| Age | Commit message (Collapse) | Author |
|---|
|
User hooks are executed using run_hookdir (defined in the /sbin/dhclient-script script) instead of run-parts. That allows user hooks to modify variables set by the dhcp client (e.g., the new_routers variable to avoid the installation of the default routes).
(cherry picked from commit 645a0e768e27912a3f46d00de31d0fc79b6fd463)
|
|
|
|
|
|
* T5195: move run, cmd, call, rc_cmd helper to vyos.utils.process
* T5195: use read_file and write_file implementation from vyos.utils.file
Changed code automatically using:
find . -type f -not -path '*/\.*' -exec sed -i 's/^from vyos.util import read_file$/from vyos.utils.file import read_file/g' {} +
find . -type f -not -path '*/\.*' -exec sed -i 's/^from vyos.util import write_file$/from vyos.utils.file import write_file/g' {} +
* T5195: move chmod* helpers to vyos.utils.permission
* T5195: use colon_separated_to_dict from vyos.utils.dict
* T5195: move is_systemd_service_* to vyos.utils.process
* T5195: fix boot issues with missing imports
* T5195: move dict_search_* helpers to vyos.utils.dict
* T5195: move network helpers to vyos.utils.network
* T5195: move commit_* helpers to vyos.utils.commit
* T5195: move user I/O helpers to vyos.utils.io
|
|
Use return instead of exit in 99-dhclient-exit-hook to allow subsequent unnumbered hooks to run (like rfc3442-classless-routes). Hooks are sourced, not executed.
|
|
Add numbers for all dhclient-exit-hooks.d to enforce script order execution
Also, move '99-run-user-hooks' to '98-run-user-hooks' due to
vyatta-dhclient-hook bug and exit with 'exit 1' it is
described in the https://vyos.dev/T4856, so we should move this hook
to the end. Rename 'vyatta-dhclient-hook' to '99-vyatta-dhclient-hook'
|
|
This commit adds a script to run user-defined hook scripts upon renewing
a DHCP lease. This can be used to, for example, dynamically define a
firewall address-group based on the dynamic IP address of an interface.
For an example of its use (as well as the use case I had in mind while
coding this), see https://vyos.dev/T2196#142394
Co-authored-by: br <git@ibeep.com>
|
|
Our python scripts use the shebang logic to set an intepreter - we should
rely on this and not use an external interpreter in front of the helper.
|
|
1. When a PPPoE session is connected, `pppd` will update
`/etc/resolv.conf` regardless of `system name-server` option unless `no-peer-dns` is set.
This is because `pppd` vendors scripts `/etc/ppp/ip-up.d/0000usepeerdns` and `/etc/ppp/ip-down.d/0000usepeerdns`,
which updates `/etc/resolv.conf` on PPPoE connection and reverts the change on disconnection.
This PR removes those scripts and adds custom scripts to update name server entries through `vyos-hostsd` instead.
2. There is a typo in `/etc/dhcp/dhclient-enter-hooks.d/04-vyos-resolvconf, which misspells variable name `new_dhcp6_name_servers` as `new_dhcpv6_name_servers`.
This causes IPv6 name server entries in `vyos-hostsd` not updated
when dhclient receives nameservers from DHCPv6.
3. Regular expressions in scripts under `/etc/dhcp/dhclient-enter-hooks.d` and
`/etc/dhcp/dhclient-exit-hooks.d/` are not enclosed in `^$`, so those
IPv4 related branches (like `BOUND`) could be mistakenly executed when an IPv6
reason (like `BOUND6`) is given.
|
|
This reverts commit 78b247b724f74bdabab0706aaa7f5b00e5809bc1.
|
|
|
|
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 67b3dd6b4715fef266eb47e68623944f8be617e0)
|
|
|
|
|
|
write_file()
|
|
This reverts commit 95bbbb8bed92a60a320ff255c8b8656145f3c540.
|
|
This is the completion of commit 50a742b5 ("IPSec: T3643: Fix path for
swanctl.conf file") that moves the generated swanctl file from non-volatile to
a volatile (tmpfs backed) storage like we do for all out configuration files.
