• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 1234
  • Last Modified:

Network Performance Help

I'm running out of ideas here.  I have a site that has a new 100meg Optiman pipe.  We recently changed all switch equipment to new Hp Procurves, old cisco router is still running the optiman

So, I have horrible network performance.  Less than 1 meg down (via speedtest) but I'm getting over 40 up.  A ping from the switch that links to the Cisco router ranges from 5-25ms (!!!!!).  I get about 5% Rx FCS errors.  We've replaced the patch cable.  I've hard set the ports to 100 FDX.  Utilization on the uplink port (from the HP side) Rx is 3% and Tx is 1%.

Any ideas?!
0
JamesonJendreas
Asked:
JamesonJendreas
  • 5
  • 2
1 Solution
 
JamesonJendreasAuthor Commented:
Hmmm.....
.
W 01/01/90 01:35:01 00328 FFI: port 36-Excessive jabbering/fragments. See Help
0
 
Robert Sutton JrSenior Network ManagerCommented:
Can you post the current firmaware version and alos let me know if you have "Jumbo Frames" enabled? Let us know.
0
 
JamesonJendreasAuthor Commented:
On the cisco side:
12.4

HP:
W.14.70

Jumbo frames are NOT enabled.  I do have flow control enabled (this solved large amounts of Tx errors)
0
Improve Your Query Performance Tuning

In this FREE six-day email course, you'll learn from Janis Griffin, Database Performance Evangelist. She'll teach 12 steps that you can use to optimize your queries as much as possible and see measurable results in your work. Get started today!

 
JamesonJendreasAuthor Commented:
From the switch log:

W 01/01/90 02:09:19 00328 FFI: port 36-Excessive jabbering/fragments. See Help
W 01/01/90 02:09:51 00328 FFI: port 36-Excessive jabbering/fragments. See Help
W 01/01/90 02:12:07 00328 FFI: port 36-Excessive jabbering/fragments. See Help
W 01/01/90 02:13:00 00328 FFI: port 36-Excessive jabbering/fragments. See Help
I 01/01/90 02:15:01 00077 ports: port 36 is now off-line
I 01/01/90 02:15:02 00435 ports: port 36 is Blocked by STP
I 01/01/90 02:15:04 00076 ports: port 36 is now on-line
W 01/01/90 02:15:16 00633 FFI: port 36 Duplex Mismatch. Reconfig port to Auto.
W 01/01/90 02:15:16 00329 FFI: port 36-Excessive CRC/alignment errors. See help.
W 01/01/90 02:15:48 00328 FFI: port 36-Excessive jabbering/fragments. See Help
I 01/01/90 02:18:51 00057 stp: Spanning Tree Protocol disabled


So I can confirm both sides are hard set 100mFdx - so the mismatch makes no sense to me.  I turned off STP for the time being on this switch.

Note, this is at 1 site of three.  All three are connected via Optiman circuit.
0
 
JamesonJendreasAuthor Commented:
Well, I cleared the stats about 10 minutes ago and now everything seems clean. I'm getting a speed test of over 30megs.  

Still getting a 5ms ping from the switch to the router, and they are directly connected.  I don't think the root issue is fixed at this point...
0
 
Reid PalmeiraTelecom EngineerCommented:
You might also want to check your 802.1q trunk ports on the switch. Maybe something is setup in portfast / edgeport mode when it shouldn't be, I've seen the "excessive jabbering" thing before on two switches that were incorrectly connected into a loop.
0
 
Robert Sutton JrSenior Network ManagerCommented:
What concerns me is that you see a "Duplex" mismatch and also the fact you have Spaning-tree disabled. Can you verify these settings?
0
 
JamesonJendreasAuthor Commented:
The Duplex mismatch is really weird - I have both sides hard set to 100FDX.  No auto-neg going on.

I disabled STP on the switch due to it shutting down the port.  This switch only has two links on it - 1 going to the router, and 1 going back to the MDF, so there is not redundant paths on the switch.  As soon as I can figure out what is going on with STP I'll turn it back on.

Also, I HAVE to use RSTP.  I cannot use standard STP, as it will break my phone system.  Since turning off RSTP a few hours ago, things are actually running much cleaner.  I'm looking at STP as part of the issue.

I don't see how I can have any redundant links - all my switches are connected via Fiber from MDF-IDF, and any IDF that has two switches (which is rare) are connected via a CX4 module.  I'm trying to think of a way that a loop could have been patched.  In fact, i intentionally patched a loop to show someone what would happen in that case - within a minute the whole network was down.

But maybe there is a loop.  Maybe I need to go back and re-evaluate how I have RSTP setup.
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

Join & Write a Comment

Featured Post

Improve Your Query Performance Tuning

In this FREE six-day email course, you'll learn from Janis Griffin, Database Performance Evangelist. She'll teach 12 steps that you can use to optimize your queries as much as possible and see measurable results in your work. Get started today!

  • 5
  • 2
Tackle projects and never again get stuck behind a technical roadblock.
Join Now