| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
|
|
|
|
The call to rc_cmd('podman inspect ...') determines whether a container image
scheduled for deletion has any ancestor containers still using it. Previously,
if the podman command wrote output to stderr, rc_cmd() would return that error
message alongside or instead of the ancestor container ID.
This caused subsequent podman calls to fail, as the error string was incorrectly
treated as a valid command argument. This change ensures only valid ancestor
IDs are returned.
This is a fix for commit a99ca6d11b5 ("op-mode: T7403: add option for forcefully
remove a container image")
|
|
syslog: T4251: Fix TLS enablement logic for syslog
|
|
is sent
Add support for specifying a custom health check port for HAProxy backend servers.
This allows health probes to target a dedicated endpoint - such as port 8080 - separate
from normal traffic ports (e.g., 80 or 443).
|
|
|
|
|
|
|
|
T3680: protocols: add dhclient hooks for dhcp-interface static routes
|
|
frr: T7664: properly set log configuration during daemon startup
|
|
T7948: always call setUp() and tearDown() base class methods
|
|
In large networks with many zones where simple allow/deny rules are not sufficient,
zones become tedious to manage. Many use cases can be simplified by providing an
ability to define a default ruleset for traffic from other zones. This change proposes
adding the follwing syntax:
set firewall zone <name> default_firewall name <name>
set firewall zone <name> default_firewall ipv6_name <name>
The proposed behavior is the following:
local in:
The default firewall ruleset for the local zone will be appended after all
from configurations.
local out:
If a non-local zone does not have a from local ruleset but does have a
default_firewall ruleset, the default_firewall ruleset will be appended using
oifname
forward:
The default firewall ruleset for the zone will be appended after all from
configurations
To keep the behavior consistent with from ruleset configurations, a return is appended
after the default_firewall ruleset.
The proposed behavior differs slightly from the default_policy configuration for the
local out chains. The default_policy applied in the out templates comes from the local
zone, not the actual outbound zone. The proposed change does not amend this, but does
make default_firewall logically consistent with the intent of the out rules.
|
|
Due to an issue in FRR where dynamic PPPoE interfaces sometimes receive an
ifIndex of 0 (which is invalid), disable the testcase and add a marker.
Test was just recently added in commit 75e9fd60c ("smoketest: T7858: add PPPoE
client tests with IPv4, IPv6 and DHCPv6-PD")
|
|
This extends commit 2c521f135 ("smoketest: T7858: make failfast main argument
dynamic").
|
|
Upgrades to FRR 10.4 have shown that there are cases where FRRs mgmtd will
SIGSEGV - to cover for this case and fail the smoketests we will monitor the
PID of the mgmtd process.
|
|
This extends commit 2c521f135 ("smoketest: T7858: make failfast main argument
dynamic").
|
|
|
|
While working on task T7664 (FRR 10.4 upgrade), I identified the need for
additional validation and safeguards around the FRR management daemon. The
most appropriate place for this logic is in the setUp() and tearDown() methods
of the smoketest base class, VyOSUnitTestSHIM.
However, during implementation, it became apparent that test cases do not
consistently invoke the base class's setup and teardown methods. This
inconsistency complicates the process of capturing the FRR mgmtd PID at the
start of a test and verifying that it remains unchanged by the end - a key step
in detecting crashes or unexpected terminations (e.g., SIGSEGV) of the FRR
management daemon during tests.
|
|
kea: T7925: Improve error handling, validate IPv6 PD prefix length
|
|
Solves the problem that vyos-configs in FRRender caches configuation and
DHCP changes are ignored.
* Add src/helpers/vyos-request-configd-update.py that requests vyos-configd
to update FRR configuration.
* Make dhclient hooks use it instead of calling protocols_static.py
* Make FRRender cache not only configuration but also DHCP gateways so
that is any of them changes, FRR configuration is updated
|
|
|
|
Execute commands for vpp sflow with API calls. Use values for polling interval and sampling rate from 'system sflow'. Add op-mode command
|
|
T7946: log redirected stdout from FRRender
|
|
T5942: Make failover support dhcp-interface
|
|
T7930: VPP: Changing NAT44 settings resets `forwarding_enabled` to False
|
|
T7929: VPP: nat44: validate that only self-twice-nat external address is in translation pool
|
|
translation pool
|
|
Related to commits:
* fca49413f - frrender: T7664: do not log unique-id in syslog messages
* c4c339fb5 - frrender: T7664: reduce log level to notifications during
normal operation
* 7ff824f44 - frrender: T7664: add "log timestamp precision 3" global option
|
|
frrender: T7927: de-nest DHCP and PPPoE interface section for VRFs
|
|
After T7855, logging of stdout from FRRender was dropped. Explicity log
redirected stdout in vyos-configd/commitd; restore parity of logging
between the two daemons.
|
|
Enable/disable NAT forwarding in vpp_nat.py script to prevent it's reset
|
|
T7942: consistent naming of "memory" in op-mode
|
|
T7878: conflict check updated to use mergify
|
|
kea: T7821: Fix subnet-id accepted range
|
|
T7941: fix DHCP client running in VRF with non-word characters
|
|
|
|
'Subnet IDs must be greater than zero and less than 4294967295.'
|
|
|
|
|
|
smoketest: T7657: check for Kernel option CONFIG_SLUB_DEBUG
|
|
The previous implementation used awk with a regex to extract the VRF name from
JSON data, relying on "(\w+)" to match the value. This broke for valid VRF
names containing hyphens or other non-word characters.
This update replaces the regex-based extraction with a jq query that reliably
parses the JSON structure, ensuring correct behavior regardless of VRF name
format. This also reduces parsing fragility by using a tool purpose-built for
JSON processing.
|
|
We have the following memory related commands:
* show hardware mem
* show system memory <cache | detail | routing-daemons>
Most of the time we call it memory so this should be also made consistent in
the op-mode CLI. This is NOT a breaking change as mem is an abbreviation to
memory, so it's backwards compatible.
|
|
This complements commit cf4666257 ("frrender: T7927: de-nest DHCP and PPPoE
interface section for VRFs") by validating the new DHCP behavior.
When a VRF bound interface obtains an IP via DHCP, the default route received
must be reflected in FRR's routing table (vtysh). Ensure the dynamic DHCP
learned route is retained even when unrelated sections of the FRR configuration
are modified.
|
|
|
|
While debugging and resolving the underlying VRF assignment issue in T7929,
this code path was refactored for clarity. Previously, the check_dhcp() helper
function relied on the parent scope’s vrf variable. Although this worked
correctly, it was not clean or explicit. The function now takes and uses its
own vrf argument, improving readability and maintainability.
|
|
This update refactors the code path to eliminate unnecessary nesting when
processing interfaces using PPPoE or DHCP that are bound to a VRF.
Previously, the configuration dictionary was only populated if a static route
was defined under the corresponding VRF node. This behavior was incorrect, as
interfaces tied to a VRF can receive dynamic routes - such as a default route
via DHCP - without having any static routing configuration. A typical use case
is a VRF dedicated to out-of-band management.
|
|
|
|
|
