Post reply

Strange SQL Cluster 2008 situation

  • September 29, 2011 at 8:03 am

    #1388381

    @perry Whittle

    The NIC is Broadcom NetExtreme, and yes the latest drivers have been installed.

    @mike-2 P Barron

    We do have separate FC switches, but I'm not sure I follow you right ... "managed out-of-band" ?

    I'm not English, so this might just be a 'language barrier' thing, but still ... care to explain in more detail?

    Thanks

  • September 29, 2011 at 8:11 am

    #1388387

    In this context I'm using the phrase out-of-band to mean; using the regular ethernet network to manage the SAN rather than over the fabric with the data. I've seen some NetApp (ONTAP) SANs that use this method.

    We had problems with teamed HP NICs as the 64-bit drivers to allow 10Gb support had issues. This was a few years ago now, so should have been addressed. It might be a problem if you're using drivers that were shipped with the card, as vendors rarely update the CDs.

  • September 29, 2011 at 8:19 am

    #1388392

    So you're definately using FC? What make/model of HBA do you have? Are those drivers up to date? Are you using multi-path I/O?

    I'm suggesting that there may be a problem with the FC switch (either hardware or its configuration). It could be a issue on the I/O path to node 1 for one or more of your host LUN maps. Who's looking after your storage?

Viewing 3 posts - 16 through 17 (of 17 total)

You must be logged in to reply to this topic. Login to reply