[prev in list] [next in list] [prev in thread] [next in thread] 

List:       linux-net
Subject:    Re: bnx2 rx_filtered_packets statistic?
From:       Charlie Brady <charlieb () budge ! apana ! org ! au>
Date:       2010-07-02 14:45:15
Message-ID: Pine.LNX.4.64.1007021042310.31895 () e-smith ! charlieb ! ott ! istop ! com
[Download RAW message or body]


On Fri, 2 Jul 2010, Brian Haley wrote:

> On 07/01/2010 11:22 PM, Charlie Brady wrote:
> >> "Filtered packets are dropped by the hardware when they don't match the
> >> L2 perfect match filters, mcast hash filters and broadcast filter.
> >> These filters are programmed based on the NIC's MAC address, multicast
> >> list from the stack and broadcast is always enabled."
> >>
> >> Under normal operation I believe the counter will go up...
> > 
> > What would cause it to go up in normal operation?
> 
> I think it's going to depend on your switch, I have a p-t-p connection that
> shows zero, but one to a switch showing 20K, so I'd assume it's periodically
> going into flood mode when systems go up/down, etc.

OK, thanks.

> >> Are you seeing any particular type of packets being dropped?  For example,
> >> can you send/receive both IPv4 and IPv6 "pings" on that NIC?
> > 
> > I don't know what is being dropped. I do know that cluster heartbeat 
> > traffic between the two nodes sometimes stops, leading to cluster failover 
> > and fencing. I don't know what is causing the traffic stalls, but NIC 
> > driver issues is high on the suspicion list. Platform is CentOS 5, not 
> > fully updates - kernel is 2.6.18-128.1.6.el5.
> 
> There have been a number of updates to the driver since then...

Yeah, I know. I have inertia to overcome to get an update in. It will help 
if I can find another problem report which matches the symptoms.

---
Charlie
--
To unsubscribe from this list: send the line "unsubscribe linux-net" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
[prev in list] [next in list] [prev in thread] [next in thread]