dreamcat4/docker-images

Dreamcat4's Docker Images (Trusted Builds)

Dreamcat4's Docker Images (Trusted Builds)

docker-images's Issues

crashing smbd with docker 1.11.1

Hi there,

the end of samba docker log shows:

nmbd version 4.3.8-Ubuntu started.
Copyright Andrew Tridgell and the Samba Team 1992-2015
STATUS=daemon 'nmbd' finished starting up and ready to serve connections
EOF on stdin
Got SIGTERM: going down...
/usr/bin/samba.sh exited 0

Is maybe this fix related?
dperson/samba@bec2314

Pipework: pipework commands not run when container is restarted

I have a container with this environment:

pipework_cmd: eth0 -i eth1 @CONTAINER_NAME@ a.x.y.z/32

This works fine when the container is created but if the container is stopped and started again the ip is no longer associated with it.

Docker Pipework API version mismatch

The new docker build seems to have done something to the docker API. I get the following message.

Error response from daemon: client is newer than server (client API version: 1.24, server API version: 1.22)

I am doing some further investigation to see what is causing this.

Additionally at the moment I am unable the get inside the container using my usual exec bash method. Trying to work out if that's something to do with my install.

dnsmasq

there have a error! make me docker image run fail
dnsmasq: failed to create inotify: Too many open files!
dnsmasq: failed to create inotify: Too many open files!
dnsmasq: failed to create inotify: Too many open files!
dnsmasq: failed to create inotify: Too many open files!

pipework_cmd should be sorted

If I have multiple pipework commands, and there is any order required, I have no control over it.

Recommend changing https://github.com/dreamcat4/docker-images/blob/master/pipework/entrypoint.sh#L348 to

    _pipework_cmds="$(env | grep -o -e '[^=]*pipework_cmd[^=]*' | sort -n )"

This allows me to do things like

    00_pipework_cmd_public=eth0 @CONTAINER_ID@ dhclient
    01_pipework_cmd_public=eth0 -i eth2 @CONTAINER_ID@ 192.168.10.10/[email protected]

What FROM?

In your README you advise to use ubuntu-debootstrap:14.04 as base, here in the example you're using another image. What is the correct one?

Initialise tv EPG grabber

Now, this is my first time using Docker, so I am probably doing something wrong but here is goes (be kind). I've run up the docker and tvheadend is doing fine, took me a second attempt because I needed to add the --net=host parameter for my sat>ip. But I am getting these errors on the EPG grabber:

2016-09-14 08:05:23.753 [ INFO] /usr/bin/tv_grab_uk_tvguide: grab /usr/bin/tv_grab_uk_tvguide
2016-09-14 08:05:23.754 [ NOTICE] START: HTS Tvheadend version 4.0.9 started, running as PID:179 UID:9981 GID:9981, CWD:/ CNF:/config
2016-09-14 08:05:23.756 [ INFO] spawn: Executing "/usr/bin/tv_grab_uk_tvguide"
2016-09-14 08:05:24.626 [ ERROR] spawn: You need to configure the grabber by running it with --configure
2016-09-14 08:05:24.638 [ ERROR] /usr/bin/tv_grab_uk_tvguide: no output detected
2016-09-14 08:05:24.638 [WARNING] /usr/bin/tv_grab_uk_tvguide: grab returned no data

From root I ran

docker exec /usr/bin/tv_grab_uk_tvguide --configure
and it asked me where to put the cache files, offering to put them in /root/.xmltv/cache, I accepted this default and it just sat there. I exited with ^C, and ran just /usr/bin/tv_grab_uk_tvguide, it errored with the warning about --configure, so I ran it again and entered /config/.xmltv/cache hoping that might help, again it just sat there after I ran the command and did nothing.

In practicality I think it is strange when programmes like tvh emit errors (or at least pass them) and don't actually handle them given that the returned error message tells you how to resolve the problem. This may not be a Docker issue but I don't know how to resolve it, please help.

Not working on Centos 6.5 and Docker 1.7.1

docker --version
Docker version 1.7.1, build 786b29d/1.7.1

docker run -v /var/run/docker.sock:/docker.sock --privileged=true --pid=host --net=host dreamcat4/pipework --help
mkdir: cannot create directory '/sys/fs/cgroup': No such file or directory
mount: mount point /sys/fs/cgroup does not exist
Could not make a tmpfs mount. Did you use -privileged?

uname -a
Linux lvu.mtl.intelerad.com 2.6.32-504.23.4.el6.x86_64 #1 SMP Tue Jun 9 20:57:37 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux

Not passing arguments to dhcp client program

Hey it is me again! Hopefully for the last time!

