This is a complete redesign of how the system creates/deletes containers
from the running-config. Containers are now removed synchronously from
confd before any interfaces they may be using are removed, and created
in parallel, using a Finit task, well after confd has finished setting
up interfaces.
The logic previously provided by execd to retry container create on any
route/address changes, or periodically every 60 seconds, is now handled
by a new 'setup' command in the container wrapper script.
Additionally, container create is now split in wget/curl/podman pull of
the image and 'podman create'. This to both consolidate image fetching
and improve user feedback since most of the retry logic (above) revolves
around the image download.
Fixes#835
Signed-off-by: Joachim Wiberg <troglobit@gmail.com>
When deleting interfaces we skip a lot of steps in netdag_gen_iface(),
this patch ensures we at least call netdag_gen_ipv4_autoconf() to drop
any running zeroconf agent in the same generation.
Signed-off-by: Joachim Wiberg <troglobit@gmail.com>
The mDNS daemon, Avahi, is very verbose in its syslog output. So much
that it often overshadows other subsystems, and on switches and routers
with more than the "normal" amount of ports and interfaces, it quickly
becomes unbearable.
The patches consist of:
- Add -l LEVEL support to avahi-daemon
- Add missing casll to setlogmask() in libdaemon:daemon_set_verbosity()
Signed-off-by: Joachim Wiberg <troglobit@gmail.com>
The -S switch to less seemed at the time a good idea -- chop off too
long lines and allow horizontal scrolling -- however, the -F option, to
skip pager for content that fit the first screen is overridden by the -S
switch. This makes commands like "show containers", and similar that
normally have waaaaay too few lines for a pager to always page.
So let's drop -S to regain the original behavior. This will cause lines
to be wrapped, but with the new raw mode in the CLI this seems to work
without any weird screen artifacts.
Signed-off-by: Joachim Wiberg <troglobit@gmail.com>
Instead of having to scroll to the end (Shift G, or page down til end),
let's instead start at the end now that we have control over paging in
the CLI ourselves.
Signed-off-by: Joachim Wiberg <troglobit@gmail.com>
As it turns out, the setup with a 802.1D bridge in combination with
VLAN uppers stacked on the bridge ports is not possible to support
on mv88e6xxx ports.
Enable VLAN filtering to ensure proper isolation of the locally
terminated VLANs.
This will fail on a system running a vanilla kernel, where the dut's
ports are backed by mv88e6xxx, because:
1. The ports are attached to the same bridge, and are thus in the same
PVT group.
2. Creation of the VLAN uppers causes the DSA layer to add both ports
to the same VTU entry.
As a result, hardware behaves as if both ports had been configured as
tagged members of VLAN 10, instead of just terminating incoming
traffic locally.
Add this test to catch hardware which behaves in this way.
This patch drops a needless restriction of IP addresses on VLAN filtering
bridges from 2024-03-06. The obvious use-case is when the bridge is an
untagged member of a VLAN and only ony management VLAN is required.
Signed-off-by: Joachim Wiberg <troglobit@gmail.com>
With the FRR upgrade to 9.1.2, OSPF sometimes reports neighbors
without reporting the routerDesignatedId and routerDesignatedBackupId
attributes. Ensure that these attributes are available before trying
to copy them to the output.
During development of the upcoming ospf_container test, an issue was
discovered where zebra's view of the system drifted from the kernel's
ditto. Specifically: when transitioning from test-config to one with
many virtual interfaces (VLANs VETH pairs etc.), zebra would not
process events for all interfaces, which meant OSPF would not pick
them up.
Our experience matches very well with a bug reported upstream in:
https://github.com/FRRouting/frr/issues/13630
> If we delete kernel interface and create it and set its ip in a
> short time, in rare cases, interface ip will be lost in zebra which
> can be confirmed by vtysh show interface brirf command. This will
> lead to abnormal behavior of other protocol daemons, for example,
> bgpd does not announce the route corresponding to interface ip even
> it was specified by network command.
The issue was marked as fixed by PR 13396, which was merged in the 9.0
cycle:
https://github.com/FRRouting/frr/pull/13396
Therefore, upgrade FRR to the latest patch release from that major.
During debugging of a reconfiguration issue, one hypothesis was that
the tracking of callbacks used to determine when a transaction has
completed (as sysrepo has no hooks for this) was out of sync, causing
us to call `initctl reload` prematurely. This was not the case, but if
it ever occurs in the future, make sure that it is a fatal error that
won't go unnoticed.
Joachim Wiberg