Is this also fixed in bionic yet?
On Tue, Apr 3, 2018 at 9:10 AM, Launchpad Bug Tracker
<1729...@bugs.launchpad.net> wrote:
> This bug was fixed in the package linux - 4.13.0-38.43
>
> ---
> linux (4.13.0-38.43) artful; urgency=medium
>
> * linux: 4.13.0-38.43 -proposed tracker (LP:
Not sure if this was supposed to also be fixed in bionic-proposed yet,
but I assumed so and
it's not working there; I'll test -proposed in artful and xenial next.
ubuntu@ubuntu:~$ cat /etc/lsb-release
DISTRIB_ID=Ubuntu
DISTRIB_RELEASE=18.04
DISTRIB_CODENAME=bionic
DISTRIB_DESCRIPTION="Ubuntu Bion
On Fri, Feb 16, 2018 at 12:14 PM, Joseph Salisbury <
joseph.salisb...@canonical.com> wrote:
> @raharper, I got side tracked and it got pushed down on my list. I can
> still send it upstream if you want. It's currently in Ubuntu as a SAUCE
> patch.
>
That would be great!
Thanks,
>
> --
> You
On Tue, Dec 12, 2017 at 5:52 AM, Dimitri John Ledkov wrote:
> Once the kernel is fixed, are there any changes that are required to
> systemd/udev?
>
No changes needed.
>
> ** Changed in: systemd (Ubuntu Bionic)
>Status: New => Incomplete
>
> --
> You received this bug notification beca
Confirmed bionic works as expected.
I suspect you can send that upstream with my SoB faster than I can.
Definitely interested in seeing if they'll take something like that.
On Thu, Dec 7, 2017 at 1:21 PM, Joseph Salisbury <
joseph.salisb...@canonical.com> wrote:
> Thanks for testing and the patc
Tested the xenial update. I had one boot where the links didn't get
created, but I cannot recreate that issue now.
On Thu, Dec 7, 2017 at 9:56 AM, Ryan Harper
wrote:
> Here's the Zesty test; all looks good.
>
> ubuntu@ubuntu:~$ cat /etc/cloud/build.info
> build_name: server
> serial: 20171207
>
Here's the Zesty test; all looks good.
ubuntu@ubuntu:~$ cat /etc/cloud/build.info
build_name: server
serial: 20171207
ubuntu@ubuntu:~$ uname -a
Linux ubuntu 4.10.0-40-generic #44~lp1729145 SMP Wed Dec 6 16:21:45 UTC
2017 x86_64 x86_64 x86_64 GNU/Linux
ubuntu@ubuntu:~$ tree /dev/bcache
/dev/bcache
Thanks for doing the cleanup; Patch looks good and I approve.
Signed-off-by: Ryan Harper
On Tue, Dec 5, 2017 at 4:59 PM, Joseph Salisbury <
joseph.salisb...@canonical.com> wrote:
> In the final patch we submit for SRU, it will also include your Signed-
> off-by, I just forgot to add it in the
Woot!
I think that did the trick.
ubuntu@ubuntu:~$ cat /etc/cloud/build.info
build_name: server
serial: 20171122
ubuntu@ubuntu:~$ lsb_release -rd
Description: Ubuntu 17.10
Release: 17.10
ubuntu@ubuntu:~$ set -o vi
ubuntu@ubuntu:~$ dpkg --list | grep linux-image
ii linux-image-4.13.0-17-generic
Thanks! I'll give it a try today.
On Tue, Dec 5, 2017 at 3:43 PM, Joseph Salisbury <
joseph.salisb...@canonical.com> wrote:
> I built an Artful test kernel with the path provided by Ryan. The test
> kernel can be downloaded from:
>
> http://kernel.ubuntu.com/~jsalisbury/lp1729145/
>
> Can those
Looks like those two kfree's in dev_run can be dropped since that was an
exit after kmalloc'ing env entries which are now only done in
bch_cached_dev_emit_change()
which is only called by dev_run after it knows that the device is not
yet running.
On Mon, Dec 4, 2017 at 11:35 AM, Ryan Harper
wr
Hi Joseph,
Sorry, I didn't give that a compile either; I just wanted to show what the
change could look like;
Let me see if I can get that to at least compile.
On Mon, Dec 4, 2017 at 11:18 AM, Joseph Salisbury <
joseph.salisb...@canonical.com> wrote:
> It looks like env[] was declared in bch_ca
Dec 01 05:07:25.679368 ubuntu systemd-udevd[474]: LINK
'disk/by-uuid/0a270acb-56b8-4498-8bad-b3bb149fe869'
/lib/udev/rules.d/60-persistent-storage.rules:79
Dec 01 05:07:25.683008 ubuntu systemd-udevd[474]: WARK:
dev_old_name=/dev/bcache/by-uuid/92d882d8-38cd-4537-847b-6f9c40ba67b4
dev_old_path=/dev
It looks like there is some ordering issues:
This is a grep through /run/udev/links ; these are checked by udev-dev
# find . -name 'b250*'
./\x2fdisk\x2fby-uuid\x2f0a270acb-56b8-4498-8bad-b3bb149fe869/b250:1
./\x2fdisk\x2fby-uuid\x2f92b0868d-7e56-4956-8e55-2c90ebee4a72/b250:0
./\x2fbcache\x2fby-u
Investigating why /dev doesn't have the correct links even if they're
present led me to look at
/usr/share/initramfs-tools/scripts/init-bottom/udev (provided via the udev
dpkg from systemd)
In there, the mount -n -o move is used to migrate the /dev from initramfs
to the rootfs/dev
This bug:
https
So, /dev/bcache/by-uuid is not getting created.
That's the same kernel bug I filed.
And, if they were, I think they'd get moved properly.
init-bottom/udev script does the following:
# Stop udevd, we'll miss a few events while we run init, but we catch up
udevadm control --exit
# move the /dev
16 matches
Mail list logo
