For now, all route attributes are stored as eattrs in ea_list. This
should make route manipulation easier and it also allows for a layered
approach of route attributes where updates from filters will be stored
as an overlay over the previous version.
As there is either a nexthop or another destination specification
(or othing in case of ROAs and Flowspec), it may be merged together.
This code is somehow quirky and should be replaced in future by better
implementation of nexthop.
Also flowspec validation result has its own attribute now as it doesn't
have anything to do with route nexthop.
This doesn't do anything more than to put the whole structure inside
adata. The overall performance is certainly going downhill; we'll
optimize this later.
Anyway, this is one of the latest items inside rta and in several
commits we may drop rta completely and move to eattrs-only routes.
The prefix hash table in BGP used the same hash function as the rtable.
When a batch of routes are exported during feed/flush to the BGP, they
all have similar hash values, so they are all crowded in a few slots in
the BGP prefix table (which is much smaller - around the size of the
batch - and uses higher bits from hash values), making it much slower due
to excessive collisions. Use a different hash function to avoid this.
Also, increase the batch size to fill 4k BGP packets and increase minimum
BGP bucket and prefix hash sizes to avoid back and forth resizing during
flushes.
This leads to order of magnitude faster flushes (on my test data).
The route scope attribute was used for simple user route marking. As
there is a better tool for this (custom attributes), the old and limited
way can be dropped.
Some tokens are both keywords and symbols. For now, we allow only
specific keywords to be redefined; in future, more of the keywords may
be added to this category.
The redefinable keywords must be specified in any .Y file as follows:
toksym: THE_KEYWORD ;
See proto/bgp/config.Y for an example.
Also dropped a lot of unused terminals.
Changes in internal API:
* Every route attribute must be defined as struct ea_class somewhere.
* Registration of route attributes known at startup must be done by
ea_register_init() from protocol build functions.
* Every attribute has now its symbol registered in a global symbol table
defined as SYM_ATTRIBUTE
* All attribute ID's are dynamically allocated.
* Attribute value custom formatting hook is defined in the ea_class.
* Attribute names are the same for display and filters, always prefixed
by protocol name.
Also added some unit testing code for filters with route attributes.
Implicit paddings have undefined values in C. We want the eattr blocks
to be comparable by memcmp and eattrs settable directly by structrure
literals. This check ensures that all paddings in eattr and bval are
explicit and therefore zeroed in all literals.
This commit removes the EAF_TYPE_* namespace completely and also for
route attributes, filter-based types T_* are used. This simplifies
fetching and setting route attributes from filters.
Also, there is now union bval which serves as an universal value holder
instead of private unions held separately by eattr and filter code.
Before this change, fetch-update-write and bitmasking was hardcoded in
attribute access code cased by the attribute type. Several filter
instructions are used to do it instead.
As this is certainly going to be a little bit slower than before, the
switch block in attribute access code should be completely removed in
near future, helping with both performance and code cleanliness.
The user interface should have stayed intact.
The interface pointer was improperly converted to u32 and back. Fixing
this by explicitly allocating an adata structure for it. It's not so
memory efficient, we'll optimize this later.
The prune loop may may rebuild the prefix trie and therefore invalidate
walk state for asynchronous walks (used in 'show route in' cmd). Fix it
by adding locking that keeps the old trie in memory until current walks
are done.
In future this could be improved by rebuilding trie walk states (by
lookup for last found prefix) after the prefix trie rebuild.
When rtable is pruned and network fib nodes are removed, we also need to
prune prefix trie. Unfortunately, rebuilding prefix trie takes long time
(got about 400 ms for 1M networks), so must not be atomic, we have to
rebuild a new trie while current one is still active. That may require
some considerable amount of temporary memory, so we do that only if
we expect significant trie size reduction.
Implement flowspec validation procedure as described in RFC 8955 sec. 6
and RFC 9117. The Validation procedure enforces that only routers in the
forwarding path for a network can originate flowspec rules for that
network.
The patch adds new mechanism for tracking inter-table dependencies, which
is necessary as the flowspec validation depends on IP routes, and flowspec
rules must be revalidated when best IP routes change.
The validation procedure is disabled by default and requires that
relevant IP table uses trie, as it uses interval queries for subnets.
When output of 'show route' command was generated, the net_format() was
called for each network prematurely, even if the result was not needed.
Fix the code to call net_format() only when needed. This makes queries
that process many networks but show only few (e.g. 'show route where ..',
or 'show route count') much faster (like 5x - 10x faster).
Attach a prefix trie to IP/VPN/ROA tables. Use it for net_route() and
net_roa_check(). This leads to 3-5x speedups for IPv4 and 5-10x
speedup for IPv6 of these calls.
