** Changed in: docker.io (Ubuntu)
Status: New => Invalid
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1966889
Title:
docker-build containers do not have logging driver configure
To manage
There are global defaults for logging which you can configure in
daemon.json (log-driver, log-level, log-opt), but none of these are used
when building images.
This was intentionally changed in
https://github.com/moby/moby/pull/29552, which has been included since
Docker 17.04.0 -- I imagine your
Unfortunately for your use case, this is an intentional feature of
`docker build` itself:
https://github.com/moby/moby/blob/0a3336fd7d19f7114fce2ff849a8989ed33e2059/builder/dockerfile/internals.go#L377
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is
In your second (working) example, I see you run "service docker stop"
before deleting the contents of "/var/lib/docker" but I do not see that
in your first example, which would definitely explain the behavior you
see ("service docker start" would be a no-op because Docker's already
running, and
After some discussion with mwhudson, I tried the following:
- add proposed
- install the libc6/libc-bin updates (which breaks a lot of stuff)
- restart the container
- stuff is working again
- install docker.io (from proposed)
- profit
--
You received this bug notification because you are a
I can reproduce, but I can even reproduce lots of failures by only
upgrading "libc6" and "libc-bin" (which come in with "docker.io"),
without Docker even installed or being installed. Lots of other
services then try to restart and fail to do so, and "apt update" even
starts failing to resolve
** Bug watch added: Debian Bug tracker #976959
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=976959
** Also affects: systemd (Debian) via
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=976959
Importance: Unknown
Status: Unknown
--
You received this bug notification
This is kind of an old issue now, but it's still outstanding on both
Bionic and Focal -- any chance it could get some love? It breaks our
ability to build certain images on ppc64le (rabbitmq as a notable
example).
--
You received this bug notification because you are a member of Ubuntu
Bugs,
Just to verify, I installed docker.io version 19.03.8-0ubuntu1.20.04.1,
started up dockerd, verified /var/lib/docker got some content in it, and
then stopped dockerd and ran "apt purge docker.io", and
"/var/lib/docker" was properly emptied. It was not removed completely
(in case it's a mount
in 1.11.2-0ubuntu6:
docker.io (1.11.2-0ubuntu6) yakkety; urgency=medium
...
- remove "/var/lib/docker" on purge
...
-- Tianon Gravi Fri, 19 Aug 2016 10:15:14 -0700
It sounds like maybe there's a bug where this isn't working correctly in
all cases?
--
You receive
Aha, this sounds like it's probably the same root cause that led to
https://github.com/docker/docker-ce-packaging/pull/508.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1870514
Title:
update
I think the core of this is fixed, but there's some situations still
where it triggers -- here's a new reproducer that can still trigger it
reliably with Docker (since I can't reproduce with the simplified steps
in #31 anymore):
```
$ docker pull ubuntu:20.04
20.04: Pulling from library/ubuntu
I'm running into this building a Docker image that's "FROM ubuntu:20.04"
and just trying to install libc6:i386. Here's the simplest reproducer
I've come up with:
```
$ docker pull ubuntu:20.04
20.04: Pulling from library/ubuntu
Digest:
That's entirely fair, but in that case isn't going snap-only for
"chromium-browser" a bit short-sighted, given the constraints on running
snaps?
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1869963
The "zfs" graph driver is part of Docker itself
(https://github.com/moby/moby/tree/89382f2f20745b9e63bed6c066f104980dff4396/daemon/graphdriver/zfs),
so https://github.com/moby/moby/issues would be the appropriate place to
file an issue (or enhancement request) against it.
--
You received this
Has this been reported upstream? The fix (in postinst) is large enough
that it seems like something upstream should at least be aware of, and
is probably something they should implement a fix for as well. Do you
think it's something that could/should be fixed in "dockerd" itself?
