Solved

Latency across link

Posted on 2011-09-14
17
502 Views
Last Modified: 2012-06-21
I have a 1 gb link but FTP transfers get about 3MB per second and windows file transfers get about 2MB per second across the link.

We have sent 100MB across with an analyzer and it showed no packet loss.

Wireshark shows some errors but no warnings.

What are some ways to determine if this is a network or server issue?

Do I need to capture all the traffic on the link with wireshark?

Currently I am just capturing the traffic destined to the local machine receiving the file transfers.

There is some broadcast traffic but it does not seem to be excessive but I am not sure how to determine that.
0
Comment
Question by:Dragon0x40
  • 6
  • 5
  • 2
  • +4
17 Comments
 
LVL 12

Accepted Solution

by:
FarWest earned 72 total points
ID: 36540574
use tracert to check if routers in the middle and which one is causing the delay if any
if no letancy appears with tracert, then check if you have antivirus that monitor the packets,
and check NIC drivers in both source and distination

good luck
0
 
LVL 8

Assisted Solution

by:SeeMeShakinMyHead
SeeMeShakinMyHead earned 72 total points
ID: 36541944
are the errors you are seeing in WireShark checksum errors.  This is typically normally with today's NIC's that do checksum offloading at the NIC.  Also, try running iPerf on  both locations and send a large amount of data and see what it reports.  @fryezz - I had a similary issue with CIFS transfers where Symantec Enterprise Protection (SEP) was hindering the traffic flow.  
0
 
LVL 22

Assisted Solution

by:eeRoot
eeRoot earned 72 total points
ID: 36542633
How many network devices are there between the 2 devices?  At least one of them may have QOS or some speed limiting setting.  Also, do a read/write speed test on the 2 machines to verify neither is having an issue with its hard drive(s).  If one of the machine has very slow read/write results, it may be a workstation issue, rather then a network issue
0
Connect further...control easier

With the ATEN CE624, you can now enjoy a high-quality visual experience powered by HDBaseT technology and the convenience of a single Cat6 cable to transmit uncompressed video with zero latency and multi-streaming for dual-view applications where remote access is required.

 
LVL 57

Assisted Solution

by:giltjr
giltjr earned 71 total points
ID: 36544901
Its' a bit unusual that FTP gets slower throughput than Windows (SAMBA) File copies.

Are you sure the full path is 1 Gbps between the client and the server?  Assuming you mean mega bytes when you say 2-3 MBps, that is 16 to 24 mega-bits per second.  Sounds almost like wireless, cable, or a fractional DS3 WAN link.
0
 
LVL 17

Assisted Solution

by:rochey2009
rochey2009 earned 142 total points
ID: 36546240
Hi,

What does the wireshark capture show? Have a look at TCP round trip times, retransmissions, maximum segment size, TCP window size etc.
0
 
LVL 17

Assisted Solution

by:rochey2009
rochey2009 earned 142 total points
ID: 36546274
TCP window scaling should allow better utilization of the bandwidth.
0
 
LVL 28

Assisted Solution

by:mikebernhardt
mikebernhardt earned 71 total points
ID: 36550586
you haven't yet explained the topology of your network, but since your analyzer can send 100mpbs I'm assuming that this is all LAN with no WAN/wireless links in between. So make sure that:
1. The speed of the link is actually 1GB on both sides. Look at the Local Area Connection status.
2. The duplex setting on the 2 machines match the duplex setting on the switch ports. A mismatch has an enormous impact on performance. This isn't normally an issue with Gig links (should always be full duplex) but it's worth looking at.
0
 
LVL 57

Expert Comment

by:giltjr
ID: 36552769
Actually the analyzer sent a 100MB file, he does not say how fast it sent it.  

I was assuming that it sent it at 2-3 MBps (16-24Mbps).  Which is about what I would expect from a 54Mbps wireless link with WPA encryption or some time of WAN link.
0
 

Author Comment

by:Dragon0x40
ID: 36555504
I have attached a RTT graph from Wireshark.

We did send a 100MB test signal with an analyzer across the link and did not have any packet loss.

Also looking at a routing problem that may be sending traffic through the layer 2 cloud twice.

This is going through a layer 2 cloud so how many hops in between I don't know.
RTT-graph.png
0
 

Author Comment

by:Dragon0x40
ID: 36555523
0
 
LVL 57

Expert Comment

by:giltjr
ID: 36559433
Assuming you are running Windows, from one of the computers you are testing with can you provide the output from the command:

     pathping -n x.x.x.x

where x.x.x.x is the IP address of the of the other computer you are testing with?
0
 

Author Comment

by:Dragon0x40
ID: 36915091
I think this may be a routing issue with extra hops through the layer 2 cloud. I will post when I find out.
0
 
LVL 57

Expert Comment

by:giltjr
ID: 36916197
Can you please provide the output from pathping, or even just a ping.

The only time a "layer 2 could" should come into play is if layer 2 is bridged over a WAN.
0
 

Author Comment

by:Dragon0x40
ID: 36959951
I wil do a pathping and post the results.
0
 

Author Comment

by:Dragon0x40
ID: 36965574
C:\Users\xxxx>pathping 1.2.3.4

Tracing route to 1.2.3.4 over a maximum of 30 hops

  0  2.3.4.5
  1  7.8.9.10
  2  3.4.5.6
  3  4.5.6.7
  4  5.6.7.8
  5  6.7.8.9
  6  1.2.3.4

Computing statistics for 150 seconds...
            Source to Here   This Node/Link
Hop  RTT    Lost/Sent = Pct  Lost/Sent = Pct  Address
  0                                           2.3.4.5
                                0/ 100 =  0%   |
  1    0ms     0/ 100 =  0%     0/ 100 =  0%  7.8.9.10
                                0/ 100 =  0%   |
  2    2ms     0/ 100 =  0%     0/ 100 =  0%  3.4.5.6
                                0/ 100 =  0%   |
  3    2ms     0/ 100 =  0%     0/ 100 =  0%  4.5.6.7
                                0/ 100 =  0%   |
  4   60ms     0/ 100 =  0%     0/ 100 =  0%  5.6.7.8
                                0/ 100 =  0%   |
  5   60ms     0/ 100 =  0%     0/ 100 =  0%  6.7.8.9
                                0/ 100 =  0%   |
  6   60ms     0/ 100 =  0%     0/ 100 =  0%  1.2.3.4

Trace complete.

C:\Users\xxxx>ping 1.2.3.4

Pinging 1.2.3.4 with 32 bytes of data:
Reply from 1.2.3.4: bytes=32 time=60ms TTL=59
Reply from 1.2.3.4: bytes=32 time=60ms TTL=59
Reply from 1.2.3.4: bytes=32 time=60ms TTL=59
Reply from 1.2.3.4: bytes=32 time=60ms TTL=59

Ping statistics for 1.2.3.4:
    Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
    Minimum = 60ms, Maximum = 60ms, Average = 60ms
0
 
LVL 57

Expert Comment

by:giltjr
ID: 36969788
O.K, unless you work for a real rich company or a major telco company, you do not have a 1Gbps link between the two computers in question.  Your computer may have a 1Gbps NIC and the other computer may have a 1 Gbps NIC, but there is a WAN between the two which is running MUCH slower than 1 Gbps.

So you need to find out what the WAN bandwidth is.  I am going to assume that you are not in the networking group, or you would have already known that there is a WAN between the two computers.  So you need to contact somebody in your networking group and find out what the WAN bandwidth is.

I am also assuming that you, or your company, owns/controls both computers that are involved.

Based on the 60ms RTT, my guess is that these two computers are a few thousand miles apart.  The company I work for has a location just outside Washington DC and one in London and the RTT avg. 90-100 ms.
0
 

Author Closing Comment

by:Dragon0x40
ID: 36983543
Was a routing issue with an incorrect next hop in the routing table. The traffic was having to go thru the mpls cloud twice instead of once.
0

Featured Post

Free Tool: Site Down Detector

Helpful to verify reports of your own downtime, or to double check a downed website you are trying to access.

One of a set of tools we are providing to everyone as a way of saying thank you for being a part of the community.

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

Suggested Solutions

Title # Comments Views Activity
Connecting Servers to L2 OR L3 Switch 6 68
Performance monitors 6 44
Why isn't my network passing a certain vlan. 24 39
Router speed limit 7 56
Load balancing is the method of dividing the total amount of work performed by one computer between two or more computers. Its aim is to get more work done in the same amount of time, ensuring that all the users get served faster.
I recently attended Cisco Live! in Las Vegas, a conference that boasted over 28,000 techies in attendance, and a week of hands-on learning hosted by a solid partner with which Concerto goes to market.  Every year, Cisco displays cutting-edge technol…
After creating this article (http://www.experts-exchange.com/articles/23699/Setup-Mikrotik-routers-with-OSPF.html), I decided to make a video (no audio) to show you how to configure the routers and run some trace routes and pings between the 7 sites…
In this tutorial you'll learn about bandwidth monitoring with flows and packet sniffing with our network monitoring solution PRTG Network Monitor (https://www.paessler.com/prtg). If you're interested in additional methods for monitoring bandwidt…

679 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question