Age | Commit message (Collapse) | Author |
|
In case if a CLI configuration is not available, dhclient cannot add
nameservers to a `resolv.conf` file, because `vyos-hostsd` requires that
an interface be listed in the `set system name-server` option.
This commit introduces two changes:
* `vyos-hostsd` service will not be started before Cloud-Init fetch all
remote data. This is required because all meta-data should be available
for Cloud-Init before any of VyOS-related services start since it is
used for configuration generation.
* the `vyos-hostsd-client` in the `dhclient-script` will be used only if
the `vyos-hostsd` is running. In other words - if VyOS services already
started, dhclient changes `resolv.conf` using `vyos-hostsd`; in other
cases - does this directly.
These changes should protect us from problems with DHCP during system
boot if DHCP is required by third-party utils.
|
|
|
|
(cherry picked from commit eceaa3a787929f5a514b9c45da52936c0d4d4a54)
|
|
Custom OpenVPN options moved back to the command line from a
configuration file. This should keep full compatibility with the
`crux` branch, and allows to avoid mistakes with parsing options
that contain `--` in the middle.
The only smart part of this - handling a `push` option. Because
of internal changes in OpenVPN, previously it did not require an
argument in the double-quotes, but after version update in
`equuleus` and `sagitta` old syntax became invalid. So, all the
`push` options are processed to add quotes. The solution is still
not complete, because if a single config line contains `push` with
other options, it will not work, but it is better than nothing.
(cherry picked from commit 3fd2ff423b6c6e992b2ed531c7ba99fb9e1a2123)
|
|
When `dhclient` with the `-x` option is used to stop running DHCP client
with a lease file that is not the same as in the new `dhclient` process,
it requires a `-lf` argument with a path to the old lease file to find
information about old/active leases and process them according to
instructions and config.
This commit adds the option to the `02-vyos-stopdhclient` hook, which
allows to properly process `dhclient` instances started in different
ways.
(cherry picked from commit 393970f9ee5b3dfc58e0e999d3d5941a198b2c6f)
|
|
We can no longer use bash veriable string code vor string manipulation. Move to
a more robust "cut" implementation.
(cherry picked from commit 513e951f3e1358ec6ff5424d03e8f4e9aa7c3388)
|
|
|
|
Backported commits:
13abffe43b2a5c41bb4ec4675c227f6cf1f868da
01158a8eaa574c48c726c20693479e4aa6e18ee6
This allows finding all running dhclient processes properly.
|
|
This is a follow-up commit to 65398e5c8 ("vrrp: keepalived: T616: move
configuration to volatile /run directory") as it makes no sense to store a
static /etc/default/keepalived file marked as "Autogenerated by VyOS" that only
enabled the SNMP option to keepalived.
Better pass the --snmp switch via the systemd override file and drop all other
references/files.
|
|
Move keepalived configuration from /etc/keepalived to /run/keepalived.
(cherry picked from commit b243795eba1b36cadd81c3149e833bdf5c5bea70)
|
|
(cherry picked from commit 2647edc30f1e02840cae62fde8b44345d35ac720)
|
|
(cherry picked from commit 7e52a7079afb522d1456833023ad58fa8b05e880)
|
|
dhclient is already handled by netplug so it's removed to avoid double
renewing of dhcp leases.
|
|
|
|
(cherry picked from commit 011f8ede55311e08d42f5afed9a2e8c44fd6d301)
|
|
(cherry picked from commit c2a1c071e7d0a9ca754d7f5016eed7db188b3d1a)
|
|
Commit dce67433 ("util: T2226: rewrite resume-vm to use run") changed the way
in which the script executed system binaries in a way which could not be
processes by the underlayin infrastructure (lists are not supported, only
strings).
(cherry picked from commit fa05e4267c17d93c4e9594a0ba32e66c7b2a1d61)
|
|
(cherry picked from commit 7de7a1b2d127df85d4224162116c1d4c984b022b)
|
|
Since in some cases a dhclient command may not end with an interface name, the
way to find out a list of dhclients running for a current interface was replaced
to catch PIDs regardless of the exact command syntax.
(cherry picked from commit 13abffe43b2a5c41bb4ec4675c227f6cf1f868da)
|
|
(cherry picked from commit dd2eb5e5686655c996ae95285b8ad7eb73d63d0b)
|
|
(cherry picked from commit ce0600e97baec18c1781605f3a80c26d4ed01e2b)
|
|
Turns out we still need it, else a MC7710 card won't work on an APU4 device.
This reverts commit f9e0fb6bffd41c143ff5454c3b73cca4a588ca86.
|
|
|
|
|
|
This is no longer required after commit for the VyOS Kernel configuration
https://github.com/vyos/vyos-build/commit/3fa7fe6926a538a8b4f25
|
|
Commit b082a6fb21 ("lcd: T2564: flatten CLI interface") by accident replaced
the template for hostapd which an LCD configuration during implementation of
T2564.
This has been corrected by restoring the content of the service configuration
from commit 8efb8ba1e (one commit earlier).
|
|
|
|
|
|
|
|
* 't2564-lcd' of github.com:c-po/vyos-1x:
lcd: T2564: flatten CLI interface
system display: T2564: Added test model
system display: T2564: Dictionary code update
system display: T2564: Conf files to /run
system display: T2564: Changed "duration" to "time"
system display: T2564: py code cleanup
system display: T2564: Replace "config (enabled|disabled)" with "display disabled"
system display: T2564: Lowercase model names
system display: T2564 Extend VyOS to support appliance LCDs
|
|
* set system lcd device <device>
* set system lcd model <modeml>
Both device and model have completion helpers for supported interfaces and LCD
displays.
|
|
|
|
|
|
T2486: DNS, vyos-hostsd fixes
|
|
|
|
|
|
|
|
For more examples on the new get_config_dict() approach migrate this
implementation as it is not yet in production use. Also this serves as proof of
concept code for further migrations.
|
|
|
|
|
|
- vyos-hostsd-client syntax changed
- track changes in changes variable
- call apply only once at the end if any changes were made
- remove 'cli-shell-api existsEffective system disable-dhcp-nameservers'
condition check as the functionality was moved into vyos-hostsd
- remove comparison between old_ and new_ variables as it caused a bug
as the nameservers didn't get updated on renew or system restart,
the dhclient lease file persists across reboots, so on boot the old
variables will contain the values from previous dhclient run so they
will usually be equal to the new variables.
|
|
|
|
During testing it was discovered that on 5 out of 10 reboots the USB
enumeration/mapping from physical port to /dev/ttyUSB is different. The root
cause is that it's a FIFO so first found/loaded driver module will be assigned
ttyUSB0.
This mixed up the serial interfaces of my FTDI chips and my connected Sierra
Wireless MC7710 card which was no longer functioning as it now was mapped to
a different USB interface.
The solution is a udev rule which persistently maps the USB-tree-device to a
device file in /dev. Wait? isn't this what /dev/serial/by-{id,path} is for?
Correct, it does the very same thing but the problem is as follows:
* by-path uses device file names which also incorporate the parent bus system,
this results in "pci-0000:00:10.0-usb-0:2.4:1.0-port0"
* by-id will overwrite the assigned device symlink if a new USB device with the
same name appears. This happens to some FTDI devices with no serial number
programmed so the device added last wins and will be the only one in
the by-id folder - cruel world!
This commit adds a new directory /dev/serial/by-bus which holds the following
device files (as example):
$ ls -1 /dev/serial/by-bus/
usb0b1.3p1.0
usb0b1.3p1.2
usb0b1.3p1.3
usb0b2.4p1.0
usb0b2.4p1.1
usb0b2.4p1.2
usb0b2.4p1.3
|
|
|
|
|
|
Several improvements in processing RFC3442 routes (support for route deletion, DHCP RENEW and link-local routes)
|
|
|
|
After migrating PowerDNS to systemd and also its configuration files to a
volatile directory in commit 77d725f ("dns-forwarding: T2185: move configuration
files to volatile /run directory") the path for the control file has not
been altered and pushed to the client rec_control binary"
|
|
|
|
|