Pipework allows us to pass optional arguments to dhcp clients like this:

./pipework --direct-phys eno1v12 3b10e50e38a7 dhclient:"-H hostnametest"
In this case "-H hostname" to tell dhclient to set a custom hostname.

It is imperative we use doublequotes after dhclient: because pipework fails to pass the arguments otherwise.

Now trying to do the same with dreamcat4/pipework:latest brings us back to the doublequote issue discussed in #46 .

I am pasting logs and docker-compose files:

https://pastebin.com/EMUXS9L8

myapp:
  image: praqma/network-multitool
  environment:
      - pipework_cmd=--direct-phys eno1v12 @CONTAINER_NAME@ dhclient:"-H hostnametest"
  net: none

Suffice it to say that I also tried it without the doublequotes but it did not work, as it did not work with pipework ran straight on the host either without those doublequotes.

/entrypoint.sh: line 190: set: direct-phys: invalid option name

The documentation makes the assumption that we should be able to use all pipework options with this docker image.

Is the --direct-phys option not implemented on this?

how can I add two network card ?

Here is the command I use in docker-compose

    - pipework_cmd_that_creates_private_bridge= br1 -i eth0 @CONTAINER_NAME@ 192.168.0.1
    - pipework_cmd_that_creates_private_bridge= br2 -i eth1 @CONTAINER_NAME@ 192.168.1.1

It seems only the latest command takes affect ? how can I add two bridge ?

Looks like a great tool, however I can't get it to work on Ubuntu 15.04 with Docker 1.8.1

My docker-compose.yml currently contains:
I'm just following your example and adding an IP to the first node in my cluster.

couchbase1:
  container_name: couchbase1
  image: couchbase/server:community-3.0.1 
  ports:
    - "8091:8091"
  volumes:
    - /opt/couchbase/node1:/opt/couchbase/var
  environment:
    - pipework_cmd=eth0 @CONTAINER_NAME@ 192.168.1.101/[email protected] 0a:00:01:01:01:01

couchbase2:
  container_name: couchbase2
  image: couchbase/server:community-3.0.1
  volumes:
    - /opt/couchbase/node2:/opt/couchbase/var

couchbase3:
  container_name: couchbase3
  image: couchbase/server:community-3.0.1 
  volumes:
    - /opt/couchbase/node3:/opt/couchbase/var

pipework:
  image: dreamcat4/pipework
  restart: always
  volumes:
    - /var/run/docker.sock:/docker.sock
  privileged: true
  pid: host 
  net: host
  environment:
    - run_mode=batch,daemon
    - host_routes=true

However, as you can see it doesn't work.

Attaching to couchbase1, couchbase3, couchbase2, couchbase_pipework_1
pipework_1   | WARNING: No swap limit support
pipework_1   | ln: failed to create symbolic link '/sys/fs/cgroup/systemd/name=systemd': Operation not permitted
pipework_1   | event_line=2015-09-08T23:49:06.000000000Z d09a9ca3f2bb785763504df36fca52ba45cb2e1a9e9f12317e5e74b43d4ac072: (from dreamcat4/pipework) start
pipework_1   | WARNING: No swap limit support
pipework_1   | ln: failed to create symbolic link '/sys/fs/cgroup/systemd/name=systemd': Operation not permitted

pipework_1   | /entrypoint.sh: line 147: printf: 099: invalid octal number

Setting dnsmasq as DHCP-proxy

Hi,

i can't for the life of me get dnsmasq to act as a proxy for my current DHCP-server.
When i change the dnsmasq.conf file to the correct IP - it looks like there is a script that changes it back.

Any suggestion as to how i solve this issue?

Investigate using --cap-add=NET_ADMIN

Wondering if NET_ADMIN is sufficient to replace --privileged when running the pipework container.

Thanks!

On death cleanup would be nice

I was rather hoping that this container would clean up after itself, and it looks like it's supposed to be but isn't succeeding. Here's the relevant section of logs, I've wrapped it in code block to stop markdown eating it.

