I have a simple leaf a spine topology set up 5110 spines (irb gws) and 5100 leafs (l2 vtep only). This works fine no issues until you try and change any config! has anyone seen any behaviour like this before?
See below the behaviour:
All 3 tunnels operational and working fine.
{master:0}[edit]
root@LEAF-1# run show ethernet-switching vxlan-tunnel-end-point remote
Logical System Name Id SVTEP-IP IFL L3-Idx
<default> 0 10.10.1.3 lo0.0 0
RVTEP-IP IFL-Idx NH-Id
10.10.1.1 555 1720
VNID MC-Group-IP
600 0.0.0.0
102 0.0.0.0
101 0.0.0.0
RVTEP-IP IFL-Idx NH-Id
10.10.1.2 557 1749
VNID MC-Group-IP
600 0.0.0.0
101 0.0.0.0
102 0.0.0.0
RVTEP-IP IFL-Idx NH-Id
10.10.1.4 556 1736
VNID MC-Group-IP
600 0.0.0.0
101 0.0.0.0
102 0.0.0.0
A trunk port which connects to a server (the vlans are all assigned to a vtep)
{master:0}[edit]
root@LEAF-1# run show configuration interfaces ge-0/0/48
unit 0 {
family ethernet-switching {
interface-mode trunk;
vlan {
members [ SERVER CLIENT CLIENT2 ];
}
}
}
I then remove a VLAN from the trunk port and…..
{master:0}[edit]
root@LEAF-1# show | compare
[edit interfaces ge-0/0/48 unit 0 family ethernet-switching vlan]
- members [ SERVER CLIENT CLIENT2 ];
+ members [ SERVER CLIENT ];
{master:0}[edit]
root@LEAF-1# commit
[edit protocols]
'bgp'
warning: requires 'bgp' license
[edit vlans CLIENT]
'vxlan'
warning: requires 'vxlan' license
configuration check succeeds
commit complete
No more vteps.
{master:0}[edit]
root@LEAF-1# run show ethernet-switching vxlan-tunnel-end-point remote
Logical System Name Id SVTEP-IP IFL L3-Idx
<default> 0 10.10.1.3 lo0.0 0
{master:0}[edit]
root@LEAF-1# run show ethernet-switching vxlan-tunnel-end-point remote
Logical System Name Id SVTEP-IP IFL L3-Idx
<default> 0 10.10.1.3 lo0.0 0
If I now clear BGP the VTEPs come back up.
This is proving to be very difficult to fix and Juniper themselves are running out of options, I think I have tried every combination of firmware.
Currenlty both the 5110 and the 5100 are on 17.4
Has anyone every witnessed anything like this? i am speaking to ATAC at the moment who also right now have no idea what it can be.
This has been set up in a lab with a view of taking it in production. When it works.
E-1[1]: HMCIF: Link4: HMC Fatal Error cmd:62 lng:1 ltag:2 dinv:0 errstat:127 err_cnt:0x40000000