[j-nsp] M20 loses TTL=0 packets

[Armin Arh]

Hi,

Regarding the following mtr output, we can see huge loss at router "lon", and a little
at router "bad". Both are m20 with junos-7.2R2.4.

Hostname                                %Loss  Rcv  Snt  Last Best  Avg  Worst
 1. gw-ma-a.net.ma.bs.ka.schlund.net       0%  288  288     0    0    1    143
 2. gw-nat-a.bs.ka.schlund.net             0%  288  288     0    0    0      5
 3. v999.gw-dista.bs.ka.schlund.net        0%  288  288     0    0    4    202
 4. a0c1.gw-backbone-a.bad.schlund.net     6%  272  288     0    0    1     45
 5. so-2000.gw-backbone-a.lon.schlund.n   49%  147  288    13   13   15     97
 6. gig-gw2-UK.cableinet.net               0%  288  288    14   13   19    193
 7. 194.117.136.173                        0%  288  288    22   21   29    220
 8. 194.117.136.157                        0%  288  288    22   21   32    217
 9. 194.117.136.177                        0%  288  288    22   21   30    231
10. 194.117.136.237                        0%  288  288    26   26   33    226
11. 195.188.230.13                         0%  288  288    21   21   26    213
12. 195.188.230.50                         0%  288  288    26   26   29    212
13. 195.188.230.50                       100%    0  288     0    0    0      0

According to firewall counters, there is only 100kBit/s traffic to the
loopback (Mainly SNMP and BGP), thus there is enough capacity on the internal
100MBit/s LAN.

Does someone have an idea, what may cause this behaviour?

thx.
	Armin

-- 
Armin Arh
Schlund + Partner AG
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: not available
Url : https://puck.nether.net/pipermail/juniper-nsp/attachments/20060511/cc0419da/attachment.bin