These patches can be tested from here:
https://launchpad.net/~cheako/+archive/ubuntu/boot-subvol
Code from here:
https://github.com/cheako/os-prober/tree/aggregate
Here are corrected versions of these patches. Filesystem raid is
almost certainly a separate bug.
From 3d4b580dea4592793af3411fc0543af36de0e958 Mon Sep 17 00:00:00 2001
From: Michael Mestnik
Date: Fri, 9 Jun 2017 13:04:05 -0500
Subject: [PATCH 2/6] Process filesystem raid entries
---
common.sh
Here is an updated copy of this patch.
From 6439f5a40bd4e610a462292c646098eeb4d5bcb1 Mon Sep 17 00:00:00 2001
From: Michael Mestnik
Date: Mon, 5 Jun 2017 11:52:17 -0500
Subject: [PATCH 1/6] Add in bcache devices
---
linux-boot-probes/common/50mounted-tests | 3 +++
os-prober
Using these patches reveals they are not complete, there are two areas
that need to be addressed.
Firstly the kernel can and will use any device in a filesystem for the
contents of /proc/mount. This causes problems in multiple areas that
os-probe will have to compensate for.
Second 50mounted-tes
In that case the proposed patch is wrong, dmraid is run every time the
file exists. Not only is the conditional in test wrong, but the file
is created when it should be being removed.
Hello,
This is the result of several days working the issues and analysing
each aspect. These patches include only what is necessary to support
subvol. Care has been taken to future proof by preparing for other
filesystems, currently this only works with btrfs. These patches have
been split, s
This does look better, I love the use of operators over if statements.
I don't think using a temp file is necessary here. I also wish the
regex wouldn't ever match a device containing the name of another
device, in that it should match the surrounding bits.
{ dmraid -r -c 2>/dev/null || true } |
This patch also adds bcache to list of partitions types to skip.
Note that linux-boot-probes/common/50mounted-tests is missing
LVM2_member from said list.
From 6439f5a40bd4e610a462292c646098eeb4d5bcb1 Mon Sep 17 00:00:00 2001
From: Michael Mestnik
Date: Mon, 5 Jun 2017 11:52:17 -0500
Subject: [PA
Package: os-prober
Version: 1.75
Severity: normal
Tags: patch
Looks like bcache devices are not detected.
- -- System Information:
Debian Release: 8.7
APT prefers stable
APT policy: (500, 'stable'), (490, 'testing'), (480, 'unstable'),
(470, 'experimental')
Architecture: i386 (x86_64)
Kernel
Package: os-prober
Version: 1.75
Severity: normal
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Here is the code as found in os-prober:17
: >"$OS_PROBER_TMP/dmraid-map"
DMRAID=$(type dmraid >/dev/null 2>&1 || true)
if [ "$DMRAID" ]; then
dmraid -r -c >"$OS_PROBER_TMP/dmraid-map"
fi
The p
Forget about all of this, I'm going to simplify the design. I have in
mind to make the following changes.
1. Don't just change $1 from being the path of a block device to being
a reserved word.
a. Instead append an additional parameter.
2. Use subvolume id, instead of names.
The patches have changes that have nothing to do with btrfs, I'll
attempt to isolate these.
The patches introduce a bug.
/usr/lib/os-probes/50mounted-tests
calls fs_type on "btrfs"
I'll look into this some more.
Refresh and simplify the patches.
https://github.com/cheako/os-prober/tree/btrfs-subvol
Also note os-probe fork:
https://github.com/campadrenalin/os-prober/commit/b571c0c7c3fe8986cc17dd200173fb7d86ef0c24
This commit reorgs the project making merges with the fork difficult.
From af1753acff410e69844
Package: os-prober
Version: 1.75
Severity: wishlist
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Please consider using some form of source buitification or tidy
to remove this whitespace.
- -- System Information:
Debian Release: 8.7
APT prefers stable
APT policy: (500, 'stable'), (490, 'tes
Package: debian-installer
Version: 20170525
Severity: wishlist
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
I created bcache backing store and was unable to install into it.
- -- System Information:
Debian Release: 8.7
APT prefers stable
APT policy: (500, 'stable'), (490, 'testing'), (480,
I got this problem befour and after using partmagic, the boot flopys
don't have parted. cfdisk was blank, and when I swiched vt's I
coulden't get back I could get back to the anna screen, bottom line
blue, rest gray. ls -l proc/?/fd reported /dev/console as stdin/out/err.
If there is any thin
I have a CD less networked pc and I get told that my eepro100 needs a
driver floppy. Are tahy talking about the windows ones? Good, as I
don't have them :)
I also wanted to know, find ought for my sellf, if my windows partition
would be parted and lilo/grub confed 'n' stuff.
--
To UNSUBSCRIB
18 matches
Mail list logo
