cEOS at Podman

I've just installed podman on Centos 7 and works fine for me

[root@localhost ~]#podman import  cEOS-lab.tar.xz ceosimage:4.23.0F
Getting image source signatures
Copying blob cbd8f51e099a done
Copying config 3a55de2204 done
Writing manifest to image destination
Storing signatures
3a55de2204db97ae78dd86d7b69a71caaa187c49ed42843bef3c5b505d3df8ae

[root@localhost ~]# podman create --name=ceos1 --privileged -e INTFTYPE=eth -e ETBA=1 -e SKIP_ZEROTOUCH_BARRIER_IN_SYSDBINIT=1 -e CEOS=1 -e EOS_PLATFORM=ceoslab -e container=podman -i -t ceosimage:4.23.0F /sbin/init systemd.setenv=INTFTYPE=eth systemd.setenv=ETBA=1 systemd.setenv=SKIP_ZEROTOUCH_BARRIER_IN_SYSDBINIT=1 systemd.setenv=CEOS=1 systemd.setenv=EOS_PLATFORM=ceoslab systemd.setenv=container=podman
ab3cdc296595e1f8f9c66e495a7e5dca2bb1e1dcb2725d40a2c0a18a47258563


[root@localhost ~]# podman start ceos1
ceos1

Wait 20-30 seconds for the container and EOS to init then you can attach via Cli or bash

[root@localhost ~]# podman exec -it ceos1 Cli
ab3cdc296595>en
localhost#
localhost#
localhost#sh run
! Command: show running-config
! device: localhost (cEOSLab, EOS-4.23.0F)
!
transceiver qsfp default-mode 4x10G
!
agent PowerManager shutdown
agent LedPolicy shutdown
agent Thermostat shutdown
agent PowerFuse shutdown
agent StandbyCpld shutdown
agent LicenseManager shutdown
!
spanning-tree mode mstp
!
no aaa root
!
no ip routing
!
end

Let me know if this works for you! Thanks! Tamas

That happened to me as well if I wanted to attach to the container straight after starting it, but waiting for half a minute worked. Does it work if you go to bash and then trigger the Cli from there? In the meantime I'll try this on Fedora 31 too.

Actually, I was trying to attach to the container after 3minutes of the start. The output of "podman ps" shows this info....

I've deployed Fedora31 and hit the same issue as you did...Did a bit of a research and started comparing it with CentOS, and it seems there are known issues in Fedora31, in F30 it seems there were no such issue, and it's not related to cEOS, as alpine and other container images are breaking as well. Here's what I did: Since it was throwing runtime error, I was checking the podman config diff between CentOS and Fedora and saw that CentOS uses runc, whereas Fedora uses crun by default. I've switched to runc on Fedora too. I've modified the runtime value

/usr/share/containers/libpod.conf

and changed

# Default OCI runtime
runtime = "runc"

after this starting the container threw

[root@localhost ~]# podman start ceos1
ERRO[0000] oci runtime "runc" does not support CGroups V2: use system migrate to mitigate
Error: unable to start container "ceos1": this version of runc doesn't work on cgroups v2: OCI runtime error

one some forums I've found that I needed to run

sudo grubby --update-kernel=ALL --args="systemd.unified_cgroup_hierarchy=0"

and reboot, but that didn't help so I just did what the error told me, ie run system migrate