++ docker events --since=1451839044 --filter=event=start --filter=event=die
++ echo event_line=2016-01-03T16:38:42.000000000Z fb439faea533ad66bc94284cba8f147b81a8c4839147607ab047d6260336986e: '(from' 'kaithar/sqls:latest)' die
event_line=2016-01-03T16:38:42.000000000Z fb439faea533ad66bc94284cba8f147b81a8c4839147607ab047d6260336986e: (from kaithar/sqls:latest) die
+++ echo -e ' 2016-01-03T16:38:42.000000000Z fb439faea533ad66bc94284cba8f147b81a8c4839147607ab047d6260336986e: (from kaithar/sqls:latest) die'
+++ grep -v 'from dreamcat4/pipework'
+++ tr -s ' '
++ event_line_sanitized=' 2016-01-03T16:38:42.000000000Z fb439faea533ad66bc94284cba8f147b81a8c4839147607ab047d6260336986e: (from kaithar/sqls:latest) die'
+++ echo -e ' 2016-01-03T16:38:42.000000000Z fb439faea533ad66bc94284cba8f147b81a8c4839147607ab047d6260336986e: (from kaithar/sqls:latest) die'
+++ cut -d ' ' -f3
++ container_id=fb439faea533ad66bc94284cba8f147b81a8c4839147607ab047d6260336986e:
+++ echo -e ' 2016-01-03T16:38:42.000000000Z fb439faea533ad66bc94284cba8f147b81a8c4839147607ab047d6260336986e: (from kaithar/sqls:latest) die'
+++ cut -d ' ' -f6
++ event=die
++ '[' fb439faea533ad66bc94284cba8f147b81a8c4839147607ab047d6260336986e: ']'
++ _process_container fb439faea533ad66bc94284cba8f147b81a8c4839147607ab047d6260336986e die
+++ echo fb439faea533ad66bc94284cba8f147b81a8c4839147607ab047d6260336986e
+++ cut -c1-12
++ c12id=fb439faea533
++ event=die
+++ env
+++ grep -e '.*pipework.*'
+++ cut -d= -f1
++ unset pipework_cmd
+++ docker inspect --format '{{range $index, $val := .Config.Env }}{{printf "%s\"\n" $val}}{{end}}' fb439faea533
+++ grep -e 'pipework_cmd.*=\|^pipework_key=\|pipework_host_route.*='
+++ sed -e 's/^/export "/g'
++ _pipework_vars='export "pipework_cmd=br.external -i eth1 @CONTAINER_NAME@ <container ip>/27@<container" gw>'
++ '[' 'export "pipework_cmd=br.external -i eth1 @CONTAINER_NAME@ <container ip>/27@<container"' '] gw>'
+++ echo -e 'export "pipework_cmd=br.external -i eth1 @CONTAINER_NAME@ <container ip>/27@<container" gw>'
+++ grep -o -e '@CONTAINER_NAME@\|@CONTAINER_ID@\|@HOSTNAME@\|@INSTANCE@'
+++ uniq
+++ sort
++ _macros=@CONTAINER_NAME@
++ '[' @CONTAINER_NAME@ ']'
++ _expand_macros
++ for _macro in '$_macros'
++ case $_macro in
+++ docker inspect -f '{{.Name}}' fb439faea533
++ name=/sqls
+++ echo 'export "pipework_cmd=br.external -i eth1 @CONTAINER_NAME@ <container ip>/27@<container" gw>'
+++ sed -e 's|@CONTAINER_NAME@|sqls|g'
++ _pipework_vars='export "pipework_cmd=br.external -i eth1 sqls <container ip>/27@<container" gw>'
++ eval export '"pipework_cmd=br.external' -i eth1 sqls '<container ip>/27@<container" gw>'
+++ export 'pipework_cmd=br.external -i eth1 sqls <container ip>/27@<container gw>'
+++ pipework_cmd='br.external -i eth1 sqls <container ip>/27@<container gw>'
++ '[' '' ']'
+++ env
+++ grep -o -e '[^=]*pipework_cmd[^=]*'
+++ sort
++ _pipework_cmds=pipework_cmd
++ '[' pipework_cmd ']'
++ '[' die = die ']'
++ cleanup_wait=0
++ '[' '' ']'
++ '[' '' ']'
++ sleep 0
++ return 0
++ true

(br.external is an ovs bridge, just in case that's relevant)

Any ideas? Need further info?

dhclient does not send hostname of container

Hello,

the dhclient provided in the container does not send the hostname of the container used for dhcp setup. The hostname of the docker host system is send during the dhcp request instead.

The file /etc/dhcp/dhclient.conf specifies: send host-name = gethostname();.
This means the hostname is resolved to the hostname of the docker host system if the pipework container uses net: host.

Regards
Joke

Files on bintray

Hi there,
i was looking into your Dockerfile, I'll find your serving files from the bintray Services. Does this announcement affect your PXE Server?

report version compatibility issue on docker engine 1.7.1

new here.

It works fine in latest docker engine 1.10.x and docker compose 1.6.x (Windows DockerToolbox).

While when I use it "prod" env where docker engine 1.7.1 is used (I can't upgrade so far), no matter which version of docker-compose is used, it always reports

pipework_1 | Error response from daemon: client is newer than server (client API version: 1.21, server API version: 1.19)

Here is env

$ docker version
Client version: 1.7.1
Client API version: 1.19
Go version (client): go1.4.2
Git commit (client): 786b29d
OS/Arch (client): linux/amd64
Server version: 1.7.1
Server API version: 1.19
Go version (server): go1.4.2
Git commit (server): 786b29d
OS/Arch (server): linux/amd64

I tried for docker-compose 1.5.2/1.5.1/1.4.2, also tried with env COMPOSE_API_VERSION, it always reports this fault.

