]> git.proxmox.com Git - mirror_frr.git/log
mirror_frr.git
4 years agopimd: Use PIM EVPN MLAG Infra for syncing PIM MLAG Entries
Satheesh Kumar K [Mon, 19 Aug 2019 09:06:00 +0000 (02:06 -0700)]
pimd: Use PIM EVPN MLAG Infra for syncing PIM MLAG Entries

Initially, MLAG Sync is happened at pim_ifchannel, this is mainly to
support even config mismatches(missing configuration of dual active).
But this causes more syncs for each entry.

and also it is not In-line with PIM EVPN. to avoid that moving to
pm_upstream based syncing.

Signed-off-by: Satheesh Kumar K <sathk@cumulusnetworks.com>
4 years agopimd: run DF election only on (*, G) termination mroutes
Anuradha Karuppiah [Wed, 10 Jul 2019 15:00:04 +0000 (08:00 -0700)]
pimd: run DF election only on (*, G) termination mroutes

(S,G) entries that inherit ipmr-lo into the OIL also inherit
the DF role from the parent (*, G) entry.

This change is done primarily to simplify the sync process and
to prevent the MLAG peers from having to track (S, G) activity etc.

Signed-off-by: Anuradha Karuppiah <anuradhak@cumulusnetworks.com>
4 years agopimd: Skip nexthop lookup for register source in some cases
Donald Sharp [Fri, 22 Nov 2019 21:16:26 +0000 (16:16 -0500)]
pimd: Skip nexthop lookup for register source in some cases

There exists the possibility that a RP exists as a anycast
pair for a lan segment.  As such one side may receive
the register and properly handle the registration mechanics.
The one that does not receive the register packets will still
get S,G state and WRVIFWHOLE upcalls across the lan.  In
this case notice that we have not received the Registration
packets and prevent nexthop lookups.

Ticket: CM-27466
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
4 years agopimd: Add accidently missed code during upstreaming process
Donald Sharp [Wed, 19 Feb 2020 14:52:17 +0000 (09:52 -0500)]
pimd: Add accidently missed code during upstreaming process

There was some code missed during the upstreaming process
due to code squash.  Identify and put into a commit
to keep code consistent and correct.

Signed-off-by: Satheesh Kumar K <sathk@cumulusnetworks.com>
Signed-off-by: Anuradha Karuppiah <anuradhak@cumulusnetworks.com>
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
4 years agopimd: There is nothing to do with a WRVIFWHOLE for iifp of pimreg
Donald Sharp [Fri, 20 Dec 2019 12:57:28 +0000 (07:57 -0500)]
pimd: There is nothing to do with a WRVIFWHOLE for iifp of pimreg

When the WRVIFWHOLE callback is made with a iifp of the pimreg
device we *know* that the packet is a PIM Register packet
( see net/ipv4/ipmr.c for kernel behavior ).  As such
we know that we will shortly read the pim register packet
and handle it through those mechanics.  There is nothing
to do here so we can move along.

Ticket: CM-27729
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
4 years agopimd: Put subdir.am into alphabetical order
Donald Sharp [Wed, 19 Feb 2020 14:57:43 +0000 (09:57 -0500)]
pimd: Put subdir.am into alphabetical order

The subdir.am had lost alphabetical order.  Put this back
for easy finding in the future.

Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
4 years agoMerge pull request #5924 from donaldsharp/more_func
Renato Westphal [Fri, 6 Mar 2020 19:16:02 +0000 (16:16 -0300)]
Merge pull request #5924 from donaldsharp/more_func

*: Finish off the __PRETTY_FUNCTION__ to __func__

4 years ago*: Finish off the __PRETTY_FUNCTION__ to __func__
Donald Sharp [Fri, 6 Mar 2020 14:23:22 +0000 (09:23 -0500)]
*: Finish off the __PRETTY_FUNCTION__ to __func__

FINISH IT

Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
4 years agoMerge pull request #5920 from qlyoung/fix-srv6-repeated-attr-memleak
Donatas Abraitis [Fri, 6 Mar 2020 13:29:11 +0000 (15:29 +0200)]
Merge pull request #5920 from qlyoung/fix-srv6-repeated-attr-memleak

Fix srv6 repeated attr memleak + uninitialized refcnt

4 years agoMerge pull request #5916 from mjstapp/fix_gmtime
Quentin Young [Thu, 5 Mar 2020 22:38:13 +0000 (17:38 -0500)]
Merge pull request #5916 from mjstapp/fix_gmtime

*: use gmtime_r exclusively

4 years agoMerge pull request #5918 from ton31337/fix/__func__everywhere
Quentin Young [Thu, 5 Mar 2020 22:37:03 +0000 (17:37 -0500)]
Merge pull request #5918 from ton31337/fix/__func__everywhere

__func__ everywhere

4 years agobgpd: properly initialize SRv6 attributes
Quentin Young [Thu, 5 Mar 2020 21:15:32 +0000 (16:15 -0500)]
bgpd: properly initialize SRv6 attributes

We were using XMALLOC for these, and only initializing the refcount to 0
on one of them. Let's just use XCALLOC instead...

Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
4 years agobgpd: correctly handle repeated SRv6 attributes
Quentin Young [Thu, 5 Mar 2020 21:13:17 +0000 (16:13 -0500)]
bgpd: correctly handle repeated SRv6 attributes

Repeating SRv6 attributes N times leaks (N-1)*32 bytes of memory.

Fix it!

Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
4 years ago*: use gmtime_r, localtime_r exclusively
Mark Stapp [Thu, 5 Mar 2020 16:42:12 +0000 (11:42 -0500)]
*: use gmtime_r, localtime_r exclusively

Stop using gmtime() or localtime() everywhere.

Signed-off-by: Mark Stapp <mjs@voltanet.io>
4 years ago*: Replace __PRETTY_FUNCTION__/__FUNCTION__ to __func__
Donatas Abraitis [Thu, 5 Mar 2020 18:17:54 +0000 (20:17 +0200)]
*: Replace __PRETTY_FUNCTION__/__FUNCTION__ to __func__

Just keep the code cool.

Signed-off-by: Donatas Abraitis <donatas.abraitis@gmail.com>
4 years agoMerge pull request #5856 from pguibert6WIND/nhrp_override_fix
Mark Stapp [Thu, 5 Mar 2020 18:20:53 +0000 (13:20 -0500)]
Merge pull request #5856 from pguibert6WIND/nhrp_override_fix

zebra: when FIB_OVERRIDE flag is set, update nexthop-tracking clients properly

4 years agotools: Add Coccinelle script to replace __FUNCTION__ to __func__
Donatas Abraitis [Thu, 5 Mar 2020 18:16:47 +0000 (20:16 +0200)]
tools: Add Coccinelle script to replace __FUNCTION__ to __func__

Same is for __PRETTY_FUNCTION__

Signed-off-by: Donatas Abraitis <donatas.abraitis@gmail.com>
4 years agoMerge pull request #5911 from ton31337/fix/no_check_if_assert
Quentin Young [Thu, 5 Mar 2020 18:16:30 +0000 (13:16 -0500)]
Merge pull request #5911 from ton31337/fix/no_check_if_assert

 bgpd: Remove a test and return statement after assert

4 years agoMerge pull request #5914 from mjstapp/fix_sa_show_all
Donald Sharp [Thu, 5 Mar 2020 17:38:49 +0000 (12:38 -0500)]
Merge pull request #5914 from mjstapp/fix_sa_show_all

zebra: clean up an SA warning in show_ip_route_all

4 years agozebra: when override flag is set, do not forget to update nht
Philippe Guibert [Wed, 19 Feb 2020 16:27:06 +0000 (17:27 +0100)]
zebra: when override flag is set, do not forget to update nht

this flag can be used when one routing daemon wants to force his route
to be injected prioritary with other routes, including selected routes.
for that, do not forget to update the new_selected pointer in the zebra
nexthop tracking algorithm.

Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
4 years agonhrpd: ignore zebra updates about our routes being deleted/added
Philippe Guibert [Thu, 5 Mar 2020 14:50:37 +0000 (14:50 +0000)]
nhrpd: ignore zebra updates about our routes being deleted/added

nhrp listens for route entries to be deleted, in case some new routes
impact the current routes installed by nhrp. To prevent from
unconfiguring nhrp shortcut route, just prevent nhrp routes to be
processed.

Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
4 years agozebra: clean up an SA warning in show_ip_route_all
Mark Stapp [Thu, 5 Mar 2020 13:35:55 +0000 (08:35 -0500)]
zebra: clean up an SA warning in show_ip_route_all

SA reports a redundant NULL check; remove it.

