| Age | Commit message (Collapse) | Author |
|---|
|
`advertise-all-vni` is globally active
When `advertise-all-vni` is configured in the global/default BGP instance,
VyOS generated a `vni <id>` sub-block under each VRF BGP `address-family
l2vpn evpn` context. This conflicts with advertise-all-vni: FRR already
owns all kernel VNIs and returns `% Failed to create VNI` when frr-reload.py
attempts to apply the VRF-level vni sub-block. FRR then performs an early
exit from config processing, silently dropping the entire l2vpn evpn
address-family for all subsequent VRF BGP instances.
|
|
simultaneously
FRR only allows one BGP instance to hold `advertise-all-vni` at a time
(FRR issue #9405). When a default BGP instance is present it is always
started before named VRF instances, so if a named VRF holds
the flag FRR silently rejects it on every boot (regardless of default
EVPN config), causing the running config to diverge from what is
stored in VyOS.
Enforce the following policy in verify():
- Default BGP instance may always hold `advertise-all-vni`.
- A named VRF may hold it only when no default BGP instance exists.
- Only one BGP instance (default or named VRF) may hold it at a time.
The default BGP verify path additionally scans dependent VRFs so that
adding or modifying the default BGP instance while a named VRF already
holds the flag is caught even when the VRF node is not part of the
current commit.
|
|
Very simple configuration with link-state address family check.
|
|
Previously, VyOS hardcoded the kernel boot log console to either ttyS0 or
tty0, with no post-install CLI method to change it (manual GRUB edits
were required).
This commit adds a new CLI node:
system console device <name> kernel
When set, the selected serial console is used as the kernel boot console.
When removed, the kernel boot console falls back to tty0.
|
|
firewall: T8281: Normalize network prefixes in rule configuration
|
|
Convert non-canonical IPv4 and IPv6 prefixes in 'source/destination address'
fields to proper network addresses (e.g. 10.10.10.1/30 -> 10.10.10.0/30).
This prevents invalid configurations and ensures consistent prefix handling.
|
|
Replace unsupported characters (e.g. `,`, `+`) with `_` when migrating
from 1.3.x and handle name collisions safely to prevent commit failures.
|
|
|
|
Disabled "point-to-multipoint" mode for GRE interfaces, along with related "mode" node.
This mode relies on the 'teib' feature, which is not implemented yet.
|
|
|
|
|
|
|
|
|
|
|
|
Remove `dpdk-options` under `set vpp settings interface <ethN>`
and move its child options to the interface level:
- `num-rx-desc`
- `num-rx-queues`
- `num-tx-desc`
- `num-tx-queues`
Also remove `set vpp settings interface <ethN> dpdk-options promisc`.
|
|
commands
Add `set vpp settings interfaces-rx-mode <polling|interrupt|adaptive>`
to configure RX mode for all interfaces. Raise error if any interface
does not support the selected mode.
Also remove per-interface rx-mode commands:
- `set vpp settings interface <name> rx-mode`
- `set vpp kernel-interfaces <name> rx-mode`
|
|
Instead of `vpp settings unix poll-sleep-usec`
use `vpp settings poll-sleep-usec`.
|
|
This commit changes logging settings naming:
- before: `set vpp settings logging default-log-level <alert>`
- after: `set vpp settings logging default-level <alert>`
|
|
Handle absent `service dns dynamic address <iface> service` nodes
during 1-to-2 migration to prevent ConfigTreeError.
|
|
- Migration script removes 'driver' and 'xdp-options' nodes.
- XDP logic is commented out in config verification and CLI tests, preserving code for future use.
- The rest of XDP-related code remains untouched
|
|
T7664: support FRR 10.5
|
|
In FRR 10.5 ip protocols
* connected
* kernel
* local
* table
* table-direct
are removed. First three in fb8e399e4f66d09780f176fbecb99168089e64eb,
table* in 7fd030504be060387694e8a2af7f19ddd7dee39d.
