Dear fellow SASers, I wonder if I might pick your brains. I have an HP Storageworks Ultrium 1840 SAS tape drive and an LSI SAS9211-4i PCI Express SAS HBA which are behaving temperamentally. The SAS card uses the mpt2sas driver. There are no other devices connected to the SAS controller. After power-on the drive works swimmingly with dd, happily writing several GB of data before unpredictably aborting the write with EBUSY. All subsequent writes with dd return EBUSY yet mt operates just fine. Re-loading the tape or resetting the SAS bus with LSI's SAS2Flash utility make no difference. Rebooting clears the fault. The fault may clear itself if I wait anything from a few seconds to many minutes. Bizarrely sometimes it clears if I 'echo some_value > /proc/sys/dev/scsi/logging_level', although this only works a dozen or so times before it becomes ineffective. I'm using a plain-vanilla 3.10.28 kernel and the latest versions of the SAS card's BIOS (7.35.00.00) and firmware (18.00.00.00). I note that in 2010 HP claimed the tape drive won't work with many Linux SAS HBAs because their drivers were not reliable: http://h20566.www2.hp.com/portal/site/hpsc/template.PAGE/public/psi/advisoriesDisplay/?javax.portlet.begCacheTok=com.vignette.cachetoken&javax.portlet.endCacheTok=com.vignette.cachetoken&javax.portlet.prp_efb5c0793523e51970c8fa22b053ce01=wsrp-navigationalState%3DdocId%253Demr_na-c02236321-1%257CdocLocale%253Den_US&javax.portlet.tpst=efb5c0793523e51970c8fa22b053ce01&sp4ts.oid=4227371&ac.admitted=1391170950929.876444892.199480143 but I figure things have changed since then. Does anyone have any suggestions for debugging this problem? Yours, Larry. -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