Signed-off-by: Mark Stapp <mjs@voltanet.io>
4 years agoMerge pull request #5850 from Spantik/gr_fix
Quentin Young [Thu, 5 Mar 2020 04:40:26 +0000 (23:40 -0500)]
Merge pull request #5850 from Spantik/gr_fix

Zebra: Zebra gr dynamic client handling

4 years agoMerge pull request #5907 from ton31337/fix/replace_sizeof_coccinelle
Quentin Young [Thu, 5 Mar 2020 00:05:05 +0000 (19:05 -0500)]
Merge pull request #5907 from ton31337/fix/replace_sizeof_coccinelle

Bool related stuff

4 years agoMerge pull request #5912 from mjstapp/fix_doc_slack_link
Quentin Young [Thu, 5 Mar 2020 00:04:09 +0000 (19:04 -0500)]
Merge pull request #5912 from mjstapp/fix_doc_slack_link

doc: change slack reference link to main FRR homepage

4 years agoMerge pull request #5910 from donaldsharp/ospf_indentation
Mark Stapp [Wed, 4 Mar 2020 20:24:56 +0000 (15:24 -0500)]
Merge pull request #5910 from donaldsharp/ospf_indentation

ospfd: Fix 4 levels of unnecessary indentation

4 years agodoc: change slack reference link to main FRR homepage
Mark Stapp [Wed, 4 Mar 2020 19:13:21 +0000 (14:13 -0500)]
doc: change slack reference link to main FRR homepage

Change the overview page's link for slack info to point to
the main FRR homepage section - that's where the self-serve
link/info is.

Signed-off-by: Mark Stapp <mjs@voltanet.io>
4 years agobgpd: Remove a test and return statement after assert
Donatas Abraitis [Wed, 4 Mar 2020 18:19:12 +0000 (20:19 +0200)]
bgpd: Remove a test and return statement after assert

No need to check if the variable is NULL and return after assert.

Signed-off-by: Donatas Abraitis <donatas.abraitis@gmail.com>
4 years agotools: Add coccinelle script to remove a test case after assert()
Donatas Abraitis [Wed, 4 Mar 2020 18:17:42 +0000 (20:17 +0200)]
tools: Add coccinelle script to remove a test case after assert()

Signed-off-by: Donatas Abraitis <donatas.abraitis@gmail.com>
4 years agoospfd: Fix 4 levels of unnecessary indentation
Donald Sharp [Wed, 4 Mar 2020 17:50:57 +0000 (12:50 -0500)]
ospfd: Fix 4 levels of unnecessary indentation

Optimize display of code a tiny bit so that we can actually
have readable code.

Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
4 years agoMerge pull request #5908 from rubenk/zebra-fix-typo-in-debug-message
Quentin Young [Wed, 4 Mar 2020 16:59:49 +0000 (11:59 -0500)]
Merge pull request #5908 from rubenk/zebra-fix-typo-in-debug-message

zebra: fix typo in debug log message

4 years ago*: Use true/false instead of 1/0 when assigning variable for bool type
Donatas Abraitis [Wed, 4 Mar 2020 16:49:36 +0000 (18:49 +0200)]
*: Use true/false instead of 1/0 when assigning variable for bool type

Signed-off-by: Donatas Abraitis <donatas.abraitis@gmail.com>
4 years agotools: Coccinelle script to convert 0/1 to false/true if variable type is bool
Donatas Abraitis [Wed, 4 Mar 2020 16:47:09 +0000 (18:47 +0200)]
tools: Coccinelle script to convert 0/1 to false/true if variable type is bool

Signed-off-by: Donatas Abraitis <donatas.abraitis@gmail.com>
4 years agoMerge pull request #5906 from ton31337/feature/documentation_for_rpki
Quentin Young [Wed, 4 Mar 2020 16:39:37 +0000 (11:39 -0500)]
Merge pull request #5906 from ton31337/feature/documentation_for_rpki

doc: Add documentation for `show rpki prefix` and `show rpki as-number`

4 years agopimd: Convert pim_upstream_evaluate_join_desired type to bool
Donatas Abraitis [Wed, 4 Mar 2020 15:05:12 +0000 (17:05 +0200)]
pimd: Convert pim_upstream_evaluate_join_desired type to bool

Signed-off-by: Donatas Abraitis <donatas.abraitis@gmail.com>
4 years agozebra: fix typo in debug log message
Ruben Kerkhof [Wed, 4 Mar 2020 15:08:18 +0000 (16:08 +0100)]
zebra: fix typo in debug log message

Signed-off-by: Ruben Kerkhof <ruben@rubenkerkhof.com>
4 years agotools: Add coccinelle script to convert int to bool if returning value is bool
Donatas Abraitis [Wed, 4 Mar 2020 13:29:22 +0000 (15:29 +0200)]
tools: Add coccinelle script to convert int to bool if returning value is bool