Also noticed it is not related with docker-compose

$ docker run -v /var/run/docker.sock:/docker.sock --privileged=true --pid=host --net=host   dreamcat4/pipework --help
Error response from daemon: client is newer than server (client API version: 1.21, server API version: 1.19)
error: can't connect to unix:///docker.sock

Do you have any suggestion ? (old docker image ?)

UI not accessible

I've setup and configured this docker image, but am having problems accessing the web ui. When looking in STDOUT I find the following error repeating. I have verified the credentials, but its still not authenticating.

Ignores host_routes=false

If you set host_routes to false, it still sets up a host_route.

The problem is here https://github.com/dreamcat4/docker-images/blob/master/pipework/entrypoint.sh#L367

it creates the host route if $pipework_host_routes is non blank, or $pipework_host_route is not blank. But $pipework_host_route is set by evaluating $pipework_cmd_varname which is always non-blank, and so it always sets up the routes.

error: can't connect to unix:///docker.sock

I'm trying to run the latest dreamcat4/pipework with docker v1.8.2/CentOS7,
and facing the error, "error: can't connect to unix:///docker.sock"
Need to set some options for docker daemon?

# docker -v
Docker version 1.8.2-el7.centos, build a01dc02/1.8.2

# cat /etc/redhat-release
CentOS Linux release 7.2.1511 (Core)

# docker images | grep dreamcat4
docker.io/dreamcat4/pipework     latest              8a25bbdbf03c        10 days ago         251.9 MB
docker.io/dreamcat4/pipework     1.0.6               58d0fbb45a32        10 months ago       186.5 MB

# docker run \
   --restart=always \
   --name=pipework \
   -v /var/run/docker.sock:/docker.sock \
   --privileged=true \
  --pid=host \
   --net=host \
   -e run_mode=batch,daemon \
   -e host_route=true \
   dreamcat4/pipework --help
Error response from daemon: client is newer than server (client API version: 1.21, server API version: 1.20)
error: can't connect to unix:///docker.sock

# ls -l /var/run/docker.sock
srw-rw---- 1 root root 0  2月 15 19:11 /var/run/docker.sock

v1.0.6 works well.

# docker run \
   --restart=always \
   --name=pipework \
   -v /var/run/docker.sock:/docker.sock \
   --privileged=true \
   --pid=host \
   --net=host \
   -e run_mode=batch,daemon \
   -e host_route=true \
   -d dreamcat4/pipework:1.0.6

# docker run \
   --name=sample \
   --net=none \
   -e 'pipework_cmd_eth0=br0 -i eth0 @CONTAINER_NAME@ 172.26.210.212/24' \
   -e 'pipework_cmd_eth1=br1 -i eth1 @CONTAINER_NAME@ 192.168.100.112/24' \
   -itd busybox

# docker exec sample ip address show | grep "inet "
    inet 127.0.0.1/8 scope host lo
    inet 172.26.210.212/24 scope global eth0
    inet 192.168.100.112/24 scope global eth1

PXE - ubuntu casper-rw does not try to connect to samba

Hi,

I need some help with PXE, which I installed on a second computer in my home.

I configured only the ubuntu part, including the casper-rw and home-rw files.
During network boot, the menu appears, after selecting either the ubuntu casper-rw menu or the installer iso menu, it starts loading initrd and vmlinuz, then starts booting showing a lot of logs as usual during linux boot.
Then, the ubuntu graphical screen appears with the logo and those 5 horizontal dots changing colors between white and red and, after a minute or two, it goes back to text mode and shows the message: (initramfs) Unable to find a live file system on the network.
I used wireshark to see what happens on the network and, after downloading initrd and vmlinuz, no other packet is sent over the network. Complete silence.
It's like it does not even try to connect to samba to download the rest of the files. I don't know what to do now.

Note: in ubuntu-16.04.6-desktop-amd64.iso, there is no vmlinuz.efi or initrd.lz file, but vmlinuz and initrd. I modified boot.ipxe accordingly, besides the samba ip/share/user/pass.

I attached a screenshot from a VM:
The same happens if I try to network boot my computer itself. How to debug this?
Please advise,
Thank you

PS: I tried the chain http://boot.salstar.sk menu, it doesn't work, not even displaying their menu (screenshot attached).
I also added a new menu with chain http://boot.salstar.sk/deb.ipxe and it shows their menu, but whatever selections I make, in the end it shows a couple of errors (screenshot attached). I am a little confused about this matter.

pipework_cmd that do not start in pipework get rerun multiple times

The issue is here

https://github.com/dreamcat4/docker-images/blob/master/pipework/entrypoint.sh#L336

    unset $(env | grep -e "^pipework.*" | cut -d= -f1)

Notice how the regex is "^pipework.*" (starts with ^), so if I use c05_pipework_cmd it does not get filtered out, and keeps getting rerun with every event.