TODO:
- Rebuild the trie during rt_prune_table()
- Better way to avoid trie_add_prefix() in net_get() for existing tables
- Make it configurable (?)
Add operators .min and .max to find minumum or maximum element in sets
of types: clist, eclist, lclist. Example usage:
bgp_community.min
bgp_ext_community.max
filter(bgp_large_community, [(as1, as2, *)]).min
Signed-off-by: Alexander Zubkov <green@qrator.net>
Compare all IA_* flags that are set by sysdep iface code.
The old code ignores IA_SECONDARY flag when comparing whether iface
address updates from kernel changed anything. This is usually not an
issue as kernel removes all secondary addresses due to removal of the
primary one, but it breaks when sysctl 'promote_secondaries' is enabled
and kernel promotes secondary addresses to primary ones.
Thanks to 'Alexander' for the bugreport.
This feature is intended mostly for checking that BIRD's allocation
strategies don't consume much memory space. There are some cases where
withdrawing routes in a specific order lead to memory fragmentation and
this output should give the user at least a notion of how much memory is
actually used for data storage and how much memory is "just allocated"
or used for overhead.
Also raising the "system allocator overhead estimation" from 8 to 16
bytes; it is probably even more. I've found 16 as a local minimum in
best scenarios among reachable machines. I couldn't find any reasonable
method to estimate this value when BIRD starts up.
This commit also fixes the inaccurate computation of memory overhead for
slabs where the "system allocater overhead estimation" was improperly
added to the size of mmap-ed memory.
To access route attribute cache from multiple threads at once, we have
to lock the cache on writing. The route attributes data structures are
safe to read unless somebody tries to tamper with the cache itself.
This commit prevents use-after-free of routes belonging to protocols
which have been already destroyed, delaying also all the protocols'
shutdown until all of their routes have been finally propagated through
all the pipes down to the appropriate exports.
The use-after-free was somehow hypothetic yet theoretically possible in
rare conditions, when one BGP protocol authors a lot of routes and the
user deletes that protocol by reconfiguring in the same time as next hop
update is requested, causing rte_better() to be called on a
not-yet-pruned network prefix while the owner protocol has been already
freed.
In parallel execution environments, this would happen an inter-thread
use-after-free, causing possible heisenbugs or other nasty problems.
This basically means that:
* there are some more levels of indirection and asynchronicity, mostly
in cleanup procedures, requiring correct lock ordering
* all the internal table operations (prune, next hop update) are done
without blocking the other parts of BIRD
* the protocols may get their own loops very soon
There is a simple universal IO loop, taking care of events, timers and
sockets. Primarily, one instance of a protocol should use exactly one IO
loop to do all its work, as is now done in BFD.
Contrary to previous versions, the loop is now launched and cleaned by
the nest/proto.c code, allowing for a protocol to just request its own
loop by setting the loop's lock order in config higher than the_bird.
It is not supported nor checked if any protocol changed the requested
lock order in reconfigure. No protocol should do it at all.
In some specific configurations, it was possible to send BIRD into an
infinite loop of recursive next hop resolution. This was caused by route
priority inversion.
To prevent priority inversions affecting other next hops, we simply
refuse to resolve any next hop if the best route for the matching prefix
is recursive or any other route with the same preference is recursive.
Next hop resolution doesn't change route priority, therefore it is
perfectly OK to resolve BGP next hops e.g. by an OSPF route, yet if the
same (or covering) prefix is also announced by iBGP, by retraction of
the OSPF route we would get a possible priority inversion.
* internal tables are now more standalone, having their own import and
export hooks
* route refresh/reload uses stale counter instead of stale flag,
allowing to drop walking the table at the beginning
* route modify (by BGP LLGR) is now done by a special refeed hook,
reimporting the modified routes directly without filters
Channels have now included rt_import_req and rt_export_req to hook into
the table instead of just one list node. This will (in future) allow for:
* channel import and export bound to different tables
* more efficient pipe code (dropping most of the channel code)
* conversion of 'show route' to a special kind of export
* temporary static routes from CLI
The import / export states are also updated to the new algorithms.
Routes are now allocated only when they are just to be inserted to the
table. Updating a route needs a locally allocated route structure.
Ownership of the attributes is also now not transfered from protocols to
tables and vice versa but just borrowed which should be easier to handle
in a multithreaded environment.
Pipes copy the original rte with old values, so they require rte to be
exported with stored tmpattrs. Other protocols access stored attributes
using eattr list, so they require rte to be exported with expanded
tmpattrs. This is temporary hack, we plan to remove whoe tmpattr mechanism.
Thanks to Paul Donohue for the bugreport.
In most cases of export there is no need to store back temporary
attributes to rte, as receivers (protocols) access eattr list anyway.
But pipe copies the original rte with old values, so we should store
tmpattrs also during export.