Signed-off-by: Donatas Abraitis <donatas.abraitis@gmail.com>
4 years ago*: Return bool type for bool functions
Donatas Abraitis [Wed, 4 Mar 2020 13:05:22 +0000 (15:05 +0200)]
*: Return bool type for bool functions

Signed-off-by: Donatas Abraitis <donatas.abraitis@gmail.com>
4 years ago*: Use short version of bool expressions
Donatas Abraitis [Wed, 4 Mar 2020 13:04:23 +0000 (15:04 +0200)]
*: Use short version of bool expressions

Signed-off-by: Donatas Abraitis <donatas.abraitis@gmail.com>
4 years agotools: Add coccinelle script to check if bool functions return bool type
Donatas Abraitis [Wed, 4 Mar 2020 13:02:23 +0000 (15:02 +0200)]
tools: Add coccinelle script to check if bool functions return bool type

Signed-off-by: Donatas Abraitis <donatas.abraitis@gmail.com>
4 years agotools: Add coccinelle script to check if bool expressions use short format
Donatas Abraitis [Wed, 4 Mar 2020 13:00:51 +0000 (15:00 +0200)]
tools: Add coccinelle script to check if bool expressions use short format

Signed-off-by: Donatas Abraitis <donatas.abraitis@gmail.com>
4 years agodoc: Add documentation for `show rpki prefix` and `show rpki as-number`
Donatas Abraitis [Wed, 4 Mar 2020 09:25:57 +0000 (11:25 +0200)]
doc: Add documentation for `show rpki prefix` and `show rpki as-number`

Signed-off-by: Donatas Abraitis <donatas.abraitis@gmail.com>
4 years agoMerge pull request #5894 from patrasar/2399927
Donald Sharp [Wed, 4 Mar 2020 02:20:25 +0000 (21:20 -0500)]
Merge pull request #5894 from patrasar/2399927

pimd: Update pim register stop statistics

4 years agoMerge pull request #5896 from patrasar/2409015
Donald Sharp [Wed, 4 Mar 2020 02:13:27 +0000 (21:13 -0500)]
Merge pull request #5896 from patrasar/2409015

pimd: neighbor needed true in pim nexthop lookup

4 years agoMerge pull request #5899 from Spantik/clear_bgp_bug
Donald Sharp [Tue, 3 Mar 2020 18:16:01 +0000 (13:16 -0500)]
Merge pull request #5899 from Spantik/clear_bgp_bug

BGPD: Cores when cleared with doppelganger.

4 years agoMerge pull request #5900 from rubenk/bgpd-fix-typo
Donald Sharp [Tue, 3 Mar 2020 18:11:28 +0000 (13:11 -0500)]
Merge pull request #5900 from rubenk/bgpd-fix-typo

bgpd: fix typo in 'show bgp neighbors' output

4 years agoMerge pull request #5886 from sworleys/Trust-Kernel-Direct-Routes
Mark Stapp [Tue, 3 Mar 2020 17:10:35 +0000 (12:10 -0500)]
Merge pull request #5886 from sworleys/Trust-Kernel-Direct-Routes

zebra: trust directly connected kernel/system routes

4 years agoMerge pull request #5873 from mjstapp/zebra_lsp_dtor
Donald Sharp [Tue, 3 Mar 2020 16:48:16 +0000 (11:48 -0500)]
Merge pull request #5873 from mjstapp/zebra_lsp_dtor

zebra: add a free api for LSPs

4 years agobgpd: fix typo in 'show bgp neighbors' output
Ruben Kerkhof [Tue, 3 Mar 2020 15:46:31 +0000 (16:46 +0100)]
bgpd: fix typo in 'show bgp neighbors' output

And fix a typo in a comment while we're at it

Signed-off-by: Ruben Kerkhof <ruben@rubenkerkhof.com>
4 years agoMerge pull request #5890 from mjstapp/zapi_allow_label_num
Russ White [Tue, 3 Mar 2020 14:36:45 +0000 (09:36 -0500)]
Merge pull request #5890 from mjstapp/zapi_allow_label_num

zebra: allow multiple labels in LSP zapi message

4 years agoMerge pull request #5897 from patrasar/2415676
Russ White [Tue, 3 Mar 2020 14:35:43 +0000 (09:35 -0500)]
Merge pull request #5897 from patrasar/2415676

pimd: fix crash seen while executing igmp related configuration

4 years agoBGPD: Cores when cleared with doppelganger.
Santosh P K [Tue, 3 Mar 2020 14:27:30 +0000 (06:27 -0800)]
BGPD: Cores when cleared with doppelganger.