--
You
18.09.9 has been in stable for a while now, and 19.03.8 has been in beta
for a short time (hopefully stable soon). :)
** Changed in: docker (Ubuntu)
Status: Confirmed => Fix Released
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to
** Also affects: debconf (Ubuntu)
Importance: Undecided
Status: New
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1875882
Title:
debconf attempts interactive configuration; breaks Docker
The upstream-supported method of accomplishing this is via a systemd
drop-in which overrides ExecStart.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1858248
Title:
Thanks for testing and confirming! Fix is now released to the stable
channel.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1852720
Title:
docker snap suddenly installed by itself
To manage
My fix finished building and uploading into the edge channel, so if
folks could test it from either edge or beta (I'd suggest beta, but your
choice), that would be helpful.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
At least for the "Docker fails to run containers and fill syslog" half
of this, I've reproduced in https://forum.snapcraft.io/t/docker-
snap-18-09-9-error/14171/5 and have a fix committed waiting for
Launchpad to build so I can rush it through the channels ASAP.
--
You received this bug
This is definitely not the appropriate place for this to be filed
(because "src:podman" not existing is definitely not a *Docker* bug).
What would be more appropriate would be to take a look at
https://wiki.ubuntu.com/UbuntuDevelopment/NewPackages.
** Changed in: docker.io (Ubuntu)
>From the Docker perspective, it sounds like there's a process stuck
somewhere. I'd recommend checking the system for any leftover "docker"
processes (dockerd, containerd, shim processes, etc) via something like
"ps faux | grep docker" (and/or "containerd", "runc", "shim", etc).
The "dockerd"
Public bug reported:
After installing "gosu" on Bionic on ppc64el, attempts to run the
resulting "gosu" binary result in the following error message:
gosu: error while loading shared libraries: R_PPC64_ADDR16_HA reloc
at 0x0b41bcc7af58 for symbol `' out of range
I spoke briefly with
Anything useful in Docker's daemon logs?
As a workaround you should be able to set --dns on your container or
adjust /etc/docker/daemon.json to adjust that setting more permanently.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
Have you tried restarting the Docker daemon? (IIRC it caches the DNS
settings during startup)
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1820278
Title:
Incorrect DNS (resolv.conf) setup inside
Yep, that would be socket activation; I'd recommend explicitly enabling
"docker.service" so it starts on boot as expected.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1808266
Title:
Docker
This sounds to me like Docker restarted for some reason -- do you have
some kind of automated update process in place? If it's unattended, it
could potentially restart Docker, but the logs you've provided include a
host reboot too, which would definitely look like this externally.
--
You
Proposed patch attached -- I've tested that this:
1. does not prompt on initial install (but does start the daemon)
2. does prompt on every reinstall / upgrade (if the daemon is running)
3. does prompt during "dpkg-reconfigure docker.io" (but does _not_ [re]start
the daemon)
4. works properly
I can confirm this behavior -- there's a bug in the postinst script
where it makes sure we don't restart the daemon during "dpkg-
reconfigure" (since that should only happen during install/upgrade), but
instead it also stops us from prompting during "dpkg-reconfigure", which
isn't right.
I think
++ Absolutely agree that having "apt-get install" auto-update stale
indexes would be ideal (or even a "--update" flag to install for
combining them); looks like that discussion is over in bug 1709603. :)
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is
Looking into this deeper -- applying this patch for bionic will have
net-zero effect, given this comment:
https://github.com/moby/moby/pull/36417#issuecomment-369266565
For this patch to do anything, "libseccomp" needs to be at least version
2.3.3, and bionic is only at 2.3.1 (so the added line
> The reason this is being done as a debconf question is because neither
option is good. And *since* we are forced to ask the user to pick their
poison, I consider it insufficient to ask this question only once. The
question should by default be re-asked, at each package upgrade, so that
the user
I'm guessing you added the group as a non-system group?
(resulting in "addgroup: The group `docker' already exists and is not a system
group. Exiting.")
I think it would probably be sane to add a check similar to the one
Docker itself uses (namely whether "/etc/group" includes a line that
starts
Indeed, looks like this fix isn't in a released version at all yet
(likely to be in 18.04).
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1755250
Title:
backport statx syscall whitelist fix
To
*** This bug is a duplicate of bug 1744599 ***
https://bugs.launchpad.net/bugs/1744599
** This bug has been marked a duplicate of bug 1744599
package docker.io (not installed) failed to install/upgrade: subprocess
installed post-removal script returned error exit status 127
--
You
This was fixed upstream in 17.04+
(https://github.com/docker/docker/pull/30519) -- I could've sworn we
backported the patch, but apparently not! (I'll go work on getting that
applied now)
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to
This is still causing quite a _lot_ of downstream pain -- any chance it
could be re-prioritized?
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1711735
Title:
source.list for armhf includes
He means simply removing "trusty + armhf" for the "ros" and/or
"gazebo" images, I believe (not for the "ubuntu" image).
The harm this causes is that "apt-get update" fails, which causes the
whole "docker build" to fail.
Here's an example from ppc64el:
W: Failed to fetch
The "hello-world" image is a 64bit image; you'll need something 32bit to
run on that system (like i386/hello-world --
https://hub.docker.com/r/i386/hello-world/).
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
This was filed against the wrong Docker package (talks about
containers); didn't verify whether it's valid but wanted to get it
pointed at the right package anyhow.
** Package changed: docker (Ubuntu) => docker.io (Ubuntu)
--
You received this bug notification because you are a member of Ubuntu
** Package changed: docker (Ubuntu) => docker.io (Ubuntu)
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1657646
Title:
VG/LV are not available in /dev/mapper/ after reboot
To manage notifications
I think it's probably more likely that this is related to the changes to
the dtb address over in #1636838 (although after hacking at it for a
while trying to make those changes ad-hoc for verification, I wasn't
able to confirm that hunch for sure).
--
You received this bug notification because
: New
** Changed in: containerd (Ubuntu)
Status: New => In Progress
** Changed in: containerd (Ubuntu)
Assignee: (unassigned) => Tianon Gravi (tianon)
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
** Also affects: runc (Ubuntu)
Importance: Undecided
Status: New
** Changed in: runc (Ubuntu)
Assignee: (unassigned) => Tianon Gravi (tianon)
** Changed in: runc (Ubuntu)
Status: New => In Progress
--
You received this bug notification because you are a member of
I got the breaking-update automatically via unattended-upgrades -- no
human intervention of any kind. Simply rebooted the RPi3 after the
holidays and it failed to come back up.
To fix it, I used "qemu-user-static"'s ARM emulation to get a chroot
into the SD card's system on my laptop, and then I
Interesting. I've seen some success in situations like this from simply
restarting the Docker daemon -- mind giving that a shot?
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1644230
Title:
Can't
Is "JournalErrors.txt" the output of "journalctl" for the service? I
don't see anything useful in there regarding Docker startup output. :/
Has this machine had Docker installed previously, or is this a clean
install?
--
You received this bug notification because you are a member of Ubuntu
This is definitely a real issue, and it wasn't intentional -- I've got a
fix in progress, but it needs to happen in both the runc and the
docker.io packages.
** Changed in: docker.io (Ubuntu)
Status: New => Confirmed
** Also affects: runc (Ubuntu)
Importance: Undecided
Status:
Yeah, I think in Ubuntu changing that to "Recommends" is sensible.
- Tianon
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1631230
Title:
auplink spam in docker unit log
To manage notifications
It should already be in either recommends or suggests (can't remember
which).
IMO "depends" is inappropriate because it's only required if your kernel
both supports AUFS and it's configured or chosen as your graph driver.
BTRFS, overlayfs, and even ZFS are also supported, for example.