Thanks to Paul Donohue for the bugreport.
Add support for specifying a password in hexadecimal format, The result
is the same whether a password is specified as a quoted string or a
hex-encoded byte string, this just makes it more convenient to input
high-entropy byte strings as MAC keys.
The Babel MAC authentication RFC recommends implementing Blake2s as one of
the supported algorithms. In order to achieve do this, add the blake2b and
blake2s hash functions for MAC authentication. The hashing function
implementations are the reference implementations from blake2.net.
The Blake2 algorithms allow specifying an arbitrary output size, and the
Babel MAC spec says to implement Blake2s with 128-bit output. To satisfy
this, we add two different variants of each of the algorithms, one using
the default size (256 bits for Blake2s, 512 bits for Blake2b), and one
using half the default output size.
Update to BIRD coding style done by committer.
In general, events are code handling some some condition, which is
scheduled when such condition happened and executed independently from
I/O loop. Work-events are a subgroup of events that are scheduled
repeatedly until some (often significant) work is done (e.g. feeding
routes to protocol). All scheduled events are executed during each
I/O loop iteration.
Separate work-events from regular events to a separate queue and
rate limit their execution to a fixed number per I/O loop iteration.
That should prevent excess latency when many work-events are
scheduled at one time (e.g. simultaneous reload of many BGP sessions).
This is an implementation of draft-walton-bgp-hostname-capability-02.
It is implemented since quite some time for FRR and in datacenter, this
gives a nice output to avoid using IP addresses.
It is disabled by default. The hostname is retrieved from uname(2) and
can be overriden with "hostname" option. The domain name is never set
nor displayed.
Minor changes by committer.
If there are roa_check() calls in channel filters, then the channel
subscribes to ROA table notifications, which are sent when ROA tables
are updated (subject to settle time) and trigger channel reload or
refeed.
The patch add support for per-channel debug flags, currently just
'states', 'routes', and 'filters'. Flag 'states' is used for channel
state changes, remaining two for routes passed through the channel.
The per-protocol debug flags 'routes'/'filters' still enable reporting
of routes for all channels, to keep existing behavior.
The patch causes minor changes in some log messages.
When config structures are copied due to template application,
we need to reset list node structure before calling add_tail().
Thanks to Mikael Magnusson for patches.
Logging as a result of triggered receive limit in import table code
accesset rte->net, which was not filed yet.
Thanks to Pier Carlo Chiodi for the bugreport.
Merge multiple BFD option blocks in BGP configs instead of using the last
one. That is necessary for proper handling of templates when BFD options
are used both in a BGP template and in a BGP protocol derived from that
template.
BFD session options are configured per interface in BFD protocol. This
patch allows to specify them also per-request in protocols requesting
sessions (currently limited to BGP).
Put new non-best routes to the end of list instead of the second
position. Put updated routes to their old position. Position is changed
just by best route selection.
Most commands like 'show ospf neighbors' fail when protocol is not
specified and there are multiple instances of given protocol type.
This is annoying in BIRD 2, as many protocols have IPv4 and IPv6
instances. The patch changes that by showing output from all protocol
instances of appropriate type.
Note that the patch also removes terminating cli_msg() call from these
commands and moves it to the common iterating code.
There is nothing in RFCs specifying that id 0 is not allowed. Some
implementations does not support it, while some other use key id 0 by
default. We allow it but start with key id 1 by default.
Thanks to Kenth Eriksson for the bugreport.
Implement regex-like '+' operator in BGP path masks to match previous
path mask item multiple times. This is useful as ASNs may appear
multiple times in paths due to path prepending for traffic engineering
purposes.
Resolve neighbors using longest prefix match. Although interface ranges
should not generally collide, it may happen for unnumbered links.
Thanks to Kenth Eriksson for the bugreport.
The old code stored route verdicts and temporary routes directly in
rtable. The new code do not store received routes (it immediately
compares them with exported routes and resolves conflicts) and uses
internal bitmap to keep track of which routes were received and which
needs to be reinstalled.
By not putting 'invalid' temporary routes to rtable, we keep rtable
in consistent state, therefore scan no longer needs to be atomic
operation and could be splitted to multiple events.
Channel currently does not have independent pool and uses protocol pool,
which is freed when protocol changes state to down, while channel is
still in flushing. Move some some cleanup code to channel_do_flush()
so it is done before freeing of protocol pool.
Use a hierarchical bitmap in a routing table to assign ids to routes, and
then use bitmaps (indexed by route id) in channels to keep track whether
routes were exported. This avoids unreliable and inefficient re-evaluation
of filters for old routes in order to determine whether they were exported.
Continuation lines may use short form (with space instead of message
number), but this should not be done when previous line is final.
Thanks to Kenth Eriksson for the bugreport and analysis.
Underlying (IGP) route may lead to PtP link, in this case it does not
need gateway. Which is different than direct route without gateway.
When recursive (BGP) route uses PtP route, it should not use recursive
next hop as immediate next hop, while for direct routes it should.
The patch implements optional internal export table to a channel and
hooks it to BGP so it can be used as Adj-RIB-Out. When enabled, all
exported (post-filtered) routes are stored there. An export table can be
examined using e.g. 'show route export table bgp1.ipv4'.
Several BGP channel options (including 'next hop self') could be
reconfigured without session reset, with just route refeed/refresh.
The patch improves reconfiguration code to do it that way.
Protocol can have specified VRF, in such case it is restricted to a set
of ifaces associated with the VRF, otherwise it can use all interfaces.
The patch allows to specify VRF as 'default', in which case it is
restricted to a set of iface not associated with any VRF.
When 'graceful down' command is entered, protocols are shut down
with regard to graceful restart. Namely Kernel protocol does
not remove routes and BGP protocol does not send notification,
just closes the connection.
Support for dynamically spawning BGP protocols for incoming connections.
Use 'neighbor range' to specify range of valid neighbor addresses, then
incoming connections from these addresses spawn new BGP instances.
The temporary atttributes are no longer removed by ea_do_prune(), but
they are undefined by store_tmp_attrs() protocol hooks. This fixes
several bugs where temporary attributes were removed when they should
not or not removed when they should be. The flag EAF_TEMP is no longer
needed and was removed.
Update all protocol make_tmp_attrs() / store_tmp_attrs() hooks to use
helper functions and to handle unset attributes properly.
Also fix some related bugs like improper handling of empty eattr list.
Keep track of whether OSPF tmpattrs are actually defined for given route
(using flags in rte->pflags). That makes them behave more like real
eattrs so a protocol can define just a subset of them or they can be
undefined by filters.
Do not set ospf_metric2 for other than type 2 external OSPF routes and do
not set ospf_tag for non-external OSPF routes. That also fixes a bug
where internal/inter-area route propagated from one OSPF instance to
another is initiated with infinity ospf_metric2.
Thanks to Yaroslav Dronskii for the bugreport.
Route flags are mosty internal state of rtable, they are not significant
to whether a route has changed. With the old code, all routes received as
a part of enhanced route refresh are always re-announced to other peers
due to change in REF_STALE.
This is a major change of how the filters are interpreted. If everything
works how it should, it should not affect you unless you are hacking the
filters themselves.
Anyway, this change should make a huge improvement in the filter performance
as previous benchmarks showed that our major problem lies in the
recursion itself.
There are also some changes in nest and protocols, related mostly to
spreading const declarations throughout the whole BIRD and also to
refactored dynamic attribute definitions. The need of these came up
during the whole work and it is too difficult to split out these
not-so-related changes.
One of previous workarounds for phantom route avoidance breaks export
counters by expanding sending of spurious withdraws, which are send when
we are not sure whether we have advertised that routes in the past.
If not, then export counter is decreased, but it was not increased
before, so it overflows under zero.
The patch fixes that by sendung spurious withdraws, but not counting them
on export counter. That may lead to error in the other direction, but that
happens only as a race condition (i.e., in normal operation filters
return proper values about old route export state).
The earlier fix loosen conditions for not running filters on old
route when deciding about route propagation to a protocol to avoid
issues with ghost routes in some race conditions.
Unfortunately, the fix also caused back-propagation of withdraws. For
regular updates, back-propagation is prevented in import_control hooks,
but these are not called on withdraws. For them, import_control hooks
are called on old routes instead, changing (old, NULL) notification
to (NULL, NULL), which is ignored. By not calling export processing
in some cases, the withdraw is not ignored and is back-propagated.
This patch fixes that by contract conditions so the earlier fix is not
applied to back-propagated updates.
This protocol is highly experimental and nobody should use it in
production. Anyway it may help you getting some insight into what eats
so much time in filter processing.
The patch d506263d... blocked adding channel during reconfiguration,
that broke protocols which use the same functiona also during init.
This patch fixes that.
The patch implements optional internal import table to a channel and
hooks it to BGP so it can be used as Adj-RIB-In. When enabled, all
received (pre-filtered) routes are stored there and import filters can
be re-evaluated without explicit route refresh. An import table can be
examined using e.g. 'show route import table bgp1.ipv4'.
When a new channel is found during reconfiguration, do force restart
of the protocol, like with any other un-reconfigurable change.
The old behavior was that the new channel was added but remained in down
state, even if the protocol was up, so a manual protocol restart was
often necessary.
In the future this should be improved such that a reconfigurable
channel addition (e.g. direct) is accepted and channel is started,
while an un-reconfigurable addition forces protocol restart.
