feat: show the kernel's image loading progress while resuming from hibernation
The kernel freezes every process before it loads a hibernated image, so the screen looked hung for as long as the load took. A drop-in on systemd-hibernate-resume.service now opens the kernel console just before resume, draws a band across the middle of tty12 and confines the kernel's progress lines to it. The running bootvine stops drawing and stops taking the console back meanwhile, and repaints if no image was restored.
This commit is contained in:
17 files changed
+433
-13
No files matched your search
@@ -83,6 +83,35 @@ The `scripted-initrd` machine uses NixOS's older scripted initrd, which has no
|
||||
systemd. The initrd's lines scroll past as usual, and bootvine starts in
|
||||
stage 2, from `sysinit.target` on.
|
||||
|
||||
## Resuming from hibernation
|
||||
|
||||
Resuming from hibernation looks different from a boot, because the kernel
|
||||
freezes every process before it loads the image, bootvine included. The tree
|
||||
stops moving, and a band across the middle of the screen shows the kernel's own
|
||||
progress:
|
||||
|
||||
```
|
||||
─── Looking for a hibernated session ─────────────────────────────
|
||||
PM: Image loading progress: 60%
|
||||
|
||||
──────────────────────────────────────────────────────────────────
|
||||
```
|
||||
|
||||
The text comes straight from the kernel, which is why it is a line of text and
|
||||
not a bar. The screen stays like this until the restored system takes over.
|
||||
Bootvine does not draw again, because the restore replaces it along with the
|
||||
rest of the boot kernel's memory.
|
||||
|
||||
The last lines show how the time was spent. `Read ... in N seconds` is the
|
||||
image load. `Suspending console(s)` means the load has finished and the kernel
|
||||
is about to jump into the restored system. If the screen stays on that line for
|
||||
a long time, the wait is the restored kernel bringing devices back up, not the
|
||||
disk.
|
||||
|
||||
With no image to restore, `systemd-hibernate-resume` finishes at once and the
|
||||
band disappears. Some kernel console setups
|
||||
switch to a blank terminal for the load; the same lines then appear there.
|
||||
|
||||
## Shutdown
|
||||
|
||||
<img src="https://codeberg.org/wvhulle/bootvine/raw/branch/recordings/shutdown.gif" width="691" alt="A poweroff: units stopping under the targets waiting on them">
|
||||
@@ -174,6 +203,7 @@ shows the shutdown.
|
||||
| `nix run .#sample-vm-text-console` | the same units without a display manager, handing over at the default target | [without a display manager](#without-a-display-manager) |
|
||||
| `nix run .#sample-vm-scripted-initrd` | the same units with NixOS's scripted initrd, so bootvine starts in stage 2 only | [without a systemd initrd](#without-a-systemd-initrd) |
|
||||
| `nix run .#sample-vm-encrypted-volume` | a LUKS volume unlocked in the initrd; the passphrase is `sample passphrase` | [passphrase prompts](#passphrase-prompts) |
|
||||
| `nix run .#sample-vm-hibernate-resume` | a swap disk to hibernate to; run `systemctl hibernate`, then start the machine again | [resuming from hibernation](#resuming-from-hibernation) |
|
||||
| `nix run .#sample-vm-failures` | a firmware failure in the initrd, then units that exit with an error, crash, time out, cannot be run, restart too often, report a status, die after starting, and one that fails only through its dependency | [failures](#failures) |
|
||||
|
||||
Each machine keeps its disk in `~/.cache/bootvine-<machine>.qcow2`; delete it
|
||||
|
||||
Reference in new issue
Block a user