cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
2271
Views
0
Helpful
6
Replies

Connection speed RV120W

henryhudson
Level 1
Level 1

We are having same speed issue with RV120W as reported by a couple of other posts in that the connection speed drops near useless.

eg. alternate router ing ISP 4ms respone, RV120 40ms.

We connect through a large pipe, and I notice the was reference to RV's sometimes have problem with large pipes.

We have tried block size change, reset router, diabled all logs etc etc, but not had any success.

Has there been any advancement  or other suggestions you have to oversome this issue?

Our problem has been very frustrating, as we had to change Linksys router due to implementation on Voip system, and we have been very hamstrung trying to resolve this problem. along the way, we obtained 4 other routers, to eventually confirm that it is problem with the RV120.

thank you

Henry h

6 Replies 6

croelectronic
Level 1
Level 1

I will join to this thread with the similar problem. Actually, I think router might have problem with CPU utilization. I tried simple ping test and anything (LAN IP, WAN IP, GATEWAY, ANOTHER VLAN CLIENTS) has the same problem. The only thing that work is to ping client on the same VLAN. As you can see reply time jumps every 10-30 seconds most of the time to 50-100 ms. You can see it from attached screenshot. I have another unit already deployed, but I will get it back next week and test it. I have feeling that it will behave the same. This test was with two clients on VLAN1 and one on VLAN2 with no other activity but pinging. QoS was disabled for this test, but I had it on and it didn't make any difference.

This has to be fixed ASAP, because router is useless.

I have to say that I am a little disappointed to hear these reports after ordering one myself.  I should have it in my hands this upcoming Tuesday.  I will update the thread on my experience.

Hi

With the greatest respect gentleman,  would it be possible to collect and measure throughput with something better than ping latency or speedtest.

Something like Iperf or Jperf just works so much better. It wil give you reasonable baseline measurements on throughput from a PC host on the LAN to a PC host on the WAN interface.

Gentleman I believe you think that throughput is bad,  using an approach below, you can isolate and see if throughput from the LAN to the WAN is Good or BAD.  Also learning how to  use Iperf or Jpert is a pretty useful skill to know.

Seeing higher  latency on a couple of pings over a 30 second period doesn't really tell me anything as to the cause of the issue.

Anecdotal information mentioned in the first posting is a wee bit telling, but doesn't gel  with information post on throughput data mentioned on the RV120W datasheet.  Cisco would not have posted that sort of throughput data  on the RV120W datasheet for nothing.

Why not try something like  Jperf 2.0.2.  It's free but runs like an old requestor responder type service.

My PC on the WAN interface was a Jperf server, whilst my PC client on the LAN was a Jperf Client.

For example,  in the example below,  I ran with a cisco small business router this evening.  Writing it up here too me longer than running some tests.

The WAN interface was connected to the Jperf server.  I statically defined IP addresses to both the routers WAN interface and the PC that is directly connected to that WAN  interface.

I have showed the IP addressing scheme in my rough diagram below.

My setup was ;

        Jperf                                                         Jperf

    PC server--------------- (WAN)  ROUTER  (LAN)------------------PC client

    ip=1.1.1.1         ip=1.1.1.2   SA520    ip=192.168.75.1    ip=192.168.75.10

Here are the results from my test below;

Results shows that I pushed through 384 Megabytes   and averaged about 106 megabits/second throughput  from the LAN iPC host out through to a IP  host on the WAN interface.  I now have some 'baseline' measurements to workwith.

Please note that my 'stream was going through a NATted / firewall  interface

It would be interesting also if you would post more configuration data and  network topology data or even better a diagram

1.  Software version?

2.  factory reset after software upgrade...bla bla bla

It's really really hard to tell from the problem descriptions what might be going wrong..

Gentleman, if you wish, try out Jperf   (it's a respected freebie application)

But,  if after running Jperf you are still  concerned that the RV120W has a issue , why not open a call with the Small Business Support center, your routers are all under warranty.