RCA: When doppelganger still around and clear bgp is issued
there are chances of peer getting deleted and next pointer
is a  freed peer pointer.

Fix: Pass address of nnode to get next safe peer pointer.

Signed-off-by: Santosh P K <sapk@vmware.com>
4 years agoMerge pull request #5893 from donaldsharp/clang_mc_clang
Renato Westphal [Tue, 3 Mar 2020 11:53:33 +0000 (08:53 -0300)]
Merge pull request #5893 from donaldsharp/clang_mc_clang

ldpd: Remove double set of lif

4 years agopimd: fix crash seen while executing igmp related configuration
Sarita Patra [Tue, 3 Mar 2020 11:35:43 +0000 (03:35 -0800)]
pimd: fix crash seen while executing igmp related configuration

Issue 1:
1. Enable pim on an interface.
2. Configure query-interval or query max response time,
which results in pimd crash.

Root cause:
1. When pim is enabled on an interface, it creates a igmp socket
with querier_timer and other_querier time as NULL.
2. When query-interval/max_response_time is configured, it call the
function igmp_sock_query_reschedule() to reshedule the query. This
function check either of querier_timer or other_querier timer should
be running. Since in this case both are NULL, it results in crash.

Issue 2:
1. Enable pim on an interface.
2. Execute no ip igmp query-interval or query max response time,
which results in pimd crash.

Root cause:
1. When pim is enabled on an interface, it creates a pim interface
with querier_timer and other_querier time as NULL.
2. When no ip igmp query-interval/max_response_time is executed, it will
check either of querier_timer or other_querier timer should be running.
Since in this case both are NULL, it results in crash.

Fix:
When pim is enabled on an interface, it creates a igmp socket with
mtrace_only as true. So add a check if mtrace_only is true, then don't
reshedule the query.

Signed-off-by: Sarita Patra <saritap@vmware.com>
4 years agopimd: neighbor needed true in pim nexthop lookup
Sarita Patra [Tue, 3 Mar 2020 11:31:22 +0000 (03:31 -0800)]
pimd: neighbor needed true in pim nexthop lookup

Issue:
Client---LHR---RP
1. Add kernel route for RP on LHR. Client send join
2. (*,G) will be get created in LHR and RP.
3. Kill the FRR on all the nodes
4. Start FRR only on LHR node
5. In LHR, (*, G) will be created with iif as unknown.

Root cause:
In the step 4, When LHR will receive igmp join, it will call
the function pim_ecmp_fib_lookup_if_vif_index which will look
for nexthop to RP with neighbor needed as false. So RPF lookup will
be true as the route is present in the kernel. It will create a
(*, G) channel_oil with incoming interface as the RPF interface
towards RP and install the (*,G) mroute in kernel.
Along with this (*,G) upstream gets craeted, which call the function
pim_rpf_update, which will look for the nexthop to RP with neighbor
needed as true. As the frr is not running in RP, no neighbor is present
on the nexthop interface. Due to which this will fail and will update
the channel_oil incoming interface as MAXVIFS(32).

Fix:
pim_ecmp_fib_lookup_if_vif_index() call the function pim_ecmp_nexthop_lookup
with neighbor_needed as true.

Signed-off-by: Sarita Patra <saritap@vmware.com>
4 years agopimd: Update pim register stop statistics
Sarita Patra [Tue, 3 Mar 2020 09:46:36 +0000 (01:46 -0800)]
pimd: Update pim register stop statistics

Issue: REGISTER-STOP Rx is always displaying 0.

Root-cause: pim_ifstat_reg_stop_recv is not getting
incremented when register stop message is received.

Fix: Increment pim_ifstat_reg_stop_recv on receiving
of pim register stop packet.

Signed-off-by: Sarita Patra <saritap@vmware.com>
4 years agoldpd: Remove double set of lif
Donald Sharp [Tue, 3 Mar 2020 01:29:01 +0000 (20:29 -0500)]
ldpd: Remove double set of lif

The lif variable was being set in the if statement and
immediately copied into from xf.  No need to do this
twice.

Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
4 years agozebra: allow multiple labels in LSP zapi message
Mark Stapp [Mon, 2 Mar 2020 20:51:22 +0000 (15:51 -0500)]
zebra: allow multiple labels in LSP zapi message

The handlers for a couple of the main LSP-oriented zapi
messages explicitly limited themselves to a single out-label.
Allow multiple labels if the sender ... sends them.

