Topic: Configuring an Exabyte VXA Autopak with SME 5.1.2

[Mark Baumann-Erb]

Hi,

I am a newbie to SME Server and have done a bit of searching in the forums and how-to's but have yet to resolve my problem.  I have a Compaq Proliant ML350 with a Compaq Ultra3 SCSI card (essentially an Adaptec SCSI) with version 5.1.2, Update 3 of SME Server. I am attempting to connect an Exabyte VXA AutoPax 1x10 Autoloader (which contains an Excrix VXA-1 tape drive.)  After I connected the drive to the external SCSI port and booted, the BIOS reports both devices on the proper SCSI port on the card.  In looking at the messages log file during the bootup, it appears that SME Server also sees both devices:\
Aug 26 11:06:12 RBS kernel: (scsi0) found at PCI 0/2/0
Aug 26 11:06:12 RBS kernel: (scsi0) Wide Channel A, SCSI ID=7, 32/255 SCBs
Aug 26 11:06:12 RBS kernel: (scsi0) Downloading sequencer code... 396 instructions downloaded
Aug 26 11:06:13 RBS kernel: (scsi1) found at PCI 0/2/1
Aug 26 11:06:13 RBS kernel: (scsi1) Wide Channel B, SCSI ID=7, 32/255 SCBs
Aug 26 11:06:13 RBS kernel: (scsi1) Downloading sequencer code... 396 instructions downloaded
Aug 26 11:06:13 RBS kernel: scsi0 : Adaptec AHA274x/284x/294x (EISA/VLB/PCI-Fast SCSI) 5.1.33/3.2.4
Aug 26 11:06:13 RBS kernel:        
Aug 26 11:06:13 RBS kernel: scsi1 : Adaptec AHA274x/284x/294x (EISA/VLB/PCI-Fast SCSI) 5.1.33/3.2.4
Aug 26 11:06:13 RBS kernel:        
Aug 26 11:06:13 RBS kernel: scsi : 2 hosts.
Aug 26 11:06:13 RBS kernel:   Vendor: EXABYTE   Model: VXA AutoPak 1x10  Rev: E33a
Aug 26 11:06:13 RBS kernel:   Type:   Medium Changer                     ANSI SCSI revision: 02
Aug 26 11:06:13 RBS kernel:   Vendor: ECRIX     Model: VXA-1             Rev: 2B7B
Aug 26 11:06:13 RBS kernel:   Type:   Sequential-Access                  ANSI SCSI revision: 02

and the aic7xxx and st modules are being loaded:

[root@RBS /root]# lsmod
Module                  Size  Used by
ppp_mppe               15440   2  (autoclean)
ppp                    21296   2  (autoclean) [ppp_mppe]
slhc                    4528   0  (autoclean) [ppp]
printer                 4672   0  (unused)
appletalk              18624  12
ipsec                 241968   2
3c90x                  22656   1  (autoclean)
eepro100               17392   1  (autoclean)
ip_masq_vdolive         1536   0  (unused)
ip_masq_rtsp            4960   0  (unused)
ip_masq_pptp            4560   0  (unused)
ip_masq_irc             2112   0  (unused)
ip_masq_icq            13920   0  (unused)
ip_masq_h323            7616   0  (unused)
ip_masq_ftp             3776   0  (unused)
ip_masq_cuseeme         1248   0  (unused)
st                     25520   0  (unused)
usb-ohci               13344   0  (unused)
usbcore                45104   1  [printer usb-ohci]
cpqarray               15824   3
aic7xxx               132960   0  (unused)
[root@RBS /root]#

Finally, if I attempt to check the status of the tape drive, that appears to work:

[root@RBS /root]# mt -f /dev/nst0 status
SCSI 2 tape drive:
File number=-1, block number=-1, partition=0.
Tape block size 1024 bytes. Density code 0x80 (unknown to this mt).
Soft error count since last status=0
General status bits on (1010000):
 ONLINE IM_REP_EN
[root@RBS /root]#

Yet, when I attempt to perform a backup, the system freezes (including the console) and I have to do a hard reboot on the system.  The logs for flexbackup show nothing unusual:

