Turns out it was a multicast issue. Our network engineer discovered that ganglia was overwhelming the iDRAC (express) with multicast traffic. We disabled multicast and everything returned to normal. It seems that the iDRAC is running some Linux which is/was attempting to process the traffic.
Cheers, Ryan On Fri, Oct 2, 2009 at 11:51 AM, Ryan Dooley <[email protected]> wrote: > Turns out that Red Hat has an bug open on this for the e1000 driver > which was opened up earlier this year: > > https://bugzilla.redhat.com/show_bug.cgi?id=496248 > > > On Tue, Sep 29, 2009 at 5:26 PM, Ryan Dooley <[email protected]> wrote: >> Hi, >> I've a bunch of shiny new R710's complete with the IDRAC 6 Express and >> I'm having what look like vlan issues. Serial Over Lan consoles seem >> to work fine until the network driver for the bnx2 configures the >> interface and then it just up and stops responding. >> >> The host is on say vlan2 and the DRAC has been set to share eth0 on vlan102. >> >> I'm using the broadcom netxtreme driver on two different hosts: >> * 1.9.20b with a custom kernel 2.6.25.10 on one host >> * 1.9.3 with kernel 2.6.18-164.el5 on another >> >> Both hosts are essentially Centos 5. >> >> If the vlan is changed from 102 to another, say 106, things work fine. >> It is as if there is a bit mask issue in handling vlan tagging. >> >> The exact same configuration on the 2950 III platform doesn't have >> this issue (but it doesn't have the DRAC either, just a bare BMC). >> >> Any suggestions would be great. >> >> Cheers, >> Ryan >> > _______________________________________________ Linux-PowerEdge mailing list [email protected] https://lists.us.dell.com/mailman/listinfo/linux-poweredge Please read the FAQ at http://lists.us.dell.com/faq
