have you tried disconnecting and reconnecting the wire? sometimes this will
clear the problem (or at least generate something in the logs to help you figure
it out)
David Lang
On Thu, 6 Feb 2014, john boris wrote:
Date: Thu, 6 Feb 2014 16:04:08 -0500
From: john boris <jbori...@gmail.com>
To: Robert Lanning <lann...@lanning.cc>
Cc: "tech@lists.lopsa.org" <tech@lists.lopsa.org>
Subject: Re: [lopsa-tech] Change NIC settings on HP dl180G5 NC105 Card under
ubuntu 10.04
Robert,
Well I will be honest I am not sure how to set the ethtool settings back to
default. It seems the NIC doesn't like any commands I give it. I tried
sudo ethtool -r eth0
to restart the negotiation and that did not help.
On Thu, Feb 6, 2014 at 3:51 PM, Robert Hajime Lanning <lann...@lanning.cc>wrote:
First clear all of the ethtool settings back to their default.
GigE to GigE autonegotiations work just fine. Even across vendors.
Auto-neg was designed into the protocol from the start.
The 100Base-T and lower specs had N-Way autonegotiation tagged on after
the fact and so had quite a few compatibility issues.
GigE also will fix cross-over situations on the fly. Though there is
still nothing it can do about split pairs in the cable. If the pairing is
kept correct it figures out which pair is what, on its own.
So, make sure autoneg is on for both sides of the connection (this is
default for all devices I have ever used (Cisco,Juniper,HP,Dell,Intel...))
I might be, that with the previous switch, the server had to be hardcoded
to 100Mb Full-Duplex, because the switch was 100Mb and its N-Way neg failed.
This was a big issue with Sun HME vs. Cisco. When we went GigE, and
remembered to clear all the hardcodings, everything worked fine.
On 02.06.2014 12:31, john boris wrote:
First thanks for the reply.
Where the server sits the NOC had an upgrade and they went to all new
gigabit switches. So when I try to do any transfers from this server
(or from any of the VMs on the server) I get speeds that make a 56K
modem look like lightspeed. The server is not overloaded as there is
only one production VM on the system that is just doing a bunch of
telnet (well sssh ) sessions. The load average on the VM never goes
above 0.00.
I asked our NOC folks and they said to make sure the NIC is set to
full Duplex. So I started checking and saw the card was set to 100mb
and half duplex. Thus started my journey. I have been on with HP for
the past 1.5 hours while they step me through all the things I have
tried. I was hoping there was someone out there that might have come
across this already.
--
Mr. Flibble
King of the Potato People
http://www.linkedin.com/in/RobertLanning
_______________________________________________
Tech mailing list
Tech@lists.lopsa.org
https://lists.lopsa.org/cgi-bin/mailman/listinfo/tech
This list provided by the League of Professional System Administrators
http://lopsa.org/
_______________________________________________
Tech mailing list
Tech@lists.lopsa.org
https://lists.lopsa.org/cgi-bin/mailman/listinfo/tech
This list provided by the League of Professional System Administrators
http://lopsa.org/
_______________________________________________
Tech mailing list
Tech@lists.lopsa.org
https://lists.lopsa.org/cgi-bin/mailman/listinfo/tech
This list provided by the League of Professional System Administrators
http://lopsa.org/
