Hi Jon, Therefore I've various possibilities:
0 - To Check connections ( though the problem only occurs with the tape 04) 1 - To update the soft from my PowerVault ( from dell ) 2 - To Cleaning heads 3 - To change the tape 04 How can I check the state of de tape surface ( any command )? or to delete contents, re-initialize the tape... Best regards, ----- Original Message ----- From: "Jon LaBadie" <[EMAIL PROTECTED]> To: "Ra�l Wild-Spain" <[EMAIL PROTECTED]> Sent: Thursday, June 26, 2003 3:10 PM Subject: Re: Q: tape problem > On Thu, Jun 26, 2003 at 08:52:59AM +0200, Ra�l Wild-Spain wrote: > > Hi Jon, > > > > I add below notifications from my Debian ( log check). You can see many strange > > errors, specially "Jun 24 03:06:06 xaloc kernel: st0: Error on write filemark." > > repeatedly. > > > > I don't understand what happens ... > > > > > > > > 3. you have holding disk space available but have not set a chunk size > > > below the maximum file size limit of your holding disk. > > > > > > --- LOG CHECK --- > > > > > > Unusual System Events > > =-=-=-=-=-=-=-=-=-=-= > > Jun 24 03:06:06 xaloc kernel: st0: Error with sense data: Current st09:00: sense > > key Medium Error > > Jun 24 03:06:06 xaloc kernel: Additional sense indicates Excessive write errors > > Jun 24 03:06:06 xaloc kernel: st0: Error with sense data: Current st09:00: sense > > key Medium Error > > Jun 24 03:06:06 xaloc kernel: Additional sense indicates Excessive write errors > > Jun 24 03:06:06 xaloc kernel: st0: Error on write filemark. > > Jun 24 03:06:06 xaloc kernel: st0: Error with sense data: Current st09:00: sense > > key Medium Error > > > When you get errors like this, at the kernel level, it seldom (as in > virtually never) is due to the software running at the time. Problems > in the software generally cause things like core dumps and other > wierd things but without significant kernel messages. > > Most often it is in the driver or likely the hardware. My first thought > is, as always with SCSI devices, check connections, cables, terminators. > Nine-plus times out of ten it is that. > > Note too that last item "Medium Error". It could even be as simple as > dirty head or failed tape. > > -- > Jon H. LaBadie [EMAIL PROTECTED] > JG Computing > 4455 Province Line Road (609) 252-0159 > Princeton, NJ 08540-4322 (609) 683-7220 (fax)
