?
Solved

Procurve Errors, Can't force N IC speed on PC's.

Posted on 2007-07-26
11
Medium Priority
?
3,348 Views
Last Modified: 2012-06-21
We purchased a Procurve 2900 about half a year ago. Recently I started checking it for errors as we experience delays at times writing or reading data from the file server ( This office about 20 users, 40 systems, W2K AD Domain, 98% XP Pro clients ).  Under status I discovered half a dozen systems were throwing errors stating either 'High collision or drop rate on port ...'  or 'Excessive broadcasts detected on port ...'.  The High collision errors are far more common.  

I started troubleshooting thinking these might be the cause of the intermittent delays. My first thought was to force both ends of the connection to Gig instead of leaving them on 'Auto' however I found that NONE of the NIC's involved would allow you to force them to a speed above 100 megs. Also discovered that when I forced a client to 100 or 10 Megs that client would lose connectivity with the LAN. Running IPCONFIG would return 'media disconnected'. Rebooting did not resolve it, only going back to 'Auto' would get the system to admit it ( all of our systems ) have a static IP and then allow connectivity to the LAN.

The other thoughts on tracking the problem down include trying other ports on the switch, trying a 3rd party NIC in the PC's involved, and trying other patch cables.  

I'm a little puzzled about not being able to force a speed on the connection. Many years ago that was how we ensured the auto settings were not screwing up.

Might anyone have a work around, or other ideas on identifying the cause of these few systems continuing errors?


Thanks

Paul
0
Comment
Question by:Bluewhale042399
  • 3
  • 3
  • 2
  • +3
11 Comments
 
LVL 8

Expert Comment

by:Jim_Coyne
ID: 19574964
Set the switch to full duplex, leave the PC as auto

Speed 1000
duplex full
0
 

Author Comment

by:Bluewhale042399
ID: 19578161
Hey Jim. That's how they are now. PC's on auto, Procurve most on Auto but have tried a few at 'Auto-1000' as that's as close as the switch comes to allowing me to force the speed.


paul
0
 
LVL 17

Accepted Solution

by:
jburgaard earned 1200 total points
ID: 19578660
Prefered: auto-auto
If that does not work, then the fixed settings must be equal in both ends.
-as auto<->full will 'negotiate' to half<->full . Resulting in the errors described.

From ftp://ftp.hp.com/pub/networking/software/2900yl-MCG-0207-T_12_XX-Chap10-PrtStat-BasicCfg.pdf
switch-syntax: interface < port-list >
[speed-duplex < auto-10 |10-full | 10-half | 100-full | 100-half |auto| auto-
100 | 1000-full >]
(no auto-1000 as far as I can see)

Trying other patch cables =good idea
0
VIDEO: THE CONCERTO CLOUD FOR HEALTHCARE

Modern healthcare requires a modern cloud. View this brief video to understand how the Concerto Cloud for Healthcare can help your organization.

 
LVL 27

Expert Comment

by:pseudocyber
ID: 19578732
I've found, for GigE, both sides HAVE
0
 
LVL 27

Expert Comment

by:pseudocyber
ID: 19578737
Sorry.

I've found, for GigE, both sides HAVE to be Auto.
0
 
LVL 8

Assisted Solution

by:Jim_Coyne
Jim_Coyne earned 400 total points
ID: 19579232
High collision errors are an indication of duplex mismatch (not speed)

http://en.wikipedia.org/wiki/Duplex_mismatch
0
 
LVL 8

Expert Comment

by:Jim_Coyne
ID: 19579246
You'll need to set the switch to 1000-half if you continue to get High collision errors w/ 1000-full
0
 
LVL 2

Assisted Solution

by:mkurtzhals
mkurtzhals earned 400 total points
ID: 19582340
Gig does not function @ half duplex.
0
 

Author Comment

by:Bluewhale042399
ID: 19585799
I'll try to set both sides for 100 if the other ideas don't pan out.
0
 
LVL 4

Expert Comment

by:Mark Walden
ID: 19592991
Do you have any dumb hubs in the bldg.  because some hubs operate at 10/half they will cause a lot of collision.  See if you have any hubs hanging of the ports with high collisions.
0
 

Author Comment

by:Bluewhale042399
ID: 19595612
Turns out the switch is bad, possibly deteriorating. Moved one of the trouble PC's to another port and no errors. 2 daze later moved three more, same thing. Called HP who is overnighting a replacement out. FWIW the HP tech indicated that per RFC when using Gig ehternet at least one end of the connection must be set to 'auto'.

Thanks for the suggestions tho.


Paul
0

Featured Post

What is SQL Server and how does it work?

The purpose of this paper is to provide you background on SQL Server. It’s your self-study guide for learning fundamentals. It includes both the history of SQL and its technical basics. Concepts and definitions will form the solid foundation of your future DBA expertise.

Question has a verified solution.

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

This article will inform Clients about common and important expectations from the freelancers (Experts) who are looking at your Gig.
This program is used to assist in finding and resolving common problems with wireless connections.
There's a multitude of different network monitoring solutions out there, and you're probably wondering what makes NetCrunch so special. It's completely agentless, but does let you create an agent, if you desire. It offers powerful scalability …
In this video we outline the Physical Segments view of NetCrunch network monitor. By following this brief how-to video, you will be able to learn how NetCrunch visualizes your network, how granular is the information collected, as well as where to f…

807 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