RFC 3623 specifies the Graceful Restart enhancement to the OSPF
routing protocol. This PR implements support for the restarting mode,
whereas the helper mode was implemented by #6811.
This work is based on #6782, which implemented the pre-restart part
and settled the foundations for the post-restart part (behavioral
changes, GR exit conditions, and on-exit actions).
Here's a quick summary of how the GR restarting mode works:
* GR can be enabled on a per-instance basis using the `graceful-restart
[grace-period (1-1800)]` command;
* To perform a graceful shutdown, the `graceful-restart prepare ospf`
EXEC-level command needs to be issued before restarting the ospfd
daemon (there's no specific requirement on how the daemon should
be restarted);
* `graceful-restart prepare ospf` will initiate the graceful restart
for all GR-enabled instances by taking the following actions:
o Flooding Grace-LSAs over all interfaces
o Freezing the OSPF routes in the RIB
o Saving the end of the grace period in non-volatile memory (a JSON
file stored in `$frr_statedir`)
* Once ospfd is started again, it will follow the procedures
described in RFC 3623 until it detects it's time to exit the graceful
restart (either successfully or unsuccessfully).
Testing done:
* New topotest featuring a multi-area OSPF topology (including stub
and NSSA areas);
* Successful interop tests against IOS-XR routers acting as helpers.
Co-authored-by: GalaxyGorilla <sascha@netdef.org>
Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
Description:
OSPF does not have an option to control the maximum multiple
equal cost paths to reach a destination/route(ECMP).
Currently, it is using the system specific max multiple paths.
But Somtimes, It requires to control the multiple paths from ospf.
This cli helps to configure the max number multiple paths in ospf.
Signed-off-by: Rajesh Girada <rgirada@vmware.com>
The #if 0 code in ospfd, has not been compiled since at least
2012. If we are at least 9 years old at this point with no effort
to use or save, we should just get rid of it.
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
When P and Q spaces are adjacent then it makes sense to use adjacency SIDs to
from the P node to the Q node. There are some other corner cases where this
makes also sense like when a P/Q node adjacent to root node.
Signed-off-by: GalaxyGorilla <sascha@netdef.org>
A reverse SPF is important in the context of TI-LFA, e.g. for
computing so called Q spaces. In case the weights of the links are
symmetric there is no difference to the 'normal' SPF and hence this
patch is really just needed for the case with asymmetric link
weights.
Signed-off-by: GalaxyGorilla <sascha@netdef.org>
Commit: 1d376ff539 removed
the code to find nexthops for the POINTOMULTIPOINT and
replaced it with a generic bit of code that was
supposed to handle both POINTOPOINT and POINTOMULTIPOINT
the problem is that the ospf rfc states that the
network mask on point to multipoint should be /32
which will not allow you to properly do a prefix match
on it against the network.
Restore original behavior as much as possible and leave
the new POINTOPOINT code alone.
Fixes: #7624
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
Having 2 ABR in NSSA area where R3 is the elected translator
R3# show ip ospf
We are an ABR and the NSSA Elected Translator.
R2# show ip ospf
We are an ABR, but not the NSSA Elected Translator.
When R3 loses the Border condition by shutting down the interface
to the backbone, we end up with no translator in the NSSA area. It
is expected R2 to take over the translator role
R3# sh ip ospf
It is not ABR, therefore not Translator.
R2# show ip ospf
We are an ABR, but not the NSSA Elected Translator.
This PR forces the ABR to reevaluate the translator condition, so
R2 becomes the elected Translator
Signed-off-by: ckishimo <carles.kishimoto@gmail.com>
In the context of TI-LFA it is necessary to have multiple
representations of SPFs for so called P and Q spaces. Hence it makes
sense to start with fresh vertex lists, and only delete them when
the SPF calculation is not a 'dry run'.
Signed-off-by: GalaxyGorilla <sascha@netdef.org>
In the case of P2P links it is possible to use IP unnumbered which
yields a strong dependency to the interface data for nexthop
resolution in the SPF calculations. While the SPF code strives to
be as independent of non-LSA data as possible there is no way
around here: one has to resolve the nexthop for such a special case
using the interface data.
For this purpose a new flag 'spf_root_node' is introduced to signal
that interface data can be used for P2P links. For now this flag is
always 'true' since the SPF currently always uses the calculating
node as the root node. This will change with the introduction of
TI-LFA where other nodes can be root nodes.
Signed-off-by: GalaxyGorilla <sascha@netdef.org>
in OSPF interface data is used for the nexthop resolution
during the SPF algorithm, see RFC2328 16.1.1. However, for
certain technologies like TI-LFA it is desirable to be able
to calculate SPFs for arbitrary root nodes, not just the
calculating node. Since interface data is not available for
other nodes it is necessary to remove this dependency and
make its usage optional, depending on the intent of
changing the RIB with the generated tree (or not).
To signal that a SPF run is used without the intent to
change the RIB an additional flag `spf_dry_run` is
introduced to the ospf_area struct. This flag is currently
only used within the pure SPF code but will be extended
to the SPF postprocessing later on.
Signed-off-by: GalaxyGorilla <sascha@netdef.org>
Just non-functional changes, cosmetics, removal of eye
cancer. The intention here is to make the SPF code more
approachable.