--
You
>From what I've seen in the past, this is almost always indicative of
daemon startup errors (either "/var/log/docker.log" or "journalctl -u
docker.service" depending on init system), which are _usually_ related
to graph drivers (such as the AUFS module failing to load for some
reason, or a mix of
The important bits there are these:
Sep 30 16:42:55 einstein docker[5009]:
time="2016-09-30T16:42:55.781784363+02:00" level=error msg="[graphdriver] prior
storage driver \"aufs\" failed: driver not supported"
Sep 30 16:42:55 einstein docker[5009]:
time="2016-09-30T16:42:55.781825992+02:00"
https://github.com/tianon/docker-brew-ubuntu-core/pull/64 is relevant
here. :)
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1609087
Title:
Ubuntu docker images should create /run/container_type &
This is fixed now that we've got "TasksMax=infinity" in our systemd
service file, right?
(The commit referenced above is also included in all the 1.11 releases,
so either way we should be fixed here.)
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is
Yeah, I agree and want to get there, but I figured just dropping it
straight away would be doing our users a disservice (especially since it
might not be obvious right away that it's not being used anymore).
For example, if a user has "--bip=..." set in DOCKER_OPTS in that file,
then dropping the
https://anonscm.debian.org/cgit/docker/docker.io.git/commit/?id=0bc5ad426c84804c939804e33247140ff2636991
is the compromise I've applied in Debian recently (replacing the file's
contents with a long-form note pointing our users to the upstream-
recommended alternatives, which should invoke dpkg's
This appears to be because "hello-world" is an amd64 image and you
appear to be on an s390x system -- can you try "s390x/hello-world"
instead? :)
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1603244
IMO it's also worth noting that 1.12.0 isn't GA yet -- upstream is still
doing release candidates (currently at rc4), and the _current_ planned
date for 1.12.0 is July 28th (although it used to be the 14th, so
there's nothing to say it won't change again, especially if more issues
manifest during
I think https://github.com/raspberrypi/firmware/issues/550 has some
relevant information, especially
https://github.com/raspberrypi/firmware/issues/550#issuecomment-190780010
(RPi Foundation is not working on arm64) and
https://github.com/raspberrypi/firmware/issues/550#issuecomment-197452521
The debdiff from xenial to trusty (attached) doesn't appear terribly
interesting. :(
** Patch added: "xenial-to-trusty.debdiff"
https://bugs.launchpad.net/ubuntu/+source/chromium-browser/+bug/1563184/+attachment/4665562/+files/xenial-to-trusty.debdiff
--
You received this bug notification
Hit the same issue on my own new Raspberry Pi 3 running Ubuntu Xenial
(installed via http://www.finnie.org/software/raspberrypi/ubuntu-
rpi3/ubuntu-16.04-preinstalled-server-armhf+raspi3.img.xz linked from
https://wiki.ubuntu.com/ARM/RaspberryPi).
I managed to get a little bit further by
I ran it locally with a 4GB VM and managed to get through all the tests
without sshd dying. :(
(obviously we've got some failures that we'll need to figure out, but
probably only worth figuring out the failures if we can figure out why
sshd is getting killed D: )
OOPS: 192 passed, 120
This sounds suspiciously like OOM killing sshd, not Docker's tests
directly -- I know the integration tests used to require anywhere from
1 to 4 GB of RAM, and I imagine that's probably still the case. :(
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is
Applying https://github.com/docker/docker/pull/22000 is another option
worth considering. The patch isn't ideal (and upstream currently has
no interest in applying it), but it does do the job.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to
That output is still cut off, but does show the problem:
time="2016-04-21T12:26:54.837949976+01:00" level=fatal msg="Error
starting daemon: error initializing graphdriver: \"/var/lib/docker\"
contains other graphdrivers: d
We're missing the end of this log line which would tell us exactly
which
That output looks like the actual error message is getting cut off --
"journalctl -u docker.service | tail" might give us a slightly more
useful indication (at the very least, hopefully the full error message
from the daemon).
--
You received this bug notification because you are a member of
IMO we should just remove the patch we add that makes it apply; systemd
drop in files are the way systemd upstream recommends configuring startup
parameters like this, which is why upstream doesn't include the
EnvironmentFile directive that we've patched in.
