Gerrard Geldenhuis wrote:
I have an HP server with a onboard broacom network card and an pci intel network card.This uses bnx2(broadcom) and e1000 (intel) drivers.
I had a similar problem recently. Fail-over from one NIC to the other was unreliable. Sometimes I get a few packets lost and then ping would resume. On more than one occasion fail-over never happened. The server in question for me was an IBM x3850 with two Broadcom NICs onboard using the tg3 drivers and six Intel NICs using the e1000 drivers. RHEL 5.1 was fully up to date. An interesting observation here was that bonding never worked reliably if I had a mixed bag of slave devices, i.e. one Broadcom and one Intel. Alternatively, I tried using Intel NICs for both slaves with similar results. Fail-over worked as expected only when I used both the onboard Broadcom NICs as slaves. I think it's a problem with the Intel drivers. I tried again on a different but identical server earlier today with the same results. Has anybody else noticed something along the same lines? Should I file a bugzilla for this?
-- Regards, विवेक ज. पाटणकर (Vivek J. Patankar) Registered Linux User #374218 Fedora release 8 (Werewolf) Linux 2.6.23.8-63.fc8 x86_64
signature.asc
Description: OpenPGP digital signature
_______________________________________________ rhelv5-list mailing list [email protected] https://www.redhat.com/mailman/listinfo/rhelv5-list
