@MPStudyly note that tests are not run with sudo
- Feed Queries
- All Stories
- Search
- Feed Search
- Transactions
- Transaction Logs
All Stories
Jul 14 2025
@nicholaschaffeee Hi!
Are you able to describe your use case above?
Jul 13 2025
vi /opt/vyatta/share/vyatta-op/templates/update/container/image/node.tag/node.def
Replace the run command with
I didn't get this issue, so I created another ticket.
I fixed it.
PR
Jul 12 2025
Jul 11 2025
That's odd, I really wonder what the difference between our environments is. I'm running all tests inside the build container directly calling "sudo make <target>" one after another for the targets:
In our environment, all tests passed as during build, as installed on the test VM
DEBUG - Running Testcase: /usr/libexec/vyos/tests/smoke/cli/test_vrf.py DEBUG - test_vrf_assign_interface (__main__.VRFTest.test_vrf_assign_interface) ... ok DEBUG - test_vrf_bind_all (__main__.VRFTest.test_vrf_bind_all) ... ok DEBUG - test_vrf_conntrack (__main__.VRFTest.test_vrf_conntrack) ... ok DEBUG - test_vrf_disable_forwarding (__main__.VRFTest.test_vrf_disable_forwarding) ... ok DEBUG - test_vrf_ip_ipv6_nht (__main__.VRFTest.test_vrf_ip_ipv6_nht) ... ok DEBUG - test_vrf_ip_ipv6_protocol_non_existing_route_map (__main__.VRFTest.test_vrf_ip_ipv6_protocol_non_existing_route_map) ... ok DEBUG - test_vrf_ip_protocol_route_map (__main__.VRFTest.test_vrf_ip_protocol_route_map) ... ok DEBUG - test_vrf_ipv6_protocol_route_map (__main__.VRFTest.test_vrf_ipv6_protocol_route_map) ... ok DEBUG - test_vrf_link_local_ip_addresses (__main__.VRFTest.test_vrf_link_local_ip_addresses) ... ok DEBUG - test_vrf_loopbacks_ips (__main__.VRFTest.test_vrf_loopbacks_ips) ... ok DEBUG - test_vrf_static_route (__main__.VRFTest.test_vrf_static_route) ... ok DEBUG - test_vrf_table_id_is_unalterable (__main__.VRFTest.test_vrf_table_id_is_unalterable) ... ok DEBUG - test_vrf_vni_add_change_remove (__main__.VRFTest.test_vrf_vni_add_change_remove) ... ok DEBUG - test_vrf_vni_and_table_id (__main__.VRFTest.test_vrf_vni_and_table_id) ... ok DEBUG - test_vrf_vni_duplicates (__main__.VRFTest.test_vrf_vni_duplicates) ... ok DEBUG - DEBUG - ---------------------------------------------------------------------- DEBUG - Ran 15 tests in 308.260s DEBUG -
For example, the latest rolling:
vyos@r14:~$ /usr/libexec/vyos/tests/smoke/cli/test_vrf.py test_vrf_assign_interface (__main__.VRFTest.test_vrf_assign_interface) ... ok test_vrf_bind_all (__main__.VRFTest.test_vrf_bind_all) ... ok test_vrf_conntrack (__main__.VRFTest.test_vrf_conntrack) ... ok test_vrf_disable_forwarding (__main__.VRFTest.test_vrf_disable_forwarding) ... ok test_vrf_ip_ipv6_nht (__main__.VRFTest.test_vrf_ip_ipv6_nht) ... ok test_vrf_ip_ipv6_protocol_non_existing_route_map (__main__.VRFTest.test_vrf_ip_ipv6_protocol_non_existing_route_map) ... ok test_vrf_ip_protocol_route_map (__main__.VRFTest.test_vrf_ip_protocol_route_map) ... ok test_vrf_ipv6_protocol_route_map (__main__.VRFTest.test_vrf_ipv6_protocol_route_map) ... ok test_vrf_link_local_ip_addresses (__main__.VRFTest.test_vrf_link_local_ip_addresses) ... ok test_vrf_loopbacks_ips (__main__.VRFTest.test_vrf_loopbacks_ips) ... ok test_vrf_static_route (__main__.VRFTest.test_vrf_static_route) ... ok test_vrf_table_id_is_unalterable (__main__.VRFTest.test_vrf_table_id_is_unalterable) ... ok test_vrf_vni_add_change_remove (__main__.VRFTest.test_vrf_vni_add_change_remove) ... ok test_vrf_vni_and_table_id (__main__.VRFTest.test_vrf_vni_and_table_id) ... ok test_vrf_vni_duplicates (__main__.VRFTest.test_vrf_vni_duplicates) ... ok
Any updates on this? Just re-ran the tests to discover there are quite a lot of other/new failures. I also noticed the nightly builds workflow to be disabled since 4 months. Still there are recent nightly releases available in that repository. Did you internally switch to some other workflow that's not public (yet)? Otherwise I'd not assume you've quit running the tests yourself.
Seems fixed
<pre>vyos@r16# set load-balancing wan interface-health eth1 nexthop 192.0.2.1 [edit] vyos@r16# commit [edit] vyos@r16# exit Warning: configuration changes have not been saved. exit vyos@r16:~$ vyos@r16:~$ show wan-load-balance Interface: eth1 Status: active Last Status Change: N/A Last Interface Success: N/A Last Interface Failure: N/A Interface Failures: 0
As an example (and the reason why i try to solve this issue the way i am doing it right now), the junos module (https://docs.ansible.com/ansible/latest/collections/junipernetworks/junos/junos_config_module.html#ansible-collections-junipernetworks-junos-junos-config-module) support this exact behaviour. So it was easy to adapt this for our vyos firewalls.
It seems that community plugin netcommon network_cli breaks long lines resulting in diff. I tried ansible_terminal_width but no luck
I will be looking for some workarounds, unless there is already an open issue with Netcommon
Fix of commit-config after changes: https://github.com/vyos/vyatta-cfg/pull/107
Jul 10 2025
I can see the community module we use (necommon/network_cli) actually returns the running config with a broken long strings - it inserts extra space
Hi,
I am testing against 1.4.2, via pylibssh, - I see 4096-long one does break idempotency but I see no Exception
I will retest with vyos_user
Looking at vyos.vyos.vyos_firewall_global, i have a few issues with it:
Using this playbook i can reproduce the same bug with the vyos_user command (2048 does not change on the second run, 4096 does):
@nicolasberens You may want to give the vyos_user command a try as it should deal with this appropriately.
I think this may be related to the other T7621, since they would both involve the diff engine for the cliconf plugin.
Hi,
thanks for the reply.
Jul 9 2025
Hi @nicolasberens!
I believe the module is configured to match line by line by default and only work out on deltas in the config.
So if in the active Ansible session, the module does not sense changes after 'delete' (because there are none as yet, since the command is not applied) - At start, the module retrieves the current configuration and process it, then issues a set of commands.
Either having two separate tasks to delete first, then to set works, or the example below also works.
Hi All,
Looking into this