Two companies have a dedicated T1 between their offices,
at company A there is one OSR 5.0.7 system (A1),
at company B there are two OSR 5.0.7 (B1 and B2) and one OSR 5.0.6 (B3).
The A1, B1 and B2 are the same HP ML380s, the B3 is older HP h/w
There is a ftp-based data exchange established between system at A and all three systems at B.
The speed between A1 / B1 and A1 / B2 is around 55 Kbytes/s; the speed between A1 and B3 is around 220 Kbytes/s.
The routing scheme at company B is the same on all three boxes. The B2 and B3 are on the same subnet, B1 is on another subnet. The load on B3 is actually the highest of those three at company B.
I can't understand how is the older and fully loaded system achieves 3 times higher speed.
What should I check and tweak to ensure all three systems from B would get the same 200+ Kbytes/s speed?
sorry for the late reply, the systems are not readily accesable to me.
I will try to answer the questions here
network cards are configured to be auto on speed and on flow control
one single switch used for B1, B2 and B3
one firewall for the whole B company, no specific configs for the OSR systems
don't know how to get errors on the network, should I just run
netstat -s
the output is quite lengthy, which section should I post here?
traceroute outputs:
From B1 to A (slow)
traceroute to 12.222.222.2 (12.222.222.2), 30 hops max, 40 byte packets
1 11.22.33.50 (11.22.33.50) 0.807 ms 0.635 ms 0.684 ms
2 123.22.23.1 (123.22.23.1) 3.913 ms 4.078 ms 3.943 ms
3 123.22.23.13 (123.22.23.13) 3.780 ms 4.198 ms 4.057 ms
4 123.22.23.14 (123.22.23.14) 7.287 ms 6.699 ms 7.058 ms
5 12.222.222.2 (12.222.222.2) 6.731 ms 6.942 ms 6.553 ms
From B3 to A (fast)
traceroute to 12.222.222.2 (12.222.222.2), 30 hops max, 40 byte packets
1 11.22.33.50 (11.22.33.50) 1.089 ms 0.857 ms 0.768 ms
2 123.22.23.1 (123.22.23.1) 4.759 ms 6.764 ms 9.643 ms
3 123.22.23.13 (123.22.23.13) 27.320 ms 13.411 ms 3.507 ms
4 123.22.23.14 (123.22.23.14) 6.998 ms 7.296 ms 6.751 ms
5 12.222.222.2 (12.222.222.2) 6.911 ms 9.518 ms 37.746 ms
They look very similar to me.
Please tell me what else is needed to pinpoint the problem?
Thanks in advance.
Output from netstat -s would be good to see, even if it is a bit long.
Do you have a problem in FTP speed when you compare speeds just within company B?
We had a problem once where the auto-negotiation was wrong and we had a switch at 100M-Full duplex with a servers at 10M-Half. It all worked fine until we forced lots of data with an FTP, so it was difficult to track down.
You might need to ask the network manager to display the switch ports settings. You may be able to give IP address, or they may require the MAC address of your network card.
I regret I do not know how to display the requested and negotiated speeds on OSR. Can anyone else chip in?
Within the company B1 to B3 is around 4300 Kbytes/s, no problems there.
Here is data from B1 (slow conn speed) system:
$ netstat -s
ip:
21879416 total packets received
0 bad header checksums
0 with bad IP version number
0 with size smaller than minimum
0 with data size < data length
0 with header length < data size
0 with data length < header length
0 with unknown protocol
0 with link layer broadcast addr but unicast IP addr
0 fragments received
0 fragments dropped (dup or out of space)
0 fragments dropped after timeout
0 packets reassembled
0 packets forwarded
4730 packets not forwardable
0 no routes
0 redirects sent
0 system errors during input
21874686 packets delivered
22058749 total packets sent
0 system errors during output
0 packets fragmented
0 packets not fragmentable
0 fragments created
0 PCBs failed connect due to bad source address
icmp:
0 calls to icmp_error
0 errors not generated because old message was icmp
Output histogram:
echo reply: 588
echo: 28
0 messages with bad code fields
0 messages < minimum length
0 bad checksums
0 messages with bad length
0 bad netmasks received
Input histogram:
echo reply: 28
echo: 588
588 message responses generated
616 messages received
616 messages sent
0 messages not sent due to bad source address
0 system errors during output
igmp:
0 messages received
0 messages received with too few bytes
0 messages received with bad checksum
0 membership queries received
0 membership queries received with invalid field(s)
0 membership reports received
0 membership reports received with invalid field(s)
0 membership reports received for groups to which we belong
0 membership reports sent
0 total packets sent
0 output errors
tcp:
22056572 packets sent
22056536 packets used fast path
20399099 data packets (1672139922 bytes)
641 data packets (546447 bytes) retransmitted
56738 ack-only packets (28961 delayed)
0 URG only packets
16 window probe packets
1598492 window update packets
1586 control packets
69 resets
21712074 packets received
2851244 acks (for 1672210416 bytes)
22777 duplicate acks
0 acks for unsent data
18940013 packets (1470375583 bytes) received in-sequence
0 completely duplicate packets (0 bytes)
20574 packets with some dup. data (26697 bytes duped)
42 out-of-order packets (14850 bytes)
0 packets (0 bytes) of data after window
0 window probes
7141 window update packets
0 packets received after close
0 discarded for bad checksums
0 discarded for bad header offset fields
0 discarded because packet too short
0 system errors encountered during processing
765 connection requests
161 connection accepts
924 connections established (including accepts)
1398 connections closed (including 8 drops)
12 embryonic connections dropped
0 failed connect and accept requests
7 resets received while established
1784835 segments updated rtt (of 1785338 attempts)
296 retransmit timeouts
1 connection dropped by rexmit timeout
29 persist timeouts
0 alloc failures caused reschedule
191 keepalive timeouts
191 keepalive probes sent
0 connections dropped by keepalive
18829133 segments predicted
2604161 acks predicted
udp:
0 incomplete headers
0 bad data length fields
0 bad checksums
10196 bad ports (10196 were broadcast/multicast)
1037 input packets delivered
0 system errors during input
1304 packets sent
0 streams allocation failures