Signed-off-by: Mark Stapp <mjs@voltanet.io>
4 years agozebra: trust directly connected kernel/system routes
Stephen Worley [Mon, 2 Mar 2020 18:16:02 +0000 (13:16 -0500)]
zebra: trust directly connected kernel/system routes

We made the decision to explicitly trust kernel and system routes
of every other type with 058c16b7e239f1c50a1d4b4376a6aff6b8dad959.

So, we should trust directly connected routes the same way, assuming
the interface exists.

Old Behavior:

K   2.2.2.1/32 [0/0] is directly connected, unknown inactive, 00:00:39

New Behavior:

K>* 2.2.2.1/32 [0/0] is directly connected, test1, 00:00:03

As a bonus, this fixes the issues we were seeing with not removing
directly connected routes of certain interface types when
those interfaces go down/are deleted.

Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
4 years agoMerge pull request #5855 from ton31337/fix/allow_using_add_sub_for_local-preference
Sri Mohana Singamsetty [Mon, 2 Mar 2020 17:40:34 +0000 (09:40 -0800)]
Merge pull request #5855 from ton31337/fix/allow_using_add_sub_for_local-preference

bgpd: Allow using add/subtract for local-preference in route-maps

4 years agoMerge pull request #5780 from chiragshah6/evpn_dev1
Sri Mohana Singamsetty [Mon, 2 Mar 2020 17:39:37 +0000 (09:39 -0800)]
Merge pull request #5780 from chiragshah6/evpn_dev1

zebra:remote vtep delete handling

4 years agozebra: add a free api for LSPs
Mark Stapp [Thu, 27 Feb 2020 17:18:48 +0000 (12:18 -0500)]
zebra: add a free api for LSPs

consolidate some LSP cleanup code into a common free function.

Signed-off-by: Mark Stapp <mjs@voltanet.io>
4 years agoMerge pull request #5874 from donaldsharp/hash_backet2
Mark Stapp [Fri, 28 Feb 2020 21:26:55 +0000 (16:26 -0500)]
Merge pull request #5874 from donaldsharp/hash_backet2

*: change hash_backet to hash_bucket

4 years agoMerge pull request #5822 from rbauduin/patch-2
Donatas Abraitis [Fri, 28 Feb 2020 20:55:15 +0000 (22:55 +0200)]
Merge pull request #5822 from rbauduin/patch-2

doc: Document attribute-unchanged directive

4 years agoMerge pull request #5878 from mjstapp/fix_nhg_for_fpm
Stephen Worley [Fri, 28 Feb 2020 20:25:40 +0000 (15:25 -0500)]
Merge pull request #5878 from mjstapp/fix_nhg_for_fpm

zebra: fix nexthop_group conversion in fpm code

4 years ago*: change hash_backet to hash_bucket
Donald Sharp [Thu, 27 Feb 2020 01:03:34 +0000 (20:03 -0500)]
*: change hash_backet to hash_bucket

It's been a year search and destroy.

Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
4 years agoMerge pull request #5862 from sworleys/PBR-No-Table-Range
Donatas Abraitis [Fri, 28 Feb 2020 16:28:51 +0000 (18:28 +0200)]
Merge pull request #5862 from sworleys/PBR-No-Table-Range

pbrd: unconfigure table range

4 years agoMerge pull request #5876 from ton31337/feature/doc_vrrpd_update
Mark Stapp [Fri, 28 Feb 2020 15:57:11 +0000 (10:57 -0500)]
Merge pull request #5876 from ton31337/feature/doc_vrrpd_update

doc: Change interface name for vrrp example

4 years agozebra: fix nexthop_group conversion in fpm code
Mark Stapp [Fri, 28 Feb 2020 13:45:56 +0000 (08:45 -0500)]
zebra: fix nexthop_group conversion in fpm code

Recent commit that embedded the nhg_hash_entry's group
missed a couple of fpm modules.

Signed-off-by: Mark Stapp <mjs@voltanet.io>
4 years agoMerge pull request #5859 from donaldsharp/clang_latest
Mark Stapp [Fri, 28 Feb 2020 12:51:52 +0000 (07:51 -0500)]
Merge pull request #5859 from donaldsharp/clang_latest

Clean up clang latest SA warnings

4 years agotests: Add a test case for `set local-preference +/-LOCAL_PREF`
Donatas Abraitis [Sun, 23 Feb 2020 18:59:54 +0000 (20:59 +0200)]
tests: Add a test case for `set local-preference +/-LOCAL_PREF`

Signed-off-by: Donatas Abraitis <donatas.abraitis@gmail.com>
4 years agodoc: Use +/- for `set local-preference` to add/subtract the value
Donatas Abraitis [Sun, 23 Feb 2020 18:55:14 +0000 (20:55 +0200)]
doc: Use +/- for `set local-preference` to add/subtract the value

Signed-off-by: Donatas Abraitis <donatas.abraitis@gmail.com>
4 years agobgpd: Allow using add/subtract for local-preference in route-maps
Donatas Abraitis [Sat, 22 Feb 2020 21:04:21 +0000 (23:04 +0200)]
bgpd: Allow using add/subtract for local-preference in route-maps

We already have a generic support for add/sub in route-maps. It's already
handled in route_value_compile().

