Re: scsi bus and tape backup questions
Bill wrote:
> greetings all:
>
> i have a couple of questions i hope someone can answer for me.
>
> to start: i have a sparc station and linux (debian 2.1, 2.0.35 kernel)
>
> 1st question - re: scsi
>
> my system works pretty good. i am only curious as to why the sparc
> scsi bus locates the internal system hard drive as target 3 when i have
> external hard drives, as it is physically closer to the system. i.e.
>
> esp0: FAST chip is fas100a (family=0, version=2)
> ESP: Total of 1 ESP hosts found, 1 actually in use.
> scsi0 : Sparc ESP100A-FAST
> scsi : 1 host.
> Vendor: SEAGATE Model: ST15230N Rev: 0638
> Type: Direct-Access ANSI SCSI revision: 02
> Detected scsi disk sda at scsi0, channel 0, id 0, lun 0
>
> Vendor: SEAGATE Model: ST15230N Rev: 0638
> Type: Direct-Access ANSI SCSI revision: 02
> Detected scsi disk sdb at scsi0, channel 0, id 2, lun 0
>
> Vendor: SEAGATE Model: ST1480 SUN0424 Rev: 8628
> Type: Direct-Access ANSI SCSI revision: 02
> Detected scsi disk sdc at scsi0, channel 0, id 3, lun 0
>
> Vendor: HP Model: C1533A Rev: 9503
> Type: Sequential-Access ANSI SCSI revision: 02
> scsi : detected 3 SCSI disks total.
>
> sda and sdb are in an external box (scsi id 2) whereas the small
> internal is sdc. i am just curious why the sparc does not find and
> id the internal first, i.e. target 0 or 1; sda. ???
Well, the disks are iterated by SCSI ID, which is usually set by jumpers on the drive
to a number between 0 and 7. Physical distance is irrelevant and not really knowable
by the SCSI adapter or the OS. For some historical reason Sun boxes have always (all
the one's I've ever encountered, from Sparc 2s up through Ultras) had their internal
drives with the OS installed as SCSI ID 3.
> 2nd question: re: tape backups.
>
> i am trying to finess mt, dump, and restore to do the job for us.
> they work o.k., except that there are a few main information bits
> that seem to be missing. i went through the mans on each of
> these, and i can't see anywhere an operation that will tell me how
> much space is left on the tape.
>
> GNU mt version 2.4.2
>
> Package: dump
> Version: 0.4b4-13
Unfortunately there isn't really a way for the device or the software to know how much
space is left on the tape. The drive can't find this out. I've done the same kind of
thing as you. I've written a perl script which takes care of keeping track what's on
the tape (the script labels the tape by keeping a record at the end of what dump was
made when) and puts multiple backups on the same tape so I only have to change it once
a week. I'm sending this along in a separate email and you may do with it what you
wish.
> 3rd question: cdrom drive:
>
> i also have a cdrom drive ii want to put on the system. whenever i
> add it to the chain, however, i get a scsi bus interrupt; reset infinite
> loop when i try to boot.
>
> here's how my system looks
>
> as above for the hard drives. external case with hard drives: id 2
>
> plus: hp tape backup: scsi id 4
>
> this configuration works fine. but when i add the cdrom to the end
> of the chain, i have the problem. the last device is terminated
> properly.
>
> with respect to scsi types;
>
> sparc - (fast scsi 2) --> external drives (scsi) --> hp tape (scsi)
> --> cdrom (sun - fast scsi 2)
>
> i tried setting the cdrom as device 5 and 6, neither worked.
>
> however, the nvram probe-scsi util did find them all properly and
> it looked fine in that setting. it was only when i went to boot when
> it went into an infinite loop.
>
> could it be a driver problem? i have the sg and st modules loaded.
I doubt it's a driver problem but how far does the boot get? What's the last thing
that happens. I'm still going to guess that this is a termination problem. Is this an
external cdrom in its own enclosure? Is your chain too long (can't remember the limit
offhand).
> thanks for any assistance with this.
Sure. Good luck.
--
Jens B. Jorgensen
jjorgens@bdsinc.com
Reply to: