Warn when the in-memory filesystem is getting full
When the internal memory is full, downloads fail without warning and extractions also fail.
GNOME has a notification mechanism for when internal hard disks get full. See https://search.disconnect.me/image?l=aHR0cHM6Ly9lbmNyeXB0ZWQtdGJuMC5nc3RhdGljLmNvbS9pbWFnZXM/cT10Ym46QU5kOUdjUjdDMVE4ZkVvMHlyRnhBcGQwWUFscDBFRmhScVc2WXV0bjNNTU81RldWampXRnZuMTBPUQ==.
We should have something similar in Tails for RAM (which is our internal disk somehow). The message should be custom and recommend to:
- Remove files (downloads, etc.) or move them to the persistent volume.
- Close applications
#3 Updated by sajolida about 4 years ago
- Description updated (diff)
- Status changed from New to Confirmed
- Assignee deleted (
- Priority changed from Low to Normal
- QA Check deleted (
- Type of work changed from Debian to Code
In the description of the ticket you are mentioning "memory" getting full, but in your last comment you are mentioning "disk" getting full. I think both cases should be considered separately.
Regarding disks (external USB sticks, persistence, internal disks), I think we should have the usual GNOME notification about "low disk space": https://search.disconnect.me/image?l=aHR0cHM6Ly9lbmNyeXB0ZWQtdGJuMC5nc3RhdGljLmNvbS9pbWFnZXM/cT10Ym46QU5kOUdjUjdDMVE4ZkVvMHlyRnhBcGQwWUFscDBFRmhScVc2WXV0bjNNTU81RldWampXRnZuMTBPUQ==.
I tested and this mechanism doesn't work for the persistant volume, so I create a ticket just for that: #10064.
Regarding memory, yes, it would be super cool to have a warning about memory getting full as well. I'm surprised we don't have a ticket about that already... So I'm renaming this ticket and I'll let BitingBird mark it as a duplicate if we already have something about that elsewhere.
#5 Updated by intrigeri about 4 years ago
- Assignee set to DanOver
- QA Check set to Info Needed
When saving stuff to non-persistent places in the filesystem, since Tails 1.5 one can (at least manually) see the remaining "disk" space (really: allocated memory to the read-write branch on
df /. I think this wasn't the case with 1.4.1, so perhaps this has been fixed => DanOver, can you reproduce this problem with Tails 1.5?