I am a bit perplexed by your results,  as the RV102W  datasheet does state  the following;


regards Dave

Ok, I have this router online and I am afraid I will have to return it.  I thought surely there would be a firmware update but it is running the only version currently available.

When connected via hardwire, its hard to make a case for it being slow.  When connected via Wi-Fi in N-Only mode, speed is terrible.  I switched back to my trusty WRT54G to post this message.  However, speed isn't the only problem.

I setup access rules to allow connections to a couple of my servers.  One of these servers I only allowed RDP.  (I thought it was odd that there is no pre-configured RDP protocol for this device.  Had to create a "custom service".)  I secure my RDP session with SSL.  Using this router/firewall, I am unable to maintain a session.  Although I can create an RDP session, it will break sporadically because of data encryption errors.  If I switch back to my WRT54G, no worries..

I setup MRTG to collect usage stats for reference.  I wanted to monitor CPU but I cant seem to determine the OID for this.  Viewing the status fro the routers interface, it doesn't appear to be CPU bound but it would be nice to have good metrics for this.

Anyway, unless there is an update available, I seriously doubt a support (warranty) call will fix these problems.  That is unless I just have a defective unit and the call results in a replacement or upgrade.

This is a shame because the device has a great feature set that I was really excited about.  A lot of "bang for the buck".  But with these issues, not so much bang...

As advised, I did some tests and here are conclusions.

Configuration setup:

Notebook (Broadcom Fast Ethernet) as client (LAN side), Dell SC 430 (Broadcom Gigabit) as server (WAN side). No switches, only RV120W between.

First I tested client and server to make sure that there is no problem with them (direct connection). Throughput around 93Mb/s +-1.

First test was TCP throughput. 50-70 Mb/s, jumping up and down, very unstable. So I would say this router can handle 50 Mb/s, peeking up to 70 Mb/s More then that goes into "advertising" material only. But this is not my problem, because it's going to be used on DSL 6Mb/s connection anyway.

I was more courios about problem that I mentioned. Ping reply latency. And if there is something else behind that.

TCP is lossless protocol, at least up to certain point, so the only way  to verify packet loss and jitter is UDP test.

So I had test set to 300 seconds, transferring at the following rates: 64kb/s, 500kb/s, 1Mb/s, 5Mb/s, 10Mb/s

At the same time I was pinging router's LAN ip address to see how transfer rate, packet loss and jitter are behaving at the moments when I'm getting increased reply time.

What I found is very disturbing.

Increased jitter every time when there is increased reply time, sometimes jitter rapidly increasing second after second up to one minute until it goes back to normal. There is no packet loss at the lower speeds (64kb/s, 500kb/s), but on higher speeds there is significant packet loss. At 1Mb/s tipicaly around 0.5% up to 1%, at 5Mb/s typically 2-3% up to 5% and at 10Mb/s typically 5% up to 10%.

I did all tesst LAN -> WAN, then I repeated the same tests VLAN1-VLAN3. No difference, the same issues. VLAN1-VLAN1 didn't show the same problem, it worked fine.

To make sure I didn't get bad apple, I tested another RV120W and results are the same. Both routers are experiencing the same symptoms.

I also noticed that if there is no traffic to very light traffic those hiccups are no so frequent, but if I start to push data at the constant rate through the router are start to appear every 10-20 seconds.

I will not draw any conclusion, but I would like somebody from Cisco to step up here, do tests and give us some feedback. ASAP

I am very upset that CISCO is asking its customers to conduct tests on its equipment.  Do you not have a QA department anymore?  I began another thread because I believe I am seeing similar issues.  When my network activity peaks, I see multiple issues with DNS and data transfers.  Here is a link to the thread.

https://supportforums.cisco.com/thread/2036350

I think the tests run by Mr. Jurkovic are very conclusive.  The router struggles when throughout peaks.  While speed limitation can be tolerated, the jitters and packet loss at peak speeds are SEVERE ISSUES.  It is affecting my entire office.

I performed a very basic test.  I removed the unit from my network and my issues went away.

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: