When a nexthop is resolved via a label based nexthop, copy
the labels into the newly created recursive nexthop.
Please note that this does not fix the case where we
have a label based nexthop that is recursively resolved
through *another* nexthop that is also label based.
In this case we need to create a new label stack
for those routes.
Philippe Guibert [Wed, 10 Jan 2018 13:13:50 +0000 (14:13 +0100)]
lib: ns_id_t changed to 32 bit
Because the VRF_ID is mapped into 32 bit, and because when NETNS will be
the backend of VRF, then the NS identifier must also be encoded as 32
bit.
Also, the NS_UNKNOWN value is changed accordingly to UINT32_MAX.
Also, the NS_UNKNOWN and NS_DEFAULT values are removed from zebra_ns.h
and kept on ns.h header file.
Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
The number of vrf bitmap groups is increased so as to avoid consuming
too much memory. This fix is related to a fork memory that occured when
running pimd as daemon.
A check on memory consumed shows that the memory consumed goes from
33480ko to 46888ko with that change. This is less compared to if the
value of the bitmap groups is increased to 16 ( 852776ko).
Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
Philippe Guibert [Thu, 14 Dec 2017 15:01:36 +0000 (16:01 +0100)]
lib: increase vrf_id from 16 bit to 32 bit identifier
This is a preparatory work for configuring vrf/frr over netns
vrf structure is being changed to 32 bit, and the VRF will have the
possibility to have a backend made up of NETNS.
Let's put some history.
Initially the 32 bit was because one wanted to map on vrf_id both the
VRFLITE and the NSID.
Initially, one would have liked to make zebra configure at the same time
both vrf lite and vrf from netns in a flat way. From the show
running perspective, one would have had both kind of vrfs, thatone
would configure on the same way.
however, it leads to inconsistencies in concepts, because it mixes vrf
vrf with vrf, and vrf is not always mapped with netns.
For instance, logical-router could also be used with netns. In that
case, it would not be possible to map vrf with netns.
There was an other reason why 32 bit is proposed. this is because
some systems handle NSID to 32 bits. As vrf lite exists only on
Linux, there are other systems that would like to use an other vrf
backend than vrf lite. The netns backend for vrf will be used for that
too. for instance, for windows or freebsd, some similar
netns concept exists; so it will be easier to reuse netns
backend for vrf, than reusing vrflite backend for vrf.
This commit is here to extend vrf_id to 32 bits. Following commits in a
second step will help in enable a VRF backend.
Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
vivek [Sat, 20 Jan 2018 21:21:05 +0000 (13:21 -0800)]
zebra: Install connected routes during VRF change only if interface is up
During VRF change handling, the connected route for the interface should be
installed only if the interface is up. Otherwise, we end up with duplicate
connected routes which can lead to other problems.
Donald Sharp [Thu, 18 Jan 2018 14:17:57 +0000 (09:17 -0500)]
zebra: Add some more debug information on read issues in FPM
When we receive a read failure in handling a FPM read
let's add a bit more information to what we think has
gone wrong, in a hope that debugging will be a bit easier.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Quentin Young [Tue, 16 Jan 2018 20:44:14 +0000 (15:44 -0500)]
frr: update README
* Clarify that FRR implements as well as manages protocols
* Move IS-IS out of "early support"
* Add Babel - friends don't leave friends out of READMEs!
* Add mention of further information
Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
Chirag Shah [Sat, 30 Dec 2017 05:01:07 +0000 (21:01 -0800)]
ospfd: show ip ospf interface json output format
Current json output does not differentiate start of
interface objects. Adding "interfaces" keyword at the
beginning of the interface list. This is useful
when displaying vrf level output along with interface list.
Ticket:CM-19115
Testing Done:
show ip ospf vrf all interface json
show ip ospf vrf all interface <specific intf> json
show ip ospf interface json
show ip ospf interface <specific intf> json
Signed-off-by: Chirag Shah <chirag@cumulusnetworks.com>
Quentin Young [Fri, 1 Dec 2017 19:44:32 +0000 (14:44 -0500)]
lib: avoid crash when cancelling invalid rw job
There are some observed instances where we end up trying to cancel a rw
job based on a file descriptor that we don't have a reference on. The
specific cancel function for rw jobs assumes it's called with a file
descriptor that is valid within pollfds and will cause a segmentation
fault by buffer overrun if this is not the case.
Instead log it and move on. Since the fd does not exist this should
patch over the buggy behavior and provide additional information to help
in finding the root cause.
Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
Quentin Young [Fri, 12 Jan 2018 19:57:57 +0000 (14:57 -0500)]
bgpd: disallow invalid config at cli layer
Remove the ability to attempt to configure a couple of options on
directly connected neighbors that don't make sense for them, as well as
the soft error handling code.
Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
Quentin Young [Fri, 12 Jan 2018 17:35:19 +0000 (12:35 -0500)]
vtysh: add ability to output to file
Add ability to set file destination for all vtysh output, with the
exception of tab-complete and similar meta output. This is useful for
inline recording of some information without exiting the shell.
Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
Quentin Young [Thu, 11 Jan 2018 22:07:07 +0000 (17:07 -0500)]
bgpd: preserve admin shutdown on peer-group add
When a peer configured with administrative shutdown is added to a peer
group, the administrative shutdown status is discarded and the peer will
enter the BGP FSM. This is not what we want. Preserve the flag instead.
Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
Donald Sharp [Thu, 11 Jan 2018 00:01:57 +0000 (19:01 -0500)]
zebra: Add one-shot thread to recheck speed
There are certain interfaces that when brought up and we receive
the netlink notification about it, the speed of the interface is
not set correctly. This creates a one-shot thread that will
wait 15 seconds and then requery the speed and if it is different
it will renotify the running daemons.
The kernel should notify us on speed changes, unfortunately this
is not done currently via a netlink message as you would think.
As I understand it there is some in-fighting about the proper
way to approach this issue and due to the way the kernel release
cycle works we are a ways off from getting this fixed. This
is a `hack` to make us work correctly while we wait for the
true answer.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Don Slice [Wed, 10 Jan 2018 19:40:45 +0000 (19:40 +0000)]
bgpd: fix updating redist bitmask when vrf_id changes
Problem reported that when "systemctl restart networking" was
performed, prefixes previously redistributed into bgp from connected
were deleted from the bgp table. Determined that we were not correctly
changing the redistribution bitmask when the vrf_id of the vrf was
changed. This patch corrects that behavior.
Manual tests look good. bgp-min and vrf-min completed with no new failures.
Ticket: CM-19369 Signed-off-by: Don Slice <dslice@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
Daniel Walton [Thu, 31 Aug 2017 16:58:38 +0000 (16:58 +0000)]
bgpd: 'summary' to show NoNeg if peer did not negotiate afi/safi
If we have configured neighbor 1.1.1.1 for an afi/safi but they have not
activated that afi/safi with us then display "NoNeg" in the state column
of the summary output. This is to make troubleshooting afi/safi
easier.
Signed-off-by: Daniel Walton <dwalton@cumulusnetworks.com>
Quentin Young [Wed, 10 Jan 2018 16:20:09 +0000 (11:20 -0500)]
bgpd: move startup sync lock around
Condition needs to be set inside critical section, otherwise i/o thread
can deadlock. Also unlock mutex once finished with it, no need to hold
the lock for the life of the program.
Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
Quentin Young [Mon, 8 Jan 2018 21:40:49 +0000 (16:40 -0500)]
vtysh: optimize printout routine
When printing responses from a client, vtysh searches through every
response it receives twice. Once is to look for the terminating
sequence and the other is to translate \n to \0, which is used with the
line processing callback capability. However:
* There's no need to search all of the output for the terminator, we can
just check the last 4 bytes.
* In most scenarios we won't have a callback. Therefore we don't need to
process the output and can just dump it.
Together these optimizations have a significant impact on the total
runtime of talkative CLI commands, such as `sh ip bgp json` which runs
roughly 1000% faster when dumping 1,000,000 routes.
Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
Daniel Walton [Mon, 8 Jan 2018 21:16:18 +0000 (21:16 +0000)]
ospfd: do not complain if same area is reconfigured
Signed-off-by: Daniel Walton <dwalton@cumulusnetworks.com>
Before
------
cel-redxp-10(config)# router ospf vrf RED
cel-redxp-10(config-router)# network 1.1.1.1/32 area 0.0.0.0
cel-redxp-10(config-router)# network 1.1.1.1/32 area 0.0.0.0
There is already same network statement.
cel-redxp-10(config-router)#
When we see the "There is already same network statement." message
vtysh exits non-zero. This scenario breaks frr-reload because
the command took and it in the config, it should exit zero here.
After
-----
cel-redxp-10(config)# router ospf vrf RED
cel-redxp-10(config-router)# network 1.1.1.1/32 area 0.0.0.0
cel-redxp-10(config-router)# network 1.1.1.1/32 area 0.0.0.0
cel-redxp-10(config-router)# network 1.1.1.1/32 area 0.0.0.0
cel-redxp-10(config-router)# network 1.1.1.1/32 area 0
cel-redxp-10(config-router)#
cel-redxp-10(config-router)# network 1.1.1.1/32 area 0.0.0.1
There is already same network statement.
cel-redxp-10(config-router)#
Donald Sharp [Sat, 6 Jan 2018 19:04:35 +0000 (14:04 -0500)]
bgpd: Ensure that io thread is running after start
The BGP IO thread must be running before other threads
can start using it. So at startup check to see
that it running once, instead of before every
function call into.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
projects / mirror / frr.git / log