In `ip protocols`, `ipv6 protocols` and `vrf` VyOS supports
* connected
* kernel
* table
Remove these from CLI, add migration scripts, update tests.
|
|
- Introduced new CLI option 'data-ciphers-fallback' for OpenVPN interfaces
(used as fallback cipher in site-to-site mode)
- Adjust migration 1‑to‑2 to skip migrating 'cipher' for site‑to‑site interfaces
|
|
According to [1] rijndael-cbc@lysator.liu.se is an alias for aes256-cbc which
was standardized in RFC4253 (2006).
This changes the migrator implementation to not only delete the old
"rijndael-cbc@lysator.liu.se" cipher from the CLI and set the new, standardized
aes256-cbc SSH cipher.
1: https://github.com/openssh/openssh-portable/commit/03e93c753d7c223063a
|
|
Follow VyOS CLI best practices for using singular whenever possible to build a
CLI node. As we introduce a new migration 2 -> 3 for SSH we can correct this
minor detail.
|
|
According to an Arch Linux forum discussion, the cipher
rijndael-cbc@lysator.liu.se was removed in OpenSSH 6.7.
References:
- https://bbs.archlinux.org/viewtopic.php?id=188613
- https://www.openssh.org/txt/release-6.7
- https://github.com/openssh/openssh-portable/commit/03e93c753d7c223063a
|
|
While producing all configtest assert files for VyOS 1.4 it was noted that
the the isis-small testcase from fails. This config fragment is not properly
migrated when updating from VyOS 1.3.0-rc1 -> 1.4 and using IS-IS.
protocols {
isis FOO {
interface eth1 {
bfd
}
net 49.0001.1921.6800.1002.00
redistribute {
ipv4 {
connected {
level-2 {
route-map EXPORT-ISIS
}
}
}
}
}
}
and results in loosing IS-IS connectivity. This is due the fact that
config.rename() does not work when only using the base tagNode.
|
|
Configuration files for config tests gathered from lab installations, customers
or our own networks were placed in multiple directories - all related to
the same thing.
We have had config-tests, configs and config.no-load. This commit re-arranges
all the files and places a proper README for the users.
|
|
syslog: T8059: migrate host:port node names to the new syntax with a dedicated port option
|
|
with a dedicated port option
|
|
This adds necessary embedded certificates to validate the migration logic added
in commit 63cc76fa236 ("openvpn: T7738: avoid duplicate certs during 1.3 -> 1.4
migration").
The CA used was generated using EasyRSA as described in our documentation:
https://docs.vyos.io/en/1.3/configuration/interfaces/openvpn.html#generate-x-509-certificate-and-keys
|
|
We also do need to adjust all other smoketests/configutests which make use of
MTU after https://github.com/vyos/vyos-build/pull/1070/commits/d84ad7e2e76bc309b
the QEMU guest MTU was lowered.
|
|
Commit f08a5700e7 ("dhcpv6: T7646: restore missing default route after upgrade")
introduced a regression affecting both non-VIF and VIF interfaces.
The addressing mode check in the migration logic was incorrect for both cases.
For non-VIF interfaces, the migration was skipped entirely due to a missing
configuration test. For VIF (VLAN) interfaces, the existing test always
evaluated to true, since it incorrectly assumed that DHCPv6 was configured
even when a static address was present.
Smoketests have been extended to cover for these cases.
|
|
|
|
T7897: VPP: fix rx-mode interrupt for XDP driver with workers
|
|
Remove no-syscall-lock from CLI and enable this option if interrupt/adaptive mode is enabled and workers are configured.
Also forbid interrupt mode for ixgbevf driver
|
|
T7887: system_option: Add validation for memory hugepage-count
|
|
This reverts commit d871fe9c4c65de87232802ed54b263c9b2824391.
|
|
Move smoke test for hugepages to test_vpp.py
|
|
bgp: T7760: remove per vrf instance system-as node
|
|
Commit de44c6aef249 ("interface: T7379: do not request SLAAC default route when
only DHCPv6 is set") introduced a behavioral change while addressing an earlier
issue fixed in e9fb2078d5ea ("interface: T7375: SLAAC assigned address is not
cleared when removing SLAAC").
This change caused unintended connectivity loss after upgrading to VyOS 1.4.3.
The underlying reason is that VyOS now requires explicit configuration to
determine how IPv6 addressing and routing should be handled, rather than making
assumptions: Requesting a DHCPv6 address alone will correctly assign an address.
However, since the IPv6 default route is typically advertised via SLAAC (and not
DHCPv6), SLAAC must also be explicitly enabled to receive the default route.
Historically, this distinction was implicit and did not require additional
configuration.
To preserve backward compatibility, a configuration migrator has been added.
It inspects existing configurations that only request a DHCPv6 address and
automatically appends the required CLI node to also enable SLAAC, ensuring that
the default route is restored after upgrade.
|
|
Originating from the bug in T7665. To avoid potential issues down the line - and
given that there's no compelling technical reason to retain the system-as CLI
node under per-VRF BGP configuration, which cannot be achieved through
alternative means - the maintainers have collectively decided to deprecate the
following command:
set vrf name <name> protocols bgp system-as <asn>
Starting with VyOS 1.4.4, this CLI command will be considered deprecated. While
it will still be accepted, it will no longer have any operational effect. A
deprecation warning will be displayed at commit time, indicating that the BGP
ASN from the global BGP configuration is now used instead.
A migration script will handle the transition and perform the following actions:
* Ensure a global BGP configuration exists; if not, initialize one.
* Iterate over all configured VRFs to determine whether a BGP instance exists
* For any insance, update the configuration to use the global system-as
and apply the local-as ASN no-prepend replace-as option on all affected
neighbors to preserve existing behavior.
* If a neighbor is already configured with a local-as directive, that neighbor
will be excluded from the migration process, as it already follows a custom
configuration.
* Add allowas-in per neighbor option. Required to not deny prefix received
updates due to as-path contains our own global ASN.
|
|
|
|
|
|
|
|
firewalls should always accept
|
|
Previously, we used a lower limit of 1 and a default value of 32768 for the
nf_conntrack_buckets (conntrack hash-size) sysctl option. However, the Linux
kernel enforces an internal minimum of 1024. A configuration migrator will now
adjust the lower limit to 1024 if necessary.
The former default value of 32768 was passed as a kernel module option, which
only took effect after the second system reboot. This was due to the option being
rendered but not applied during the first boot. This behavior has been changed so
that the value is now configurable at runtime and takes effect immediately.
Additionally, since VyOS 1.4 increased the hardware requirements to 4GB of RAM,
we now align the default value of nf_conntrack_buckets with the kernel's
default for systems with more than 1GB of RAM to 65536 entries. Previously, we
only supported half that amount.
|
|
VyOS 1.4.1 implemented support for logging facilities for HAProxy. The
facilities got included from the syslog XML definition, which also added
"virtual" or non existing facilities in HAProxy, namely: all, authpriv and mark.
If any of the above facilities is set, HAProxy will not start.
The XML definition for syslog also came with an arbitrary log-level "all" that
is also unsupported in HAProxy.
This commit adds a migration script removing the illegal CLI nodes.
|
|
|
|
LLDP is a stateless protocol which does not necessitate sending to receive
advertisements. There are multiple scenarios such as provider peering links in
which it is advantageous to receive LLDP but not disclose internal information
to the provider.
Add new CLI command:
* set service lldp interface <name> mode [disable|rx-tx|rx|tx]
The default is unchanged and will be rx-tx.
Furthermore if an interface has an explicit LLDP disable configured under
"set service lldp interface <name> disable" this will be migrated to
"set service lldp interface <name> mode disable"
|
