Hi,
I am trying to checkpoint and restore a test container using https://criu.org/Docker#TTY url
, and checkpoint command gives me the "Error response from daemon: custom checkpointdir is not supported" error.

Any help will be appreciated.
[email protected]

Is there an example on how to use it for docker migration/.

Hi. I use the latest version p.haul tool to migration the number-printer docker container from one host to another, but in the destination host, in the final restore stage, an Exception occurred with the message: docker restore failed.

And here is the log:

...
12:01:38.206: 2064: Container rootfs: /var/lib/docker/aufs/mnt/a4447ddbda3f0140d5b54f90527f523e11f884c5b1f9e0585382e6024307af2c
12:01:38.206: 2064: Container config: /var/lib/docker/containers/a4447ddbda3f0140d5b54f90527f523e11f884c5b1f9e0585382e6024307af2c
12:01:38.206: 2064: Container meta: /var/run/docker/execdriver/native/a4447ddbda3f0140d5b54f90527f523e11f884c5b1f9e0585382e6024307af2c
Traceback (most recent call last):
  File "/root/p.haul/phaul/xem_rpc.py", line 42, in work
    res = getattr(self._master, "rpc_" + data[1])(*data[2])
  File "/root/p.haul/phaul/service.py", line 120, in rpc_restore_from_images
    self.htype.final_restore(self.img, self.criu_connection)
  File "/root/p.haul/phaul/p_haul_docker.py", line 183, in final_restore
    raise Exception("docker restore failed")
Exception: docker restore failed
...

And when I tried to restore the destination host's container by manually, another error occurred:

# docker restore --image-dir=/var/local/p.haul-fs/rst-kb3qan-17.07.28-12.01/img/1  a4447ddbda3f
Error response from daemon: Cannot restore container a4447ddbda3f: [2] Container does not exist: open /var/run/docker/execdriver/native/a4447ddbda3f0140d5b54f90527f523e11f884c5b1f9e0585382e6024307af2c/state.json: no such file or directory
Error: failed to restore one or more containers

Here is my docker version:

# docker version
Client:
 Version:      1.9.0-dev
 API version:  1.21
 Go version:   go1.4.2
 Git commit:   59c375a
 Built:        Wed Aug 12 16:58:42 UTC 2015
 OS/Arch:      linux/amd64
 Experimental: true

Server:
 Version:      1.9.0-dev
 API version:  1.21
 Go version:   go1.4.2
 Git commit:   59c375a
 Built:        Wed Aug 12 16:58:42 UTC 2015
 OS/Arch:      linux/amd64
 Experimental: true

criu version:

# criu -V
Version: 2.2
GitID: v2.2

I have no idea about the solution. How to solve this bug?
Thanks in advance!

Hi
I checked this wiki "https://criu.org/Live_migration", It looks like only different of normal migration is it keep the original process in stopped status. I try to install p.haul and encounter some trouble, also it looks like nobody is maintaining it anymore. Is it that possible I can do myself a live migration script since use criu is quite easy? Is there any more material about how to do live migration using Criu?

Thanks

I'm following the page https://github.com/xemul/p.haul/blob/master/test/docker/HOWTO

the docker version is 17.03.1-ce, installed at /usr/bin/docker
criu version is 3.4, installed by make and make install

Following the mtouch HOWTO step 0, protobuf & protobuf-python is installed when following CRIU installation guidance. python-rpyc is installed via pip install rpyc.

Then on both nodes, create CRIU_dir directory and command sudo criu service -o CRIU_dir/log -v4 --daemon
on destination node, start p.haul service ./p.haul-wrap service

Log for CRIU is

(00.000037) Version: 3.4 (gitid 0)
(00.000074) Warn (criu/cr-service.c:1113): Binding to local dir address!
(00.000105) The service socket is bound to ./criu_service.socket
(00.000436) Waiting for connection...

Log for p.haul is

Waiting for connection...

Am I missing something? Appreciate for your help