icmp - TTL Exceeded mpls

[Vladimir_Shenco]

Hello, colleagues. Help to solve following problem :
There is a diagram of switching on

Customers ->MX240--ldp lsp-->X670--ldp lsp-->X670--ldp lsp-->ldp lsp-->X670--ldp lsp-->MX480-->Internet

All Extreme X670 - 15.7.2.9 - intermediate LSR
MX240 - egress mpls lsr
MX480 - ingress mpls lsr

Between MX240 and MX480 bgp session is set .
Сustomer does traceroute :

traceroute to 8.8.8.8 (8.8.8.8), 30 hops max, 60 byte packets
1 10.10.10.1 (10.10.10.1) 0.327 ms 0.332 ms 0.315 ms
2 * * * here shall be first X670
3 * * * here shall be second X670
4 * * * here shall be third X670

To documentation it is written :
"In pipe TTL mode, the LSP is viewed as a point-to-point link between the ingress LSR and the egress LSR; intermediate LSRs in the MPLS network are not viewed as router hops from an IP TTL perspective. "

Why from X670 doesn't come ttl exceeded ? Why X670 generally is visible in trace ?

[Vladimir_Shenco]

Analogs mpls icmp-tunneling , mpls no-decrement-ttl on X670 I didn't find .
From hopelessness I tried iproute mpls-next-hop, ospf mpls-next-hop , but it of course didn't help.

[andreas1]

Yeah since juniper use two different styles of this I'm not sure that extreme even support this ? cause it has to happen at the LSP level.

// Andeas

[Vladimir_Shenco]

Yes.
I supposed that X670 generally won't be visible in the route

[andreas1]

Do you mean that the trace dosen't show the X670 ?


// Andrease