[plug] IDE-TAPE

[Onno Benschop]

On Mon, 2003-09-01 at 19:30, Craig Ringer wrote:
> > No, but thanks any way.
> > 
> > Any other takers?


While I'm replying to Craig's message, I'll incorporate responses to
those questions supplied by others below also.

> In dmesg, what /does/ it say for the disk device?

Nada, niente, nichts, nothing. From dmesg:

        ide0: BM-DMA at 0xffa0-0xffa7, BIOS settings: hda:DMA, hdb:DMA
        hda: MAXTOR 6L040J2, ATA DISK drive
        hdb: ST380011A, ATA DISK drive
        blk: queue f88276c0, I/O limit 4095Mb (mask 0xffffffff)
        blk: queue f88277fc, I/O limit 4095Mb (mask 0xffffffff)
        ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
        Journalled Block Device driver loaded
        hda: attached ide-disk driver.
        hda: host protected area => 1
        hda: 78177792 sectors (40027 MB) w/1819KiB Cache,
        CHS=77557/16/63, UDMA(33)
        hdb: attached ide-disk driver.
        hdb: host protected area => 1
        hdb: 156301488 sectors (80026 MB) w/2048KiB Cache,
        CHS=9729/255/63, UDMA(33)
        Partition check:
         /dev/ide/host0/bus0/target0/lun0: p1 p2 p3 p4 < p5 p6 p7 >
         /dev/ide/host0/bus0/target1/lun0: unknown partition table
        

As you can see, it's also not showing the second channel. 

Interestingly, a look through /var/log with grep -r -i ide1 shows in
syslog, kern.log and messages:

        ide1 at 0x170-0x177,0x376 on irq 15

This was previously hosting a CDRom drive which worked - see below.

> If you run
>    hdparm -adcu /dev/hdc

This doesn't work because it doesn't know about /dev/hdc

        /dev/hdc: No such device or address

> I find it interesting that you mention that the BIOS does show it. It's 
> awfully strange to have /nothing/ show up in dmesg in that case.

As I wrote originally, that's what the installer (a tech from the
supplier who was briefed and spoken to) reported to me directly on the
phone.

> When you used ide-scsi, I presume you (a) loaded the ide-scsi module as 
> well as the st and sg (sometimes needed) modules, and (b) were accessing 
> it as /dev/st0 not /dev/mt0 ?

I did - as far as I recall, but I still am of the mind that if it's not
happening in dmesg, it's not happening at all... As I understand it, in
order of detection, BIOS => dmesg => /proc => /modules => working

> Yes, this is obvious question town, but I'm digging for ideas here.

Sure.

On Mon, 2003-09-01 at 19:36, Steve Boak wrote: 
> It IS plugged in, isn't it, both IDE and power?

That's what I've been told.

On Mon, 2003-09-01 at 19:41, John Clayton wrote: 
> This may sound silly but when you detect ide devices in the bios it can
> give you the option of skipping the entry therefore leaving it blank if it
> was blank before.

Dunno. I'm in Kununurra and AFAIK I cannot tweak the BIOS without a tech
onsite :-( See below.

> Has you friend made sure that the device is actually
> enabled in the bios and not just seen that it is detected in the bios?
> Does it show up in the initial screen that shows processor type/speed,
> memory drives and the various irq entries? If not then nothing will show
> up in dmesg.

Also, dunno. As I said earlier, I don't actually think that a BIOS
detection is related to a dmesg.

This did spark another investigation. I looked at /proc/interrupts which
reports:

                   CPU0       
          0:    1353560          XT-PIC  timer
          1:          2          XT-PIC  keyboard
          2:          0          XT-PIC  cascade
          8:          3          XT-PIC  rtc
         10:       1599          XT-PIC  eth0
         14:       6138          XT-PIC  ide0
        NMI:          0 
        LOC:    1353513 
        ERR:          0
        MIS:          0

On Mon, 2003-09-01 at 19:51, Paul Arch wrote: 
> My understanding was linux accesses ide devices directly (hdds) without
> using the bios ? Is this the same for cd drives and tape drives? My
> experience is that if I have a hard disk disabled in the bios, linux will
> still find it.

That is my understanding also.

> Steves suggestion sounds pretty obvious, so I might make another : Are the
> master/slave jumpers set correctly for the device ? Also, is the tape drive
> on a standard IDE controller chipset ( my dual celery 433 has two ide
> controllers, one 'standard' UDMA 33 and a highpoint UDMA 66) - hdd's on the
> highpoint driver appear on /dev/hdg and /dev/hdh

AFAIK, the installer removed the CDRom drive from the case (which was
working and installed on channel 2 as a master, then plugged the tape
drive into the socket that became available. The CDRom had to go,
because there was not enough space in the case (low profile SVEC - one
5.25" bay)



Unless anyone has any other "ah-ha" moments, I'll see if I can persuade
someone on-site to have a squiz at the BIOS with me looking over their
shoulder. After that I'm thinking that I'll need to call the installer
again.

Any other suggestions?

PS. Thanks for the wide-ranging approaches so-far.

Onno Benschop 

Connected via Optus B3 at S15:51'18" - E128:45'05" (Crossing Falls,
Kununurra, WA)
-- 
()/)/)()        ..ASCII for Onno.. 
|>>?            ..EBCDIC for Onno.. 
--- -. -. ---   ..Morse for Onno.. 

Proudly supported by Skipper Trucks, Highway1, Concept AV, Sony Central,
Dalcon
ITmaze - ABN: 56 178 057 063 - ph: 04 1219 8888 - onno at itmaze dot com
dot au