Vijay, Since it has to be handled at SCSI (and it is a SCSI sense), shouldn't we be retrying it at SCSI level. James, what is your opinion ? It looks good to me as the effect is local to LSI rdac storage. chandra On Mon, 2010-08-02 at 13:07 +0530, Chauhan, Vijay wrote: > Hi, > > This patch adds retry for the IO returned with 06/3f/03((INQUIRY_DATA_CHANGED)) sense code in rdac_check_sense(). IO returned with 06/3f/03 from controller are currently failed by scsi mid layer, as a reason momentarily path failure is noticed by DM multipath. > > Signed-off-by: Vijay Chauhan<vijay.chauhan@xxxxxxx> > Reviewed-by: Babu Moger <babu.moger@xxxxxxx> > Reviewed-by: Bob Stankey <Robert.stankey@xxxxxxx> > --- > > diff -uprN linux-2.6.35-rc6-orig/drivers/scsi/device_handler/scsi_dh_rdac.c linux-2.6.35-rc6/drivers/scsi/device_handler/scsi_dh_rdac.c > --- linux-2.6.35-rc6-orig/drivers/scsi/device_handler/scsi_dh_rdac.c 2010-07-22 15:13:38.000000000 -0400 > +++ linux-2.6.35-rc6/drivers/scsi/device_handler/scsi_dh_rdac.c 2010-07-27 12:13:58.000000000 -0400 > @@ -738,6 +738,11 @@ static int rdac_check_sense(struct scsi_ > * Quiescence in progress , just retry. > */ > return ADD_TO_MLQUEUE; > + if (sense_hdr->asc == 0x3f && sense_hdr->ascq == 0x03) > + /* > + * INQUIRY DATA has changed, retry again. > + */ > + return ADD_TO_MLQUEUE; > break; > } > /* success just means we do not care what scsi-ml does */ > > -- -- dm-devel mailing list dm-devel@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/dm-devel
