[Bug 257965] [fib_algo][dxr] inbound dropp on interfaces
Date: Fri, 20 Aug 2021 10:27:24 UTC
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=257965
Bug ID: 257965
Summary: [fib_algo][dxr] inbound dropp on interfaces
Product: Base System
Version: 13.0-STABLE
Hardware: Any
OS: Any
Status: New
Severity: Affects Only Me
Priority: ---
Component: kern
Assignee: bugs@FreeBSD.org
Reporter: konrad.kreciwilk@korbank.pl
I have DELL R630 with ccX (Chelsio T62100-SO-CR) agreggated with lagg0 and
vlans on it. I move all vlans to JAIL (VNET). There works bird which received
full feed (900k prefixes).
When I have set:
net.route.algo.inet.algo: dxr
the inbound dropp counter for ccX is incrementing. When I use dpdk_lpm4 or
radix4 it do not happen.
graph for net.route.algo.inet.algo=radix4
https://imgur.com/a/LsKBioR
graph for net.route.algo.inet.algo=dxr
https://imgur.com/a/UNsrAk6
# netstat -i
Name Mtu Network Address Ipkts Ierrs Idrop Opkts Oerrs
Coll
cc0 1500 <Link#1> 00:07:43:64:9e:60 1133593300298 44384 3593534
1133260298112 0 0
cc1 1500 <Link#2> 00:07:43:64:9e:68 0 0 0 0 0
0
bge0* 1500 <Link#3> 14:18:77:3d:10:5d 0 0 0 0 0
0
bge1* 1500 <Link#4> 14:18:77:3d:10:5e 0 0 0 0 0
0
bge2 1500 <Link#5> 14:18:77:3d:10:5b 19362895 0 0 15602306 0
0
bge2 - 10.19.70.0/24 10.19.70.20 19320567 - - 15559975 -
-
bge3* 1500 <Link#6> 14:18:77:3d:10:5c 0 0 0 0 0
0
cc2 1500 <Link#7> 00:07:43:64:9e:60 1108050218194 120531 1639219
1107797492457 0 0
cc3 1500 <Link#8> 00:07:43:64:9e:18 0 0 0 0 0
0
lo0 16384 <Link#9> lo0 106 0 0 106 0
0
lo0 - localhost localhost 0 - - 0 -
-
lo0 - fe80::%lo0/64 fe80::1%lo0 0 - - 0 -
-
lo0 - your-net localhost 106 - - 106 -
-
lagg0 1500 <Link#10> 00:07:43:64:9e:60 2241643518492 164915 5232753
2241057790569 0 0
dmesg when net.route.algo.debug_level=6
[fib_algo] inet.0 (dxr#213) handle_fd_callout: running callout type=3
[fib_algo] inet.0 (dxr#213) dxr_change_rib_batch: processing 2 update(s)
[fib_algo] inet.0 (dxr#213) dxr_build: D16X4R, 845809 prefixes, 751 nhops (max)
[fib_algo] inet.0 (dxr#213) dxr_build: 1311.29 KBytes, 1.58 Bytes/prefix
[fib_algo] inet.0 (dxr#213) dxr_build: range table updated in 0.008 ms
[fib_algo] inet.0 (dxr#213) dxr_build: trie updated in 0.001 ms
[fib_algo] inet.0 (dxr#213) dxr_build: snapshot forked in 0.547 ms
[fib_algo] inet.0 (dxr#213) dxr_build: range table: 2%, 26970 chunks, 8 holes
[fib_algo] inet.0 (dxr#213) handle_fd_callout: running callout type=3
[fib_algo] inet.0 (dxr#213) dxr_change_rib_batch: processing 28 update(s)
[fib_algo] inet.0 (dxr#213) dxr_build: D16X4R, 845809 prefixes, 751 nhops (max)
[fib_algo] inet.0 (dxr#213) dxr_build: 1311.29 KBytes, 1.58 Bytes/prefix
[fib_algo] inet.0 (dxr#213) dxr_build: range table updated in 0.110 ms
[fib_algo] inet.0 (dxr#213) dxr_build: trie updated in 4.664 ms
[fib_algo] inet.0 (dxr#213) dxr_build: snapshot forked in 0.462 ms
[fib_algo] inet.0 (dxr#213) dxr_build: range table: 2%, 26970 chunks, 8 holes
[fib_algo] inet.0 (dxr#213) handle_fd_callout: running callout type=3
[fib_algo] inet.0 (dxr#213) dxr_change_rib_batch: processing 2 update(s)
[fib_algo] inet.0 (dxr#213) dxr_build: D16X4R, 845807 prefixes, 751 nhops (max)
[fib_algo] inet.0 (dxr#213) dxr_build: 1311.29 KBytes, 1.58 Bytes/prefix
[fib_algo] inet.0 (dxr#213) dxr_build: range table updated in 0.005 ms
[fib_algo] inet.0 (dxr#213) dxr_build: trie updated in 0.001 ms
[fib_algo] inet.0 (dxr#213) dxr_build: snapshot forked in 0.473 ms
[fib_algo] inet.0 (dxr#213) dxr_build: range table: 2%, 26970 chunks, 8 holes
[fib_algo] inet.0 (dxr#213) handle_fd_callout: running callout type=3
[fib_algo] inet.0 (dxr#213) dxr_change_rib_batch: processing 56 update(s)
[fib_algo] inet.0 (dxr#213) dxr_build: D16X4R, 845807 prefixes, 751 nhops (max)
[fib_algo] inet.0 (dxr#213) dxr_build: 1311.29 KBytes, 1.58 Bytes/prefix
[fib_algo] inet.0 (dxr#213) dxr_build: range table updated in 0.092 ms
[fib_algo] inet.0 (dxr#213) dxr_build: trie updated in 2.845 ms
[fib_algo] inet.0 (dxr#213) dxr_build: snapshot forked in 0.545 ms
[fib_algo] inet.0 (dxr#213) dxr_build: range table: 2%, 26970 chunks, 8 holes
--
You are receiving this mail because:
You are the assignee for the bug.