[j-nsp] Anybody else seen this?

[Stefan Fouant]

Seems there might be an issue with the MTR function using 'traceroute
monitor' when used in conjunction with Logical Routers.  If I do a regular
traceroute from within a Logical router, the output is as expected:

r1 at lab:r1# run traceroute 10.0.9.6 
traceroute to 10.0.9.6 (10.0.9.6), 30 hops max, 40 byte packets
 1  10.0.4.13 (10.0.4.13)  0.405 ms  0.345 ms  0.314 ms
 2  10.0.9.6 (10.0.9.6)  0.308 ms  0.337 ms  0.295 ms

However, if I do a traceroute using the 'monitor' switch, it seems to want
to source from the fxp0 interface in use on the actual router, not the
logical router.  In fact, it doesn't seem to want to use the route table
from the logical router at all:

r1@ lab:r1# run traceroute monitor 10.0.9.6    

                                                My traceroute  [v0.69]
lab (0.0.0.0)(tos=0x0 psize=64 bitpattern=0x00)
Mon Nov 23 01:50:33 2009
Keys:  Help   Display mode   Restart statistics   Order of fields   quit
 
Packets               Pings
 Host
Loss%   Snt   Last   Avg  Best  Wrst StDev
 1. 172.16.100.1
0.0%    13    2.4   2.4   2.3   2.5   0.1
 2. 172.16.98.1
0.0%    13    2.8   3.0   2.8   3.9   0.3
 3. 38.104.59.49
0.0%    13   12.7  13.7  10.8  17.4   1.7
 4. 38.20.35.249
15.4%    13   13.0  15.9  11.3  26.4   5.3
 5. 154.54.29.117
0.0%    13   14.1  17.1  10.6  35.2   7.2
 6. 66.28.67.226
0.0%    13   15.1  15.7  12.4  23.5   2.7
 7. 66.28.31.114
0.0%    13   12.1  14.7  12.1  16.8   1.4
 8. 66.28.16.86
0.0%    13   16.1  18.8  13.8  28.1   4.8
 9. 66.28.16.85
0.0%    13   13.1  16.0  11.9  27.5   3.8
10. 66.28.16.86
0.0%    13   18.1  18.9  11.8  35.1   6.8
11. 66.28.16.85
0.0%    13   15.1  17.4  12.9  31.4   5.5

Specifying a source address did not appear to resolve the issue either.  I'm
running JUNOS 9.6R1.13.  This appears to be a bug.  I'll open a JTAC case
but wanted to see if anyone else observed this behavior and get a sanity
check.

Regards,

Stefan Fouant
www.shortestpathfirst.net
GPG Key ID: 0xB5E3803D