Just convert to string (allow passing (-) minus sign) - works like expected.

Signed-off-by: Donatas Abraitis <donatas.abraitis@gmail.com>
4 years agodoc: Change interface name for vrrp example
Donatas Abraitis [Fri, 28 Feb 2020 11:58:38 +0000 (13:58 +0200)]
doc: Change interface name for vrrp example

Looks like copy-paste error.

Signed-off-by: Donatas Abraitis <donatas.abraitis@gmail.com>
4 years agoMerge pull request #5875 from donaldsharp/support_rfcs
Donatas Abraitis [Fri, 28 Feb 2020 07:23:44 +0000 (09:23 +0200)]
Merge pull request #5875 from donaldsharp/support_rfcs

doc: Add notice of rfc-4893 in supported rfc's.

4 years agoMerge pull request #5857 from mjstapp/embed_nhg_in_nhe
Stephen Worley [Thu, 27 Feb 2020 23:43:59 +0000 (18:43 -0500)]
Merge pull request #5857 from mjstapp/embed_nhg_in_nhe

zebra,lib: Embed lib nexthop-group in zebra hash entry

4 years agodoc: Add notice of rfc-4893 in supported rfc's.
Donald Sharp [Thu, 27 Feb 2020 22:51:30 +0000 (17:51 -0500)]
doc: Add notice of rfc-4893 in supported rfc's.

Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
4 years agozebra: Embed lib nexthop-group in zebra hash entry
Mark Stapp [Tue, 25 Feb 2020 13:29:46 +0000 (08:29 -0500)]
zebra: Embed lib nexthop-group in zebra hash entry

Embed nexthop-group, which is just a pointer, in the zebra
nexthop-hash-entry object, rather than mallocing one.

Signed-off-by: Mark Stapp <mjs@voltanet.io>
4 years agoMerge pull request #5872 from Jafaral/backet
Donald Sharp [Thu, 27 Feb 2020 15:56:27 +0000 (10:56 -0500)]
Merge pull request #5872 from Jafaral/backet

Revert "*: change hash_backet to hash_bucket"

4 years agoRevert "*: change hash_backet to hash_bucket"
Jafar Al-Gharaibeh [Thu, 27 Feb 2020 15:47:02 +0000 (09:47 -0600)]
Revert "*: change hash_backet to hash_bucket"

This reverts commit 3895c42a2eb534a70ef25412547e798c1c1ef5a1.

LabN CI update needs to be coordinated before merging this

Signed-off-by: Jafar Al-Gharaibeh <jafar@atcorp.com>
4 years agoldpd: Cleanup indentation in merge_nbrps
Donald Sharp [Mon, 24 Feb 2020 14:12:17 +0000 (09:12 -0500)]
ldpd: Cleanup indentation in merge_nbrps

We had a very deeply nested function, reduce the indentation
for easier reading.

Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
4 years agobgpd: Cleanup indentation in bgp_route_refresh_receive
Donald Sharp [Mon, 24 Feb 2020 14:08:54 +0000 (09:08 -0500)]
bgpd: Cleanup indentation in bgp_route_refresh_receive

Some code in bgp_route_refresh_receive was spread across several
lines because of an end of line commit.  Move comment to a place
to allow better formating.

Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
4 years agoMerge pull request #5870 from donaldsharp/hash_backet
Jafar Al-Gharaibeh [Thu, 27 Feb 2020 15:02:35 +0000 (09:02 -0600)]
Merge pull request #5870 from donaldsharp/hash_backet

*: change hash_backet to hash_bucket

4 years agobgpd: agg_node_lookup not unlocking
Donald Sharp [Mon, 24 Feb 2020 13:45:36 +0000 (08:45 -0500)]
bgpd: agg_node_lookup not unlocking

SA has found a case where we did a table lookup of a rn( and
associated lock of that node ) where we did not unlock it.

Unlock the node before moving on to the next one.

Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
4 years agonhrpd: Cleanup set but unused variables
Donald Sharp [Mon, 24 Feb 2020 13:47:46 +0000 (08:47 -0500)]
nhrpd: Cleanup set but unused variables

There existed some variables set but never used.  Clean this up.

Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
4 years agoospfd: Cleanup set but unused variables
Donald Sharp [Mon, 24 Feb 2020 13:37:34 +0000 (08:37 -0500)]
ospfd: Cleanup set but unused variables

There existed some variables set but never used.  Clean this up.

Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
4 years agozebra: Cleanup set but unused variables
Donald Sharp [Mon, 24 Feb 2020 13:36:09 +0000 (08:36 -0500)]
zebra: Cleanup set but unused variables

There existed some variables set but never used.  Clean this up.

Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
4 years agoldpd: Cleanup set but unused variables
Donald Sharp [Mon, 24 Feb 2020 13:34:53 +0000 (08:34 -0500)]
ldpd: Cleanup set but unused variables

There existed some variables set but never used.  Clean this up.

Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
4 years agolib: Cleanup set but unused variables
Donald Sharp [Mon, 24 Feb 2020 13:29:27 +0000 (08:29 -0500)]
lib: Cleanup set but unused variables

There existed some variables set but never used.  Clean this up.

Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
4 years agobgpd: Cleanup set but unused variables
Donald Sharp [Mon, 24 Feb 2020 13:29:15 +0000 (08:29 -0500)]
bgpd: Cleanup set but unused variables

There existed some variables set but never used.  Clean this up.

Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
4 years agoMerge pull request #5864 from kuldeepkash/route_map
Donatas Abraitis [Thu, 27 Feb 2020 08:03:16 +0000 (10:03 +0200)]
Merge pull request #5864 from kuldeepkash/route_map

tests: Optimizing route-map test cases

4 years ago*: change hash_backet to hash_bucket
Donald Sharp [Thu, 27 Feb 2020 01:03:34 +0000 (20:03 -0500)]
*: change hash_backet to hash_bucket

It's been a year search and destroy.

Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
4 years agotests: Modified scripts to optimize execution time for route-map test suite
Kuldeep Kashyap [Tue, 25 Feb 2020 20:30:10 +0000 (20:30 +0000)]
tests: Modified scripts to optimize execution time for route-map test suite

1. Modified scripts to optimize execution time for route-map test suite
2. Removed unnecessary print statements from /lib/bgp.py

Signed-off-by: Kuldeep Kashyap <kashyapk@vmware.com>
4 years agoMerge pull request #5866 from qlyoung/fix-ipaddr-isset
Mark Stapp [Wed, 26 Feb 2020 20:34:50 +0000 (15:34 -0500)]
Merge pull request #5866 from qlyoung/fix-ipaddr-isset

lib: fix ipaddr_isset

4 years agolib: fix ipaddr_isset
Quentin Young [Wed, 26 Feb 2020 17:14:21 +0000 (12:14 -0500)]
lib: fix ipaddr_isset

Meaning inverted by mistake

Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
4 years agodoc: document attribute-unchanged directive
rbauduin [Mon, 17 Feb 2020 12:13:11 +0000 (13:13 +0100)]
doc: document attribute-unchanged directive

This directive was undocumented, but is the way to configure the
next-hop to be left unchanged for ipv6 setups (in ipv4, a route-map with
`set next-hop unchanged` works fine, but this is unavailable for ipv6) .

Signed-off-by: Raphael Bauduin <rb@raphinou.com>
4 years agopbrd: unconfigure table range
Stephen Worley [Mon, 24 Feb 2020 20:51:39 +0000 (15:51 -0500)]
pbrd: unconfigure table range

Implement the [no] version of `pbr table range`. We had the command
but were doing nothing with it.

This just calls the set_table_range API again using the defaults.

Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
4 years agoMerge pull request #5783 from ton31337/fix/bad_formatting_bgpd_gr
Russ White [Tue, 25 Feb 2020 16:01:27 +0000 (11:01 -0500)]
Merge pull request #5783 from ton31337/fix/bad_formatting_bgpd_gr

bgpd: Format properly `show bgp neighbors` for graceful restart stuff

4 years agoMerge pull request #5270 from NaveenThanikachalam/rmap
Donald Sharp [Tue, 25 Feb 2020 14:15:49 +0000 (09:15 -0500)]
Merge pull request #5270 from NaveenThanikachalam/rmap

lib: Optimizing route-maps

4 years agoMerge pull request #5771 from pguibert6WIND/show_route_table_protocol
Donald Sharp [Tue, 25 Feb 2020 14:14:56 +0000 (09:14 -0500)]
Merge pull request #5771 from pguibert6WIND/show_route_table_protocol

vty: add command to get route with table, vrf and protocol