Signed-off-by: GalaxyGorilla <sascha@netdef.org>
Remove mid-string line breaks, cf. workflow doc:
.. [#tool_style_conflicts] For example, lines over 80 characters are allowed
for text strings to make it possible to search the code for them: please
see `Linux kernel style (breaking long lines and strings)
<https://www.kernel.org/doc/html/v4.10/process/coding-style.html#breaking-long-lines-and-strings>`_
and `Issue #1794 <https://github.com/FRRouting/frr/issues/1794>`_.
Scripted commit, idempotent to running:
```
python3 tools/stringmangle.py --unwrap `git ls-files | egrep '\.[ch]$'`
```
Signed-off-by: David Lamparter <equinox@diac24.net>
* Change sr_prefix structure in ospf_sr.h to add support to ECMP
* Add new Segment Routing information to ospf_paths in ospf_route.h
* Backport MPLS label configuration from IS-IS Segment Routing implementation
* Re-write log message in ospf_sr.c and ospf_ext.c
Signed-off-by: Olivier Dugeon <olivier.dugeon@orange.com>
Some logging systems are, er, "allergic" to tabs in log messages.
(RFC5424: "The syslog application SHOULD avoid octet values below 32")
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
The order of ECMP nexthops currently depends on whatever order the
pqueue code returns the vertices in, which is essentially random since
they compare as equal. While this shouldn't cause issues normally, it
is nondeterministic and causes the ldp-topo1 test to fail when the
ordering comes up different. Also, nondeterministic behaviour is not a
nice thing to have here in general.
Just sort by nexthop address; realistic numbers of ECMP nexthops should
hopefully not make this a performance issue. (Also, nexthops should be
hot in the caches here.)
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
There is no need to check for failure of a ALLOC call
as that any failure to do so will result in a assert
happening. So we can safely remove all of this code.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
The following types are nonstandard:
- u_char
- u_short
- u_int
- u_long
- u_int8_t
- u_int16_t
- u_int32_t
Replace them with the C99 standard types:
- uint8_t
- unsigned short
- unsigned int
- unsigned long
- uint8_t
- uint16_t
- uint32_t
Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
This is an implementation of draft-ietf-ospf-segment-routing-extensions-24
and RFC7684 for Extended Link & Prefix Opaque LSA.
Look to doc/OSPF_SR.rst for implementation details & known limitations.
New files:
- ospfd/ospf_sr.h: Segment Routing structure definition (SubTLVs + SRDB)
- ospfd/ospf_sr.c: Main functions for Segment Routing support
- ospfd/ospf_ext.h: TLVs and SubTLVs definition for RFC7684
- ospfd/ospf_ext.c: RFC7684 Extended Link / Prefix implementation
- doc/OSPF-SRr.rst: Documentation
Modified Files:
- doc/ospfd.texi: Add new Segment Routing CLI command definition
- lib/command.h: Add new string command for Segment Routing CLI
- lib/mpls.h: Add default value for SRGB
- lib/route_types.txt: Add new OSPF Segment Routing route type
- ospfd/ospf_dump.[c,h]: Add OSPF SR debug
- ospfd/ospf_memory.[c,h]: Add new Segment Routing memory type
- ospfd/ospf_opaque.[c,h]: Add ospf_sr_init() starting function
- ospfd/ospf_ri.c: Add new functions to Set/Get Segment Routing TLVs
Add new ospf_router_info_lsa_upadte() to send Opaque LSA to ospf_sr.c()
- ospfd/ospf_ri.h: Add new Router Information SR SubTLVs
- ospfd/ospf_spf.c: Add new scheduler when running SPF to trigger
update of NHLFE
- ospfd/ospfd.h: Add new thread for Segment Routing scheduler
- ospfd/subdir.am: Add new files
- vtysh/Makefile.am: Add new ospf_sr.c file for vtysh
- zebra/kernel_netlink.c: Add new OSPF_SR route type
- zebra/rt_netlink.[c,h]: Add new OSPF_SR route type
- zebra/zebra_mpls.h: Add new OSPF_SR route type
Signed-off-by: Olivier Dugeon <olivier.dugeon@orange.com>
Convert the list_delete(struct list *) function to use
struct list **. This is to allow the list pointer to be nulled.
I keep running into uses of this list_delete function where we
forget to set the returned pointer to NULL and attempt to use
it and then experience a crash, usually after the developer
has long since left the building.
Let's make the api explicit in it setting the list pointer
to null.
Cynical Prediction: This code will expose a attempt
to use the NULL'ed list pointer in some obscure bit
of code.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Fix ANVL-OSPF-5.1 reported ospfd crash.
vertex_nexthop_free was added as valgrind reported potential
memory leak, but in some cases nexthop would not be available
freed.
The actual nexthop free is part of ospf_canonical_nexthops_free(),
upon trying to free, qfree checks mtype count becomes 0 and asserts.
Removing vertex_nexthop_free() from ospf_spf_flush_parents().
Signed-off-by: Chirag Shah <chirag@cumulusnetworks.com>
Remove assert in path of router_id_update:
Upon configuring same router-id as neighbor's
assert would cause a crash. Log a warning message
and neighborship would not come up.
Address memory leaks
Signed-off-by: Chirag Shah <chirag@cumulusnetworks.com>
The FSF's address changed, and we had a mixture of comment styles for
the GPL file header. (The style with * at the beginning won out with
580 to 141 in existing files.)
Note: I've intentionally left intact other "variations" of the copyright
header, e.g. whether it says "Zebra", "Quagga", "FRR", or nothing.
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
