On 03/11/09 21:19, E Lesage wrote:
Some one else on the team is looking into these
problems but the following may be useful.
Feb 4 10:40:28 rob-039v smbd[463]: [ID 766186
daemon.error]
NbtDatagramDecode[11]: too small packet
A NetBIOS datagram header should be 14 bytes but this
indicates that we received 11 bytes.
If nothing else has changed in your network
environment, this may be related to
6765829 Integration of project Volo PSARC/2007/587
because PSARC 2007/587
(Volo -- Low Latency Socket Framework) and PSARC
2008/694 (Volo Interfaces
Amendment) delivered into snv_106.
In order to pursue this, it would be really useful if
we could capture these packets on the
network so that we can check where the message was
truncated (on the network or
within the TCP/IP stack).
Hi,
While I can't be sure whether this is the same issue as the original poster, I
have the same messages here, every 12 minutes, exactly the broadcast interval
for the NetBIOS announcements. I am running OpenSolaris 2008.11 inside
VirtualBox, and I've run a wireshark on the host. The messages are NetBios
datagram error messages, in this case sent by my printer back to the
Opensolaris virtual guest (not sure why). (Ethernet addresses partially masked).
0000 08 00 27 XX XX XX 00 14 38 XX XX XX 08 00 45 00 ..'XXX.. 8XXX..E.
0010 00 27 a1 9a 00 00 40 11 bd bd 0a 01 01 64 0a 06 .'....@. .....d..
0020 06 04 00 8a 00 8a 00 13 3c c7 13 02 00 e8 0a 06 ........ <.......
0030 06 04 00 8a 82 00 00 00 00 00 00 00 ........ ....
The 11 bytes NetBIOS datagram message is 13 02 00 e8 0a 06 06 04 00 8a 82 and
decodes as:
Message Type: Datagram error (19)
More fragments follow: No
This is first fragment: Yes
Node Type: B node (0)
Datagram ID: 0x00e8
Source IP: 10.6.6.4 (10.6.6.4)
Source Port: 138
Error code: Destination name not present
Hope this helps. Regards,
Yes, thanks. Our datagram service isn't handling this response
correctly, which result in that spurious log message.
Please ignore this (too small packet) message for now. We'll fix
the service to deal with these responses.
Alan
_______________________________________________
cifs-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/cifs-discuss