The correct line should be:

    unset $(env | grep -e ".*pipework_cmd.*" | cut -d= -f1)

CentOS PXE

Anyway to use this to PXE boot CentOS 7.3?

PXE server wont build

It Seems that dl.bintray.com no longer exists --

Is there somewhere else i could grab these files from possibly???

=> ERROR [3/8] RUN wget -O /tmp/dnsmasq.tar.gz https://dl.bintray.com/dreamcat4/linux/dnsmasq/dnsmasq-latest_lin 1.0s

[3/8] RUN wget -O /tmp/dnsmasq.tar.gz https://dl.bintray.com/dreamcat4/linux/dnsmasq/dnsmasq-latest_linux-x86_64.tar.gz && tar zxf /tmp/dnsmasq.tar.gz -C / --skip-old-files && rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/*:
#6 0.771 --2021-10-01 11:50:10-- https://dl.bintray.com/dreamcat4/linux/dnsmasq/dnsmasq-latest_linux-x86_64.tar.gz
#6 0.773 Resolving dl.bintray.com (dl.bintray.com)... 52.43.246.207, 52.34.245.248
#6 0.844 Connecting to dl.bintray.com (dl.bintray.com)|52.43.246.207|:443... connected.
#6 0.882 HTTP request sent, awaiting response... 403 Forbidden
#6 0.893 2021-10-01 11:50:10 ERROR 403: Forbidden.
#6 0.893

---

executor failed running [/bin/sh -c wget -O /tmp/dnsmasq.tar.gz https://dl.bintray.com/dreamcat4/linux/dnsmasq/dnsmasq-latest_linux-x86_64.tar.gz && tar zxf /tmp/dnsmasq.tar.gz -C / --skip-old-files && $_clean]: exit code: 8
bdrew@hpee:~/source/docker-images/pxe$

Pipework: 'Device "eth1" does not exist.'

Hello again!

I ran the pipework container using

docker run --restart=always -v /var/run/docker.sock:/docker.sock --privileged=true --pid=host --net=host \ 
>   -e run_mode=batch,daemon -e host_routes=true dreamcat4/pipework

I have been trying to attach both physical and macvlan interfaces to a container, both with docker run and docker-compose but I am having no luck and I always get Device "ethX" does not exist. on the pipework container output. I am also getting ln: failed to create symbolic link '/sys/fs/cgroup/systemd/name=systemd': Operation not permitted but I believe that this is irrelevant.

Here is my docker-compose:

  image: praqma/network-multitool
  environment:
    - pipework_cmd=eno1  test_myapp_1 dhcp

Here is the full portainer container output with the docker-compose above:

event_line=2019-12-06T18:14:26.958833946Z container start 3457c051cc316ce52498b79e66ad8dcd8d3dc931ba8aa1f94c9bd39fafdab3c7 (com.docker.compose.config-hash=da20c53a2310685289f63617bba678a836e080f6d91c549dd2dd96c3d8fc2018, com.docker.compose.container-number=1, com.docker.compose.oneoff=False, com.docker.compose.project=test, com.docker.compose.project.config_files=docker-compose.yml, com.docker.compose.project.working_dir=/root/docker_misc/test, com.docker.compose.service=myapp, com.docker.compose.version=1.25.0, image=praqma/network-multitool, name=test_myapp_1)
Device "eth1" does not exist.

Here is another try with docker run and physical (SR-IOV) passthrough:
docker run --net=none -e 'pipework_cmd= direct-phys eno1v2 test10 10.0.10.240/24' praqma/network-multitool

I also tried running e.g. pipework eno1 test10 10.0.10.240/24 inside the pipework container through a protainer console and I always get Cannot connect to the Docker daemon. Is the docker daemon running on this host? as a result.

Any ideas what I am doing wrong?

Edit: Also trying to run the pipework container with -e run_mode=manual or -e run_mode=all I get this output:

# docker run --restart=always -v /var/run/docker.sock:/docker.sock --privileged=true --pid=host --net=host   -e run_mode=manual -e host_routes=true dreamcat4/pipework
ln: failed to create symbolic link '/sys/fs/cgroup/systemd/name=systemd': Operation not permitted
/entrypoint.sh: line 530: set: --: invalid option
set: usage: set [-abefhkmnptuvxBCHP] [-o option-name] [--] [arg ...]
/entrypoint.sh: line 532: inspect: command not found
/entrypoint.sh: line 190: set: --: invalid option
set: usage: set [-abefhkmnptuvxBCHP] [-o option-name] [--] [arg ...]
docker: "inspect" requires a minimum of 1 argument.
See 'docker inspect --help'.