--
You received this bug notification
I think this is an upstream bug because last I checked Docker shells out to
iptables directly.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1503674
Title:
depends on iptables instead of nftables
I'm actually really concerned by this change (speaking as both a Debian
maintainer on the package, and as the maintainer of the build scripts
upstream). We're actively working upstream on getting proper first-
class gccgo compilation support into Docker's build system
I'm actually really concerned by this change (speaking as both a Debian
maintainer on the package, and as the maintainer of the build scripts
upstream). We're actively working upstream on getting proper first-
class gccgo compilation support into Docker's build system
Even the current patch headers are misleading:
Description: Add gccgo build support for Power
This patch is based on one that has been accepted upstream
in trunk with some minor additions to deal with the way
that the dh-golang helper works in Debian/Ubuntu.
Origin:
Even the current patch headers are misleading:
Description: Add gccgo build support for Power
This patch is based on one that has been accepted upstream
in trunk with some minor additions to deal with the way
that the dh-golang helper works in Debian/Ubuntu.
Origin:
The current patch seems like it comes more from
https://github.com/docker/docker/pull/10569 than from the referenced
https://github.com/docker/docker/pull/10665, and basically this same
diff has been rejected upstream multiple times.
--
You received this bug notification because you are a member
The current patch seems like it comes more from
https://github.com/docker/docker/pull/10569 than from the referenced
https://github.com/docker/docker/pull/10665, and basically this same
diff has been rejected upstream multiple times.
--
You received this bug notification because you are a member
https://bugs.debian.org/775775 :D
(more reproducing, and a patch)
** Bug watch added: Debian Bug tracker #775775
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=775775
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
Has this also been reported upstream in Debian? I can't find anything
related in the BTS.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1396760
Title:
ca-certificates-java doesn't create
Paul can't take all the credit for that being missing. :)
I left it out because we leave out dh_auto_build in our
override_dh_auto_build, so I figured whatever future behavior
dh_auto_clean might include we probably want to choose to include
here manually, since the Docker build system likely
As was noted on the upstream issue, this is working as designed. You
cannot run a 64bit image on a 32bit host. Also, upstream does not
officially support 32bit hosts, so you will need to use something like
debootstrap to create your own.
--
You received this bug notification because you are a
I agree completely, but upstream isn't planning to change that in the
near future (especially since this particular configuration is not
only unsupported, but the Debian/Ubuntu package includes a 100%
unapproved-by-upstream patch to make it even possible).
_Maybe_ today's Microsoft news will help
Yes
♥,
- Tianon
On 16 Sep 2014 10:31, Paul Tagliamonte paul...@ubuntu.com wrote:
Nice! :D
On Tue, Sep 16, 2014 at 12:13 PM, James Page james.p...@ubuntu.com
wrote:
Thanks Iain
Sync'ing away!
--
You received this bug notification because you are a member of Docker
Ubuntu
This would probably be good to report upstream. It might also be wise
(and they'll request it upstream) to test it against 1.2.0 (or at least
1.0.1, which was pushed into utopic and is in trusty-updates too).
Paul and I have also been playing with nightly builds of the docker.io
package at
You're on a 32bit system (i386) and trying to run a 64bit image (amd64).
Systems that aren't 64bit are officially unsupported upstream, so to use
one, you'll have to make your own images (including the base images), or
find some that someone else has made with the appropriate architecture.
--
Just to note, this was also reported upstream at
https://github.com/docker/docker/issues/7462
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1353923
Title:
Installing Docker Breaks KVM
To manage
Paul: we've got the dh_gencontrol stuff in our Git repo that adds
apparmor to Recommends for Ubuntu - couldn't we at least start by
backporting that one fix while the deps for a newer Docker version get
sorted out (Bug #1338768)?
(For reference, it's commit
86 matches
Mail list logo
