap enabled? after same amount of time passed with the same usage | 16:17:06 |
| Andrew | it will just be marked differently in the tracked file | 16:17:41 |
