Problem with Adaptec 29320LPE

2008-11-24 Thread Greg Byshenk
Is there a problem with the Adaptec 29320LPE (PCIe x1, single-channel Ultra320)
SCSI controller under FreeBSD 7?

I've recently received a server with this controller, which is intended to be
used to connect to Sony AIT tape libraries for backup. Unfortunately, it does
not seem to function properly.

It sees the connected devices without any difficulty, but fails to write to
any connected drives, and produces very strange errors when attempting to
address the libraries. That is, when attempting to write to a drive, the
drive is seen as present, but any attempt actually to write results in an
error (an end of tape is reported) without any data being written (mt status
reports the tape at File Number 0, Record number 0).

Additionally, attempting to address the changers produces erratic results.
Sometimes, the result is normal, but at other times the results are garbled,
and syslog reports a string of errors from the controller, followed by a
long string of errors on 'ch' (see below).

I am reasonably certain that the errors are not related to the tape
libraries, as a) the libraries worked normally on the old server, and
b) after installing a different controller (Adaptec 29160), the libraries
function properly on the new machine. And I am reasonably sure that the
problem is not a 320/160 problem, as setting the new controller to 160 in
the BIOS does not help.

The system is currently running FreeBSD 7.1-PRERELEASE: Wed Nov 19 11:33:15
CET 2008, from sources csup'ed immediately prior to the build. The kernel
is very close to GENERIC, but with various cardbus, wlan, and usb support
removed.

Searching has indicated some similar-looking errors reported, but all from
rather a long time ago (2000-2002).



backuphost# camcontrol devlist
SONY LIB-162 0208at scbus0 target 0 lun 0 (pass0,ch3)
SONY SDX-1100 0102   at scbus0 target 1 lun 0 (sa3,pass1)
SONY LIB-162 0203at scbus0 target 2 lun 0 (pass2,ch4)
SONY SDX-900V 0102   at scbus0 target 3 lun 0 (sa4,pass3)
AMCC 9650SE-16M DISK 3.06at scbus1 target 0 lun 0 (da0,pass4)
AMCC 9650SE-16M DISK 3.06at scbus1 target 0 lun 1 (da1,pass5)

backuphost# chio -f /dev/ch2 status
picker 0:
slot 0: ACCESS,FULL
slot 1: ACCESS,FULL
slot 2: ACCESS,FULL
slot 3: ACCESS,FULL
slot 4: ACCESS
slot 5: ACCESS,FULL
slot 6: ACCESS,FULL
slot 7: ACCESS,FULL
slot 8: ACCESS,FULL
slot 9: ACCESS,FULL
slot 10: ACCESS,FULL
slot 11: ACCESS,FULL
slot 12: ACCESS,FULL
slot 13: ACCESS,FULL
slot 14: ACCESS,FULL
slot 15: ACCESS,FULL
drive 0: ACCESS,FULL
backuphost# chio -f /dev/ch2 status
picker 0:
slot 8: ACCESS,FULL
slot 9: ACCESS,FULL
slot 10: ACCESS,FULL
slot 11: ACCESS,FULL
slot 12: ACCESS,FULL
slot 13: ACCESS,FULL
slot 14: ACCESS,FULL
slot 15: ACCESS,FULL
slot 8: ACCESS,FULL
slot 9: ACCESS,FULL
slot 10: ACCESS,FULL
slot 11: ACCESS,FULL
slot 12: ACCESS,FULL
slot 13: ACCESS,FULL
slot 14: ACCESS,FULL
slot 0:
drive 0: ACCESS,FULL
backuphost#


Nov 20 17:53:08 backuphost kernel: ahd0: Adaptec AIC7902 Ultra320 SCSI 
adapter port 0x4400-0x44ff,
0x4000-0x40ff mem 0xda60-0xda601fff irq 18 at device 4.0 on pci10
Nov 20 17:53:08 backuphost kernel: ahd0: [ITHREAD]
Nov 20 17:53:08 backuphost kernel: aic7902: Ultra320 Wide Channel A, SCSI Id=7, 
PCI-X 101-133Mhz, 51
2 SCBs
Nov 20 15:01:16 backuphost kernel: ahd0: Transmission error detected
Nov 20 15:01:16 backuphost kernel: LQISTAT1[0x0] LASTPHASE[0x40]:(P_DATAIN) 
SCSISIGI[0x40]:(P_DATAIN
)
Nov 20 15:01:16 backuphost kernel: PERRDIAG[0xd0]:(PARITYERR|HIPERR|HIZERO)
Nov 20 15:01:16 backuphost kernel:  Dump Card State Begins 