Usage:  docker inspect [OPTIONS] CONTAINER|IMAGE|TASK [CONTAINER|IMAGE|TASK...]

Return low-level information on a container, image or task
Cannot open network namespace "ip": No such file or directory

Can't load boot menu

Hello!
I trying boot to standard "boot menu" for the test, but "boot menu" give me error is below.
Can you help me fix that?

Thanks!

[pipework] daemon mode hangs when container is restarted

Investigated.

jpetazzo/pipework#152

Fixed in pipework-v1.09 image:

5b57bb5

what is the docker run command for PXE

hi, could anyone just give us the docker run command for the dreamcat4/PXE sample?

pipework - Error response from daemon: Cannot stop container pipework: no such process

This is currently happening in daemon mode, v1.0.1.

$ docker stop pipework
Error response from daemon: Cannot stop container pipework: no such process
FATA[0000] Error: failed to stop one or more containers 
$ docker kill pipework
Error response from daemon: Cannot kill container pipework: no such process
FATA[0000] Error: failed to stop one or more containers 

The cause is not known. But may be related to piping 2nd command output (subshell) into while read loop. Here:

https://github.com/dreamcat4/docker-images/blob/master/pipework/entrypoint.sh#L349

The only known solution ATM seems to be:

sudo service docker restart

standard_init_linux.go:187: exec user process caused "exec format error"

When I try to run:

docker run -v /home/alarm/recordings/:/recordings dreamcat4/tvheadend

I get

standard_init_linux.go:187: exec user process caused "exec format error"

Im trying to run on Raspberry Pi maybe that is the issue?

pipework: host side connnection timeouts and intermittent network freezing when host_routes is enabled

Detected on: ubuntu-14.10

If host_routes option is enabled. It seems that after a while, network connectivity suffers on the host. Network intermittently stops responding. And then 'wakes up again'. With sometimes messages like no route to host. Or just nothing. The connectivity of inside docker containers seems to be unaffected. The connection problems are only seen on the host.

Suspected to be something to do with Ubuntu's auto-configuration of ipv6 addresses and host-side macvlan circuits which get auto-assigned an ipv6 address in ubuntu.

Short term solution:

Disable the host_routes=true environment variable of your pipework configuration.

Long term solution:

A new option to selectively disable ipv6 routes setup. With a safer "ipv4-only" default configuration. It depend the exact nature of the issue and if the issue is related the host's ipv6 configuration.

More is testing is being conducted. Takes time.

Please use Python 3 instead of Python 2 for Limnoria.

Limnoria supports Python 3 better than Python 2 for awhile now; and Python 2 support may be dropped in the future.

iscsi: non-persistent LUN-numbers when adding new files/targets

When adding new files, the new targets are exported as expected.
However, the lun-numbers change for each target.

As there is only one LUN for one target, having LUN No=0 should be enough.

I need samples to run pxe server.

A nice project has been created to install both linux and windows over the network via pxe. However, I couldn't even run the samba-iso container despite trying for days. Can you explain a detailed example of this? My only purpose is to install linux and windows over the network.

how to support project based network using variable in compose ?

I have setup like this (under docker 1.7.1/docker-compose 1.5.2)

 environment:
    - pipework_cmd_create_br_lscx=br_lscx -i eth1 @CONTAINER_NAME@ 169.254.0.3/24
    - pipework_cmd_create_br_rscx=br_rscx -i eth0 @CONTAINER_NAME@ 169.254.1.3/24

And I want to run multi project in compose (-p) to enable multi instance based on same compose file. But now the bridge name is fixed br_lscx/br_rscx, which makes it conflict.

@CONTAINER_NAME@ looks interesting.

Can I use below or other ways to achieve it?

 environment:
    - pipework_cmd_create_br_lscx=br_@PROJECT_NAME@_lscx -i eth1 @CONTAINER_NAME@ 169.254.0.3/24
    - pipework_cmd_create_br_rscx=br_@PROJECT_NAME@_rscx -i eth0 @CONTAINER_NAME@ 169.254.1.3/24

[pipework] daemon mode doesn't work / stops working

Investigated.

moby/moby#12023

Fixed in pipework-v1.09 image:

5b57bb5

Upgrade to 4.2

It would be really appreciated if you could update the image to 4.2.

According to maximilianwollnik, it should be possible without further issue:

"I have cloned your stable repo, have replaced the line
&& echo deb https://dl.bintray.com/dreamcat4/ubuntu ${tvh_release} main > /etc/apt/sources.list.d/tvheadend.list \

with

&& echo deb https://dl.bintray.com/tvheadend/deb xenial stable-4.2 > /etc/apt/sources.list.d/tvheadend.list \

and rebuilt the image. Everything works perfectly with the latest official release."