Thus it is ensured after a reboot or service deprecation there are no accidential
leftovers from previous configurations stored on the system.
|
|
selectors, and selectors with VTI.
|
|
dhcp-interface
|
|
dhclient fails to execute the script without the extension.
|
|
|
|
There were two problems with VRF support inside dhclient-script:
- VRF check inside the `01-vyos-cleanup` hook was needless because it will be done inside the `03-vyos-ipwrapper` anyway;
- VRF was ignored for in-kernel routes in `03-vyos-ipwrapper`. Theoretically, there must be no situation now when this can leads to a real problem, but better will be to keep both kernel and FRR backends in sync.
Also, the way to get and use a VRF name was changed to an easier one.
|
|
|
|
|
|
- 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.
|
|
Several improvements in processing RFC3442 routes (support for route deletion, DHCP RENEW and link-local routes)
|
|
The vyatta-dhclient-hook is not differentiating between v4 and v6
and attempting to use the same lease file for both protocol.
Therefore only dhcp v4 or v6 could be setup on an interface and
not both.
|
|
This changeset contains multiple changes in structure, logic, and bugfixes for dhclient-script. It should provide better compatibility with new Debian versions and flexibility in controlling and changing VyOS-related functions.
1. Structure change:
* All VyOS-related functionality was moved from dhclient-script itself to separated hook files.
* Old vyatta-dhclient-hook was moved from vyatta-cfg to vyos-1x.
* This change allows discard dhclient-script replacing and use the original one from Debian without any changes. So, we do not need to track all changes in upstream so carefully.
* To provide compatibility between original dhclient-script and VyOS, two internal commands/functions are repaced in hooks: ip and make_resolv_conf. So, in all places where used ${ip} or make_resolv_conf, actually using VyOS-tuned functions instead original.
* `ip` function is a wrapper, which automatically chooses what to use: transparently pass a command to /usr/sbin/ip, change a route in kernel table or FRRouting config via vtysh.
* `make_resolv_conf` function main logic was copied from current VyOS implementation and use vyos-hostsd-client for making changes
2. Added:
* Logging. Now is possible to log all changes, what is doing by dhclient-script. Logs can be saved to the journal and displayed in stderr (for debugging purposes). By default, logging to the journal is enabled (at least for some time) to provide a way to collect enough information in case if some bug in this new implementation will be found. This can be changed in the 01-vyos-logging file.
3. Fixed/Changed:
* If DHCP lease was expired, released or dhclient was stopped, dhclient-script will try to delete default route from this lease.
* Instead of blindly killing all dhclients in case if FRRouting daemon is not running, now used more intelligent logic:
* dhclients are stopping natively (with all triggers processing), instead of killing;
* dhclient-script will not kill parent dhclient process. This allows to fix the problem when systemd inform about failing to rise up interfaces at early boot stages (used in Cloud-init images);
* dhclient-script will not touch dhclients, which are not related to the current interface or IP protocol version.
* For getting FRRouting daemon status used native way via watchfrr.sh, instead of the previous trick with vtysh accessibility.
* before adding a new route to FRRouting configuration, this route will be deleted from the kernel (if it is presented there). This allows to properly replace routes, added at early boot stages, when FRR not available.
* Routes in FRRouting are adding with "tag 210". This allows protecting static routes, added via CLI, from deletion when old routes are deleting by DHCP.
* DNS servers will be reconfigured only when $new_domain_name_servers are not the same as $old_domain_name_servers. Previously, this was done during each RENEW procedure.
* Replacing MTU for preconfigured one was changed to Python (via vyos.config). The previous version with vyatta-interfaces.pl was obsoleted and seems to be broken.
|
