Move bfd_opts grammar inside BFD parser code to avoid dependences between
nest and BFD grammars, which breaks when BFD build is disabled.

Add dummy bfd_opts grammar rule, so protocols can use this nonterminal
even with BFD disabled.

Thanks to Yuri Honegger for the bugreport.

In BIRD 1, we used DD-MM-YYYY, while in BIRD 2 we switched to the usual
format YYYY-MM-DD.

Thanks to Janne Pisilä for the bugreport.

In OSPFv3-IPv4 there is no requirement that link-local next hop announced
in Link-LSA must be in interface address range. Therefore, for interfaces
that do not have IPv4 address we can use some loopback IP address and
announce it as a next hop. Also we should accept such address.

BFD requires defined local IP, but for nexthop with onlink there might
not be such address. So we reject this combination of nexthop options.
This prevent crash where such combination of options is used.

This reverts commit 31aa62ae.

We can easily extend command completion to handle also keywords for
command options. Help for command options is not yet supported.

Add route attribute gw_mpls_stack to make MPLS stack of route nexthop
accessible from filters. Its type is T_CLIST, which is really not correct
(as it is a list, while T_CLIST is a set). Therefore, we keep this
attribute *undocumented* and it will be *changed* without further notice.

Based on a patch from Trisha Biswas <tbiswas@fastly.com>, thanks!

Add several arguments to 'show bfd sessions' command to filter
the list of sessions.

Job Snijders authored 1 year ago and Ondřej Zajíček committed 1 year ago

Allow to explicitly configure the source IP address for RPKI-To-Router
sessions. Predictable source addresses are useful for minimizing the
holes to be poked in ACLs.

Changed from 'source address' to 'local address' by committer.

BGP route attributes have flags (Optional, Transitive) that are validated
on decode and set to valid value on export. But if such attribute is
modified by filter or set internally by BGP during import, then its flags
would be zero in local tables. That usually does not matter, as they are
not used locally and they were fixed on export, but invalid flags leaked
in BMP and MRT dumps.

Keep route attribute flags set to valid values even when set by filters
or modified by BGP.

Job Snijders authored 1 year ago and Ondřej Zajíček committed 1 year ago

IANA registered an "Early Allocation" BGP Error code for 'Send Hold
Timer Expired' event. Update BIRD to use that error code.

Allow to define both nexthop and interface using iproute2-like syntax,
e.g.: route 10.0.0.0/16 via 10.1.0.1 dev "eth0";

Now we can avoid to use link-local scope hack (e.g. 10.1.0.1%eth0)
for cases where both nexthop and interface have to be defined.

Thanks to Marcin Saklak for the suggestion.

BIRD route attribute for RTAX_SSTHRESH metric was krt_sstresh instead of
krt_ssthresh. Fix that and keep old name as an depreacted alias.

- Add krt_fastopen_no_cookie atttibute
- Add missing krt_lock_* bits
- Fix krt_feature_allfrag
- Fix type of EA_KRT_LOCK and EA_KRT_FEATURES

Allow to set TCP congestion control algorithm using krt_congctl
route attribute.

Based on patch from Trisha Biswas <tbiswas@fastly.com>, thanks!

Alexander Zubkov authored 1 year ago and Ondřej Zajíček committed 1 year ago

We can distinguish BGP sessions if at least one side uses a different IP
address. Extend olock mechanism to handle local IP as a part of key, with
optional wildcard, so BGP sessions could local IP in the olock and not
block themselves.

Inrin authored 1 year ago and Ondřej Zajíček committed 1 year ago

The help command triggered by '?' keeps the message in readline buffer,
so it could be edited. For noninteractive shell it leads to an unexpected
side effect that `echo <cmd> ? | birdc` executes the command <cmd> after
showing its help. Avoid this by clearing the readline buffer in such case.

Increase max length of notification data in error logs from 16 to 128.
There is already enough space in the buffer.

Thanks to Marco d'Itri for the suggestion.

When a recursive route with MPLS-labeled nexthop was exported to kernel
and read back, the nexthop_same() failed due to different labels_orig
field and kernel protocol reinstalled it unnecessarily.

For comparing hext hops, route cache has to distinguish ones with
different labels_orig, but KRT has to ignore that, so we need two
nexthop compare functions.

Thanks to Marcel Menzel for the bugreport.

Katerina Kubecova authored 1 year ago and Ondřej Zajíček committed 1 year ago

Minor changes by committer.

With very busy deployments, RPKI may kill cache connection too early.
Instead of that, we want it to keep loading if any data is waiting to
be read and the reason for delay is just our congestion.

Also, when we kill the session because of actually slow cache, we want
to reload from scratch as the data we have is unreliable and nobody
knows whether the state is still valid.

In attribute cache, adata structures were stored densely in one memory
block, without regard to alignment. Let's force at least u32 alignment.