Newbie requests help

Hi,
I'm new to using containers and was wondering if you could provide me with a little information to get me going. I've installed the tvheadend container on my QNAP 253A and everything seems to be working except... I have no devices listed under /dev/dvb. I've got 3 DVB-T 395U usb devices connected and the output of dmesg | grep "usb" is:-

[213507.536532] usb 1-4: new high-speed USB device number 8 using xhci_hcd
[213507.989515] usb 1-4.1: new high-speed USB device number 9 using xhci_hcd
[213508.122361] [usb.001.009] /sys/devices/pci0000:00/0000:00:14.0/usb1/1-4/1-4.1 added.
[213508.157517] usb 1-4.3: new high-speed USB device number 10 using xhci_hcd
[213508.531494] usb 1-4.3.2: new high-speed USB device number 11 using xhci_hcd
[213508.664655] [usb.001.011] /sys/devices/pci0000:00/0000:00:14.0/usb1/1-4/1-4.3/1-4.3.2 added.
[213508.700513] usb 1-4.3.3: new high-speed USB device number 12 using xhci_hcd
[213508.836272] [usb.001.012] /sys/devices/pci0000:00/0000:00:14.0/usb1/1-4/1-4.3/1-4.3.3 added.

I also read I may need to do --device

• /dev/dvb/adapter0/demux0
• /dev/dvb/adapter0/dvr0
• /dev/dvb/adapter0/frontend0
• /dev/dvb/adapter0/net0

But being so new to containers I've no idea where I add these commands.

Sorry to be a pain but are you able to provide me with some instruction so I can get tvheadend working with my 3 usb tv tuners (anyone?)

Many thanks

UglyGit

[samba] The permission of created file via samba is 744.

Hello, I have a question.

I created a file via samba and I checked the permission of it. The permission is 744.
But I want to set the permission of the file to non-executable.

So I checked the configuration file:

[global]
...
   directory mask = 0775
   force create mode = 0664
   force directory mode = 0775

I think global setting is that executable permission will be not set for normal files to create, but it does not work.

tvh:master sundtek dvb

Hello,

is it possible to add sundtek dvb driver to your tvheaded docker-image?

http://wiki.sundtek.de/index.php/TreiberInstallation

regards
tristan

pxe docker image - DHCP fails to give an address

Hi!
there is very little documentation on how you run your docker-image to provide a PXE boot to devices :p I found out that you took inspiration from this image: https://github.com/jpetazzo/pxe, therefore I decided to use the same method as jpetazzo:

docker build -t pxe .
PXECID=$(docker run --name pxe --cap-add NET_ADMIN -d pxe)
sudo pipework br0 $PXECID 192.168.242.1/24
sudo vconfig add eth0 72
sudo brctl addif br0 eth0.72

Then I start virtualbox for testing, starting a new image with PXE boot, and its eth0 bridged to the host br0. It actually do something in the container logs:

$ docker logs -f pxe
[s6-init] making user provided files available at /var/run/s6/etc...exited 0.
[s6-init] ensuring user provided files have correct perms...exited 0.
[fix-attrs.d] applying ownership & permissions fixes...
[fix-attrs.d] done.
[cont-init.d] executing container initialization scripts...
[cont-init.d] done.
[services.d] starting services
[services.d] done.
dnsmasq: started, version 2.76 DNS disabled
dnsmasq: compile time options: IPv6 GNU-getopt no-DBus no-i18n no-IDN DHCP DHCPv6 no-Lua TFTP no-conntrack ipset auth no-DNSSEC loop-detect inotify
dnsmasq-dhcp: DHCP, proxy on subnet 172.17.0.2
dnsmasq-tftp: TFTP root is /pxe 
dnsmasq-dhcp: no address range available for DHCP request via eth1
dnsmasq-dhcp: no address range available for DHCP request via eth1
dnsmasq-dhcp: no address range available for DHCP request via eth1
dnsmasq-dhcp: no address range available for DHCP request via eth1

But there is an error and it cannot go further. Do you have any clue on this?
Thank you a lot!

ubuntu-debootstrap is deprecated

I'm using your pipework image again and I notice that it's based on ubuntu-debootstrap which has been deprecated. You really should switch to the official ubuntu image.

i.e. change FROM ubuntu-debootstrap:15.04 to FROM ubuntu:15.04

How do you use --direct-phys?