root@localhost ~]# podman info  --format '{{ .host.OCIRuntime.name }}'
runc
[root@localhost ~]# podman system migrate
stopped 5677b149c1cfed6e8554cdab311a50dc384d6cf124cd9f77d15d2c54f0a2f9d2
Error: Error shutting down container storage: A layer is mounted: layer is in use by a container
[root@localhost ~]#
[root@localhost ~]#
[root@localhost ~]#
[root@localhost ~]# podman system migrate
[root@localhost ~]#
[root@localhost ~]#
[root@localhost ~]#
[root@localhost ~]#
[root@localhost ~]# podman ps
CONTAINER ID  IMAGE  COMMAND  CREATED  STATUS  PORTS  NAMES
[root@localhost ~]#
[root@localhost ~]#
[root@localhost ~]# podman ps -a
CONTAINER ID  IMAGE                                COMMAND               CREATED         STATUS                       PORTS  NAMES
5677b149c1cf  docker.io/library/ceosimage:4.23.0F  /sbin/init system...  11 minutes ago  Exited (137) 13 seconds ago         ceos1
[root@localhost ~]#
[root@localhost ~]#
[root@localhost ~]# podman start ceos1
5677b149c1cfed6e8554cdab311a50dc384d6cf124cd9f77d15d2c54f0a2f9d2
[root@localhost ~]#
[root@localhost ~]#
[root@localhost ~]#
[root@localhost ~]# podman exec -it ceos1 bash
bash-4.2#
bash-4.2#
bash-4.2# cd /mnt/flash/
bash-4.2# ls -l
total 20
-rw-rw-r--+ 1 root root 462 Nov 19 14:31 AsuFastPktTransmit.log
drwxrwxr-x+ 2 root root   6 Nov 19 14:29 Fossil
-rw-rw-r--+ 1 root root 284 Nov 19 14:31 SsuRestore.log
-rw-rw-r--+ 1 root root 284 Nov 19 14:31 SsuRestoreLegacy.log
drwxrwx---+ 3 root root  77 Nov 19 14:30 debug
drwxrwxr-x+ 2 root root   6 Nov 19 14:29 fastpkttx.backup
-rw-rw-r--+ 1 root root 161 Nov 19 14:31 kickstart-config
drwxrwxr-x+ 3 root root  72 Nov 19 14:31 persist
-rw-rwx---+ 1 root root  13 Sep 26 20:27 zerotouch-config
bash-4.2# ^C
bash-4.2#
bash-4.2#
bash-4.2# exit
Error: non zero exit code: 130: OCI runtime error
[root@localhost ~]# podman exec -it ceos1 Cli
5677b149c1cf>en
% Authorization denied for command 'enable': Default authorization provider rejects all commands
5677b149c1cf>
5677b149c1cf>en
% Authorization denied for command 'enable': Default authorization provider rejects all commands
5677b149c1cf>en
5677b149c1cf#
localhost#
localhost#sh run
! Command: show running-config
! device: localhost (cEOSLab, EOS-4.23.0F)
!
transceiver qsfp default-mode 4x10G
!
agent PowerManager shutdown
agent LedPolicy shutdown
agent Thermostat shutdown
agent PowerFuse shutdown
agent StandbyCpld shutdown
agent LicenseManager shutdown
!
spanning-tree mode mstp
!
no aaa root
!
no ip routing
!
end
localhost#

As you see I was able to to login to the Cli after a few seconds (10-20 seconds). Let me know if this works for you! Thanks Tamas

non-root user at CentOS has the same issue when CLI starts ....

it has nothing to do with the distros, you need to set up root-less podman as per https://github.com/containers/libpod/blob/master/docs/tutorials/rootless_tutorial.md Thanks for Rob Martin to pointing this out! We've been trying this out but hitting some permissions issues:

[tamas@localhost containers]$ podman logs ceos1
systemd 219 running in system mode. (+PAM +AUDIT +SELINUX +IMA -APPARMOR +SMACK +SYSVINIT +UTMP +LIBCRYPTSETUP +GCRYPT +GNUTLS +ACL +XZ +LZ4 -SECCOMP +BLKID +ELFUTILS +KMOD +IDN)
Detected virtualization other.
Detected architecture x86-64.

Welcome to CentOS Linux 7 (AltArch)!

Set hostname to .
Failed to read AF_UNIX datagram queue length, ignoring: No such file or directory
Failed to create root cgroup hierarchy: Permission denied
Failed to allocate manager object: Permission denied
[!!!!!!] Failed to allocate manager object, freezing.
systemd 219 running in system mode. (+PAM +AUDIT +SELINUX +IMA -APPARMOR +SMACK +SYSVINIT +UTMP +LIBCRYPTSETUP +GCRYPT +GNUTLS +ACL +XZ +LZ4 -SECCOMP +BLKID +ELFUTILS +KMOD +IDN)
Detected virtualization other.
Detected architecture x86-64.

Welcome to CentOS Linux 7 (AltArch)!

Set hostname to .
Failed to read AF_UNIX datagram queue length, ignoring: No such file or directory
Failed to create root cgroup hierarchy: Permission denied
Failed to allocate manager object: Permission denied
[!!!!!!] Failed to allocate manager object, freezing.

This seems to be an issue with most container images, saw many threads of various RHEL and github forums...trying to figure out how to fix these. If anyone else finds the solution in the meantime feel free to chime in! Thanks, Tamas