[root@RBS flexbackup]# more all.0.20020825
|------------------------------------------------
| Doing level 0 backup of all using dump
| Retensioning tape...
| Rewinding & erasing tape...
| Creating index key 200208250901.11
| Tape #0
| Filesystems = / /boot
|------------------------------------------------
At block 0.
|------------------------------------------------
| File number 1, index key 200208250901.11
| Backup of: /
| Date of this level 0 backup: Sun Aug 25 09:01:59 2002
| Date of last level 0 backup: the epoch
|------------------------------------------------
| (dump -0 -b 10 -a -f - / | gzip -4) | buffer -m 3m -s 10k -u 100 -t -p 75 -B \
|  -o /dev/st0
|------------------------------------------------
  DUMP: Date of this level 0 dump: Sun Aug 25 09:01:59 2002
  DUMP: Date of last level 0 dump: the epoch
  DUMP: Dumping /dev/ida/c0d0p6 (/) to standard output
  DUMP: Label: none
  DUMP: mapping (Pass I) [regular files]
  DUMP: mapping (Pass II) [directories]
  DUMP: estimated 399299 tape blocks.
  DUMP: Volume 1 started at: Sun Aug 25 09:02:44 2002
  DUMP: dumping (Pass III) [directories]
  DUMP: dumping (Pass IV) [regular files]
[root@RBS flexbackup]#
                                                                               
and nothing unusualy appears in the messages log file - it just stops after dumping mysql tables:

Aug 26 08:48:49 RBS e-smith[2476]: Processing event: mysql-delete-dumps
Aug 26 08:48:49 RBS e-smith[2476]: Running event handler: /etc/e-smith/events/mysql-delete-dumps/S10mysql-delete-dumped-tables
Aug 26 08:48:49 RBS e-smith[2476]: S10mysql-delete-dumped-tables=action|Event|mysql-delete-dumps|Action|S10mysql-delete-dumped-tables|Start|1030366129 459128|End|1030366129 485837|Elapsed|0.026709
Aug 26 08:48:49 RBS e-smith[2480]: Processing event: mysql-dump-tables
Aug 26 08:48:49 RBS e-smith[2480]: Running event handler: /etc/e-smith/events/mysql-dump-tables/S10mysql-dump-tables
Aug 26 08:48:49 RBS e-smith[2480]: S10mysql-dump-tables=action|Event|mysql-dump-tables|Action|S10mysql-dump-tables|Start|1030366129 513073|End|1030366129 685685|Elapsed|0.172612
Aug 26 08:57:54 RBS syslogd 1.3-3: restart.

I have attempted doing the backup on both the /dev/nst0 and /dev/st0 devices with equal failure.  Any help or suggestions would be greatly appreciated.

Thanks!
Mark

[Boris]

Try to change settings in the SCSI BIOS to disable WIDE for tape drive. Its still plenty of speed, but puts it in more compatible mode. I am not about arguing whether VXA-1 supports wide SCSI (per their manual), but it often freeze on ALL my servers then wide is enabled and works great then its disabled. We had this problem regardless of OS or Backup software used (NT, NW, Linux, ArcServe, Tapeware)
Good Luck.

[bart]

Mark Baumann-Erb wrote:


Finally, if I attempt to check the status of the tape drive, that appears to work:

[root@RBS /root]# mt -f /dev/nst0 status
SCSI 2 tape drive:
File number=-1, block number=-1, partition=0.
Tape block size 1024 bytes. Density code 0x80 (unknown to this mt).
Soft error count since last status=0
General status bits on (1010000):
ONLINE IM_REP_EN



Now, that tells me the tape does _not_ work.
It should state "File number=0, block number=0, partition=0."
(at least, when a blank tape is inserted)

Also, you could try [ mt -f /dev/nst0 tell ]

Regards, Bart

[Mark Baumann-Erb]

I have resolved the problem I was experiencing with the tape drive.  The problem turned out to be the on-board SCSI provided by Compaq is incompatible with the drive.  The incompatibility arises because the built-in SCSI supports RAID which is will not work with the Exabyte autoloader/VXA tape drive.  I purchased an Adaptec 29160 SCSI adapter and installed it in the system and the tape drive worked great (transfer rate up to 1990 Kb/sec so far.)

Mark