I'm trying to pass an SR-IOV interface to a test container but adding --direct-phys seems to be parsed incorrectly.

myapp:
  image: alpine
  command: sleep inf
  environment:
    - pipework_cmd=--direct-phys eth5 @CONTAINER_NAME@ udhcpc
  net: none

pipework:
  image: dreamcat4/pipework:latest
  restart: always
  volumes:
    - /var/run/docker.sock:/docker.sock
  privileged: true
  pid: host # THIS REQUIRES COMPOSE v1.3.0 & DOCKER v1.6.0
  net: host
  environment:
    #- debug=true
    - run_mode=batch,daemon
    - host_routes=true
    - route_add_delay=1

Output from docker-compose up:
(Note the pipework_1 | /entrypoint.sh: line 190: set: --: invalid option line)

Recreating test_pipework_1 ... done
Starting test_myapp_1      ... done
Attaching to test_pipework_1, test_myapp_1
pipework_1  | ln: failed to create symbolic link '/sys/fs/cgroup/openrc/name=openrc': Operation not permitted
pipework_1  | /entrypoint.sh: line 190: set: --: invalid option
pipework_1  | set: usage: set [-abefhkmnptuvxBCHP] [-o option-name] [--] [arg ...]
pipework_1  | Device "eth1" does not exist.
pipework_1  | event_line=2022-05-08T20:33:35.700228242Z container start f86c1f1a102f483f802ffa0db3f9bee29be14fc91b5385380789ac955fad216d (com.docker.compose.config-hash=4c7d33be91c4b5e95432c49c4faa425f9378546b38a366f1704e594a9ccbd773, com.docker.compose.container-number=1, com.docker.compose.oneoff=False, com.docker.compose.project=test, com.docker.compose.project.config_files=docker-compose.yml, com.docker.compose.project.working_dir=/opt/docker/test, com.docker.compose.service=pipework, com.docker.compose.version=1.29.2, image=dreamcat4/pipework:latest, name=test_pipework_1)
pipework_1  | event_line=2022-05-08T20:33:35.787127797Z container start f6dfcc6f7ed3eee8631dcb7cbd382962445120d901ac8ac16e485834122a472c (com.docker.compose.config-hash=81a7c0e72c42a72d5316b36c8961531b614684662726684d402794969a89de26, com.docker.compose.container-number=1, com.docker.compose.oneoff=False, com.docker.compose.project=test, com.docker.compose.project.config_files=docker-compose.yml, com.docker.compose.project.working_dir=/opt/docker/test, com.docker.compose.service=myapp, com.docker.compose.version=1.29.2, image=alpine, name=test_myapp_1)
pipework_1  | /entrypoint.sh: line 190: set: --: invalid option
pipework_1  | set: usage: set [-abefhkmnptuvxBCHP] [-o option-name] [--] [arg ...]
pipework_1  | Device "eth1" does not exist.

Pipework support for Docker 1.10 and above

I realise that there are new networking features in the Docker 1.10 but these do no do what I need for my system. Hopefully they will in the future.
This container work on startup but not on live events in Docker 1.10 but there has been a change in the order of the information on the event_line.

The old line looked like
2015-06-10T16:38:12.000000000Z 753ce24472db2af099328ad161f1af70da0f4bc9fa00af2a4e82625f56eb67f2: (from dreamcat4/tvheadend:latest) start

The new line looks like:
2016-08-30T15:22:50.222511284+09:30 container start 49c7aa4ef512bf488821d6118ecf4d101461db1a3d7f7346a5cee2200ad1bb56 '(image=dreamcat4/pipework,' 'name=pipework)

I have isolated the issue to lines 444 and 445.
They currently read:
container_id="$(echo -e "$event_line_sanitized" | cut -d ' ' -f3)"
event="$(echo -e "$event_line_sanitized" | cut -d ' ' -f6)"

For Docker 1.10 need to be:
container_id="$(echo -e "$event_line_sanitized" | cut -d ' ' -f5)"
event="$(echo -e "$event_line_sanitized" | cut -d ' ' -f5)"

One fix would probably read the docker version and apply the correct template based on the version.
Another would be to tag a new version for 1.10 on docker hub.

dnsmasq: failed to create inotify: Too many open files

my ulimit -n 65535;
when docker restart ,this happend！

Update the image on Dockerhub?

The Dockerfile for pipework on Dockerhub is 3 years old while the Dockerfile here is 4 month old.

Any particular reason about that?

Support for multiple pipework commands?

Is there support for multiple pipework commands? I can easily do

pipework_cmd=eth0 -i eth2 192.168.10.10/[email protected]

To add a single fixed macvlan interface to my container, but what if I want to add 2 or 3 interfaces, each with its own requirements?

The code looks like it supports it based on the loop at https://github.com/dreamcat4/docker-images/blob/master/pipework/entrypoint.sh#L359 but how do I indicate that I want to run multiple commands?