Nov 20 15:01:16 backuphost kernel: ahd0: Dumping Card State at program address 
0x3b Mode 0x22
Nov 20 15:01:16 backuphost kernel: Card was paused
Nov 20 15:01:16 backuphost kernel: INTSTAT[0x8]:(SCSIINT) SELOID[0x0] 
SELID[0x10] HS_MAILBOX[0x0]
Nov 20 15:01:16 backuphost kernel: INTCTL[0xc0]:(SWTMINTEN|SWTMINTMASK) 
SEQINTSTAT[0x10]:(SEQ_SWTMRT
O)
Nov 20 15:01:16 backuphost kernel: SAVED_MODE[0x11] 
DFFSTAT[0x19]:(CURRFIFO_1|FIFO0FREE)
Nov 20 15:01:16 backuphost kernel: SCSISIGI[0xb6]:(P_MESGOUT|REQI|BSYI|ATNI) 
SCSIPHASE[0x4]:(MSG_OUT
_PHASE)
Nov 20 15:01:16 backuphost kernel: SCSIBUS[0xc0] LASTPHASE[0x40]:(P_DATAIN) 
SCSISEQ0[0x0]
Nov 20 15:01:16 backuphost kernel: SCSISEQ1[0x12]:(ENAUTOATNP|ENRSELI) 
SEQCTL0[0x0] SEQINTCTL[0x0]
Nov 20 15:01:16 backuphost kernel: SEQ_FLAGS[0x20]:(DPHASE) SEQ_FLAGS2[0x0] 
QFREEZE_COUNT[0x40a]
Nov 20 15:01:16 backuphost kernel: KERNEL_QFREEZE_COUNT[0x40a] 
MK_MESSAGE_SCB[0xff00]
Nov 20 15:01:16 backuphost kernel: MK_MESSAGE_SCSIID[0xff] 
SSTAT0[0x2]:(SPIORDY) SSTAT1[0x11]:(REQIN
IT|PHASEMIS)
Nov 20 15:01:16 backuphost kernel: SSTAT2[0x0] SSTAT3[0x0] PERRDIAG[0x0] 
SIMODE1[0xac]:(ENSCSIPERR|E
NBUSFREE|ENSCSIRST|ENSELTIMO)
Nov 20 15:01:16 backuphost kernel: LQISTAT0[0x0] LQISTAT1[0x0] LQISTAT2[0x0] 
LQOSTAT0[0x0]
Nov 20 15:01:16 backuphost kernel: LQOSTAT1[0x0] LQOSTAT2[0x0]
Nov 20 15:01:16 backuphost kernel:
Nov 20 15:01:16

Re: Problem with Adaptec 29320LPE

2008-11-24 Thread Rink Springer
Hi Greg,

On Mon, Nov 24, 2008 at 12:42:49PM +0100, Greg Byshenk wrote:
 backuphost# camcontrol devlist
 SONY LIB-162 0208at scbus0 target 0 lun 0 (pass0,ch3)
 SONY SDX-1100 0102   at scbus0 target 1 lun 0 (sa3,pass1)
 SONY LIB-162 0203at scbus0 target 2 lun 0 (pass2,ch4)
 SONY SDX-900V 0102   at scbus0 target 3 lun 0 (sa4,pass3)
 AMCC 9650SE-16M DISK 3.06at scbus1 target 0 lun 0 (da0,pass4)
 AMCC 9650SE-16M DISK 3.06at scbus1 target 0 lun 1 (da1,pass5)

Are these volumes perhaps 2TB ? If so, it won't work...  we stumbled on
this at work a few weeks ago, and once we resized the volumes so that'd
all be 2TB, the controller worked fine...

As far as I know, this is the only workaround - I couldn't see relevant
patches in Open/NetBSD either that might have fixed this issue :-(

Regards,

-- 
Rink P.W. Springer- http://rink.nu
Anyway boys, this is America. Just because you get more votes doesn't
 mean you win. - Fox Mulder
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to [EMAIL PROTECTED]


Re: Problem with Adaptec 29320LPE

2008-11-24 Thread Greg Byshenk
On Mon, Nov 24, 2008 at 12:49:12PM +0100, Rink Springer wrote:
 Hi Greg,
 
 On Mon, Nov 24, 2008 at 12:42:49PM +0100, Greg Byshenk wrote:
  backuphost# camcontrol devlist
  SONY LIB-162 0208at scbus0 target 0 lun 0 (pass0,ch3)
  SONY SDX-1100 0102   at scbus0 target 1 lun 0 (sa3,pass1)
  SONY LIB-162 0203at scbus0 target 2 lun 0 (pass2,ch4)
  SONY SDX-900V 0102   at scbus0 target 3 lun 0 (sa4,pass3)
  AMCC 9650SE-16M DISK 3.06at scbus1 target 0 lun 0 (da0,pass4)
  AMCC 9650SE-16M DISK 3.06at scbus1 target 0 lun 1 (da1,pass5)

 Are these volumes perhaps 2TB ? If so, it won't work...  we stumbled on
 this at work a few weeks ago, and once we resized the volumes so that'd
 all be 2TB, the controller worked fine...
 
 As far as I know, this is the only workaround - I couldn't see relevant
 patches in Open/NetBSD either that might have fixed this issue :-(
 
The volume da1 is indeed 2TB, but it is not connected to the controller;
it (along with da0) is actually a RAID-10 array connected to a 3Ware/AMCC 
SATA controller.  The Adaptec contoller is used only for the tape drives
(the SDX-900V is AIT4; the SDX-1100 is AIT5), and they are 2TB.

-- 
greg byshenk  -  [EMAIL PROTECTED]  -  Leiden, NL
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to [EMAIL PROTECTED]