On Thu, 15 Apr 2010, Darren Reed wrote:

> So there's one other issue that it could be and that is if
> there are too many entries in the same hash bucket.

Got an update from Sun support, they're thinking it's a bug with
connections reusing the same source port...


-----------------------------------------------------------------------

I think I found a relationship with lost and blocked packets and what
triggers it.
It looks like it is an issue related to TCP ports being reused - similar
to this bug which was fixed a while ago.
  6531894 IPF blocks TCP SYN packets for connections in TIME_WAIT state
-> some clients can't reconnect

 04/14/10-15:59  270 in use  lost 1466
 04/14/10-16:00  323 in use  lost 1467

Apr 14 16:00:13 kyle ipmon[117]: [ID 702911 local0.warning]
16:00:13.167168 e1000g0 @20:11 b 134.71.247.43,43034 -> 134.71.247.14,80
PR tcp len 20 60 -S IN

 10 15:58:43.44771 134.71.247.14 -> 134.71.247.43 TCP D=43034 S=80 Fin
Ack=495580729 Seq=1186132000 Len=0 Win=49232 Options=<nop,nop,tstamp
171551837 1687035435>
 11 15:58:43.44781 134.71.247.43 -> 134.71.247.14 TCP D=80 S=43034
Ack=1186132001 Seq=495580729 Len=0 Win=54 Options=<nop,nop,tstamp
1687035435 171551837>
 - >12 16:00:13.16715 134.71.247.43 -> 134.71.247.14 TCP D=80 S=43034 Syn
Seq=1913409678 Len=0 Win=5840 Options=<mss 1460,sackOK,tstamp 1687057864
0,nop,wscale 7>
 13 16:00:13.16718 134.71.247.14 -> 134.71.247.43 TCP D=43034 S=80 Rst
Ack=1913409679 Win=0

 04/14/10-16:04  251 in use  lost 1467
 04/14/10-16:05  345 in use  lost 1468

 Apr 14 16:04:26 kyle ipmon[117]: [ID 702911 local0.warning]
16:04:25.604612 e1000g0 @20:10 b 134.71.68.209,813 -> 134.71.247.14,2049
PR tcp len 20 60 -S IN

 32 16:02:58.58262 134.71.247.14 -> 134.71.68.209 TCP D=813 S=2049 Fin
Ack=3325530777 Seq=1680234158 Len=0 Win=49232 Options=<nop,nop,tstamp
171577350 27868927> 33 19:02:58.58301 134.71.68.209 -> 134.71.247.14 TCP
D=2049 S=813 Ack=1680234159 Seq=3325530777 Len=0 Win=578
Options=<nop,nop,tstamp 27868927 171577350>
-> 34 16:04:25.60459 134.71.68.209 -> 134.71.247.14 TCP D=2049 S=813 Syn
Seq=3782135592 Len=0 Win=5840 Options=<mss 1460,sackOK,tstamp 27955964
0,nop,wscale 7>
 35 16:04:25.60463 134.71.247.14 -> 134.71.68.209 TCP D=813 S=2049 Rst
Ack=3782135593 Win=0
 36 16:04:31.60337 134.71.68.209 -> 134.71.247.14 TCP D=2049 S=813 Syn
Seq=3782135595 Len=0 Win=5840 Options=<mss 1460,sackOK,tstamp 27961964
0,nop,wscale 7>

 04/14/10-16:06  325 in use  lost 1468
 04/14/10-16:07  287 in use  lost 1469

 Apr 14 16:06:53 kyle ipmon[117]: [ID 702911 local0.warning]
16:06:53.145654 e1000g0 @20:10 b 134.71.68.243,844 -> 134.71.247.14,2049
PR tcp len 20 60 -S IN

  45 16:04:55.57943 134.71.247.14 -> 134.71.68.243 TCP D=844 S=2049 Fin
Ack=146698100 Seq=2607094693 Len=0 Win=49232 Options=<nop,nop,tstamp
171589050 27774756>
46 19:04:55.57983 134.71.68.243 -> 134.71.247.14 TCP D=2049 S=844
Ack=2607094694 Seq=146698100 Len=0 Win=501 Options=<nop,nop,tstamp
27774756 171589050>
-> 47 16:06:53.14563 134.71.68.243 -> 134.71.247.14 TCP D=2049 S=844 Syn
Seq=1694971762 Len=0 Win=5840 Options=<mss 1460,sackOK,tstamp 27892323
0,nop,wscale 7>
 48 16:06:53.14567 134.71.247.14 -> 134.71.68.243 TCP D=844 S=2049 Rst
Ack=1694971763 Win=0
 49 16:06:59.14540 134.71.68.243 -> 134.71.247.14 TCP D=2049 S=844 Syn
Seq=1694971765 Len=0 Win=5840 Options=<mss 1460,sackOK,tstamp 27898323
0,nop,wscale 7>


-- 
Paul B. Henson  |  (909) 979-6361  |  http://www.csupomona.edu/~henson/
Operating Systems and Network Analyst  |  [email protected]
California State Polytechnic University  |  Pomona CA 91768

Reply via email to