Solved

Cumulus Switch Can't Ping anything


Just uploaded this and entered the ip address. commit the changes. Can't ping self. This is in vmware using the ova template.

icon

Best answer by Attilla 13 July 2019, 22:04

The ssh credentials are on the bottom right. That allows you to ssh directly into the environment and you shouldnt have such limitations.
View original

14 replies

Hello
As your eth0 is inside mgmt vrf, you need to tell ping to go through this vrf:
ping -i mgmt 192.168.1.58

You can find more info about vrf and executing command inside vrf at the end of https://docs.cumulusnetworks.com/display/DOCS/Virtual+Routing+and+Forwarding+-+VRF

Regards
kage
Its ping -I (capital) but thanks. If that isn't already showing my skill set in this. Do I always have to use the interface switch?

But I'm still unable to ping the virtual IP from the server.

I changed it to class C and it still times out.
Userlevel 2
Could you share the output of "net show configuration"?
cumulus@LEAF1:mgmt-vrf:~$ net show config

time

zone
Etc/UTC

ntp

servers
0.cumulusnetworks.pool.ntp.org iburst
1.cumulusnetworks.pool.ntp.org iburst
2.cumulusnetworks.pool.ntp.org iburst
3.cumulusnetworks.pool.ntp.org iburst

source
eth0

snmp-server
listening-address localhost

dns

nameserver
10.0.2.3

ptp

global

slaveOnly
0

priority1
255

priority2
255

domainNumber
0

logging_level
5

path_trace_enabled
0

use_syslog
1

verbose
0

summary_interval
0

time_stamping
hardware

frr version 4.0+cl3u11

frr defaults datacenter

hostname cumulus

username cumulus nopassword

service integrated-vtysh-config

log syslog informational

vrf routehub-vrf
vni 104001
exit-vrf

interface eth0
description mgmt
address 192.168.101.163/24
alias mgmt
gateway 192.168.101.254
vrf mgmt

line vty

interface lo
# The primary network interface
address 10.0.0.3/32

interface swp1
mtu 9216

interface swp2
mtu 9216

interface swp6
bridge-access 201

interface bridge
bridge-ports routehub swp6
bridge-vids 201 4001
bridge-vlan-aware yes

interface mgmt
address 127.0.0.1/8
vrf-table auto

interface routehub
bridge-access 4001
mstpctl-bpduguard yes
mstpctl-portbpdufilter yes
vxlan-id 104001
vxlan-local-tunnelip 10.0.0.3

interface routehub-vrf
vrf-table auto

interface vlan201
address 172.17.201.3/24
address-virtual 00:00:00:00:00:aa 172.17.201.1/24
vlan-id 201
vlan-raw-device bridge
vrf routehub-vrf

interface vlan4001
vlan-id 4001
vlan-raw-device bridge
vrf routehub-vrf

hostname LEAF1

dot1x
mab-activation-delay 30
eap-reauth-period 0

radius
accounting-port 1813
authentication-port 1812
Userlevel 2
At first glance the port configuration looks correct except for the virtual Mac range you're using. I assume you want to build an evpn configuration? That would need some more configuration, but let's first get the connectivity to the host working.

You said you couldnt reach the virtual ip, can you reach .3? First we should see if the host Mac is learned in the vlan with "net show bridge mac vlan 201"

Also, how does "arp -a" on the host look like? You can also start your ping and check with "tcpdump -I swp6 icmp" to see if you see any icmp packets going in or out.
I changed the mac address and it still times out. I'm using a video series and I'm about half way thru and stuck on this. This is all very strange. Right now leaf 1 only sees the other pc and not what is connected to it. Leaf 2 sees no mac addresses.

cumulus@LEAF1:mgmt-vrf:~$ sudo arp
Address HWtype HWaddress Flags Mask Iface
172.17.201.55 ether 50:00:00:03:00:00 C vlan201-v1 192.168.101.254 (incomplete) eth0
172.17.201.55 ether 50:00:00:03:00:00 C vlan201

cumulus@LEAF1:mgmt-vrf:~$ net show bridge mac vlan 201

VLAN Master Interface MAC TunnelDest State Flags LastSeen
---- ------ --------- ----------------- ---------- --------- ----- --------
201 bridge bridge 00:00:00:00:00:aa permanent 00:15:26
201 bridge bridge 00:00:5e:00:01:01 permanent 00:15:26
201 bridge bridge 50:00:00:04:00:06 permanent 03:14:53
201 bridge swp6 50:00:00:03:00:00 00:08:12
Userlevel 2
Which videos are you looking at?

If I'm not mistaken, you're using gns3. I assume this uses VirtualBox as backend? I haven't used that myself recently and can't help you troubleshoot what is causing the wrong connection.

Perhaps deploying the topology again would solve this.
what about using vmware w/s?
Userlevel 2
If that works for you, that is fine. Given the topology, you might want to give Cumulus in the Cloud a try though. Make sure the configuration is working there and then you can troubleshoot the vm environment you build yourself.
I did try it. But I can only copy and paste one line at a time. I havent figured out how to pull a config from github yet.
Userlevel 2
The ssh credentials are on the bottom right. That allows you to ssh directly into the environment and you shouldnt have such limitations.
Great. Don't quit your day job or I'll never figure this out! Thanks
Userlevel 2
You might want to consider joining our Slack channel at https://slack.cumulusnetworks.com quite some people there that would be willing to help.

Also, if you're testing for any project, let us know where you're based so I can ask one of my colleagues to reach out.

Reply