lockfree.c: Solution for lfjour_cleanup_hook() too often scheduling. This function was scheduled each time any of the journal recipients
reached end of a block of journal items or read all of journal items. lfjour_cleanup_hook() can clean only journal items every recipient has processed,
so it was often called uselessly.
This commit restricts most of the unusefull scheduling. Only some consumers are given a token alowing them to try to schedule the lfjour_cleanup_hook().
When a consumer wants to schedule the lfjour_cleanup_hook(), it checks whether it has a token. If yes, it decrements number of tokens the journal has given (issued_tokens)
and discards its own token. If issued_tokens reaches zero, the consumer is allowed do schedule the lfjour_cleanup_hook().
There is a maximum number of tokens a journal can give to its customers (max_tokens). A new customer is given a token in its init, if the maximum number of tokens was not reached.
The rest of tokens is given to customers in lfjour_cleanup_hook(). In lf_jour_cleanup_hook(), it is increased the issued_tokens number in order not to call the hook
before it finishes. Then, tokens are given to the slowest recipients (but never to more than max_token recipients). Before leaving lfjour_cleanup_hook(),
the issued_tokens number is decreased. If no other tokens are given, we have to make sure the lfjour_cleanup_hook will be called again. If every item in journal
was read by every recipient, tokens are given to random recipients. If all recipients with tokens managed to finish until now, we give the token to first
unfinished customer we find or we call the hook again.
Ondrej Zajicek [Sun, 1 Dec 2024 23:00:36 +0000 (00:00 +0100)]
BMP: Refactor route monitor message serialization
Instead of several levels of functions, just have two functions
(one for routes, the other for end-of-rib), this allows to create
messages in a simple linear fashion.
Also reduce three duplicite functions to construct BGP header for
BMP messages to just one.
Maria Matejka [Tue, 17 Sep 2024 14:27:54 +0000 (16:27 +0200)]
BMP: simplified update queuing and better memory performance
This commit is quite a substantial rework of the underlying layers in
BMP TX:
- several unnecessary layers of indirection dropped, including most of
the original BMP's buffer machinery
- all messages are now written directly into one protocol's buffer
allocated for the whole time big enough to fit every possible message
- output blocks are allocated by pages and immediately returned when
used, improving the overall memory footprint
- no intermediary allocation is done from the heap altogether
- there is a documented and configurable limit on the TX queue size
Maria Matejka [Thu, 28 Nov 2024 13:10:40 +0000 (14:10 +0100)]
RPKI: refactored pdu to host byte order conversion
We shouldn't convert bytes 2 and 3 of the PDU blindly, there are several
cases where these are used by bytes. Instead, the conversion is done
only where needed.
This fixes misinterpretation bug of ASPA PDU flags on little endian
architectures.
Maria Matejka [Wed, 27 Nov 2024 11:01:58 +0000 (12:01 +0100)]
String tests: fixed too strict strcmp checks
The strcmp function is not guaranteed to return -1 or +1
but any negative or positive value if the input strings
are different. Fixed the false assumption which triggered
a build bug on emulated arm64.
Ondrej Zajicek [Wed, 27 Nov 2024 03:15:16 +0000 (04:15 +0100)]
Fixed crash if logging happened in unit tests
The patch initializes logging in unit tests. Previously, unit tests did
not initialize logging and other subsystems, just resources. But
resource_init() could under certain circumstances trigger logging and
cause crash.
The bug was Found by Jakub Ruzicka, dissected by David Petera and Maria
Matejka, disguised as failing build for Debian arm64 in pbuilder
emulation which did not like disabling THP.
Ondrej Zajicek [Tue, 26 Nov 2024 16:46:27 +0000 (17:46 +0100)]
RPKI: Fix PDU length check
The END_OF_DATA PDU was extended in version 1, so it has different length
in different versions. We should do the PDU length check according to its
version.
Maria Matejka [Mon, 25 Nov 2024 11:02:13 +0000 (12:02 +0100)]
Protocol restart timer reworked.
The restart timer was racy and didn't allow for immediate restarts
from limits. Now the protocols stores the last restart time and in case
of too frequent autorestarts caused by exceeded limits, the protocol
gets disabled with an error message.
Maria Matejka [Fri, 22 Nov 2024 13:49:13 +0000 (14:49 +0100)]
Protocol and BGP state information cleanup and fixes
There were some nasty problems with deferred protocol state updates and
race conditions on BGP startup, shutdown, and also with referencing the
cached states.
To allow reading of protocol states from other protocols or completely
different routines, we have to export these states to data structures
not requiring to lock the protocol loops.
On one hand, this doesn't give the reader the actual state "right now",
on the other hand, getting that is impossible in a properly
multithreaded environment and you will always get the information with
some (little but noteworthy) delay.
This implementation handles only the basic state information of the
protocols, common for all the protocols. Adding protocol-specific state
information should be done by implementing the protocol hook init_state().
Channel information is stored but not announced, as we don't need the
announcements for now.
For the upcoming rework of protocol state information propagation,
we need some more eattr types to be defined.
These types are probably not defined completely and before using
them for route attributes, you should check that they don't lack
some crucial methods.
Piotr Wydrych [Tue, 5 Nov 2024 17:53:22 +0000 (18:53 +0100)]
CLI: Add timeformat command
Adds ability to override time format of show commands for current CLI session
so that it does not depend on configuration and may ease parsing when CLI is
called from tools.
Eric Long [Wed, 16 Oct 2024 19:32:36 +0000 (21:32 +0200)]
Flowspec: Fix IPv6 prefix when offset is not multiple of 8
Current implementation handles flowspec prefix length and offset only
in bytes, but RFC 8956 (Dissemination of Flow Specification Rules for
IPv6) Section 3.1 [1] and example in Section 3.8.2 [2] states the
pattern should begin right after offset *bits*.
For example, pattern "::1:1234:5678:9800:0/60-104" is currently
serialized as "02 68 3c 01 12 34 56 78 98", but it should shift its
pattern 4 more bits to the left: "02 68 3c 11 23 45 67 89 80".
This patch implements shifting left/right for IPv6 type and use it to
correct the behaviour. Test data are replaced with the correct ones.
Ondrej Zajicek [Tue, 8 Oct 2024 17:34:51 +0000 (19:34 +0200)]
Filter: Fixes and improvements related to case/sets
Unify grammar for set_atom and switch_atom to avoid inconsistencies
between them. Fix errors in documentation related to case statement
and set type. Change 'vpnrd' to 'rd' to be consistent with the filter
language.
Job Snijders [Thu, 3 Oct 2024 13:43:12 +0000 (15:43 +0200)]
RPKI: Add TCP-MD5 authentication option
RPKI-To-Router (RTR) sessions seem to be similar security-sensitivity as
IBGP sessions. BIRD already offered a choice of either "plain TCP" (meh)
or "SSH" (secure, albeit a bit more hassle to set up than TCP-MD5).
The patch adds TCP-MD5 as another option. TCP-MD5 for RTR is specified
through RFC 6810 section 7.3 and RFC 8210 section 9.3.
Before this commit, on kernel shutdown, the routes were re-exported by
the regular export but treated as withdraw. This was too hairy and
caused unnecessary complexity of the protocol's state machine.
Instead of that, we found out that it makes more sense to just refeed
the routes synchronously and convert to withdraw. This is done by the
direct export access instead of the channel.
It would (maybe) make more sense to run export filters on this in case
the export filter updates the krt_metric attribute, but as this doesn't
work on regular withdraw anyway, it's better for now to just let it be
and maybe somebody in the future fixes this issue.