I am getting ready to pull my hair out on this one, I bought a nice 4u rackmount server to donate to a small school and I was going to put linux on it to run a nice squirrelmail server but I can't get the scsi controller card to work, this system has a onboard adaptec aic7896 scsi controller on it, this system worked with win 2k server so I think the drives are teminated correctly, I have tryed loading rh 7.1 7.2 7.3, md 8.1 8.2 on this system redhat 7.1 and 7.2, and md 8.1 all freeze right at the begging of the install, if you look at the kernal messages it just keeps looping and saying the device is not ready and it is reseting, but on redhat 7.3 the whole install goes fine and then on the first reboot when it trys to load the adaptec driver it keeps looping saying device offline and some other error messages, I can't capture the messages to post here. I have searched the internet for any fix and have not found any as of yet, I am hoping that someone here will have a direction for me to go, thank you for any help.

I'm willing to bet the devices aren't terminated correctly. That's one of the "old reliable" cards from nearly the 2.0 days. Maybe Windows was handling it in software. What does the SCSI bios say?

Also, in the displayed dmesg, how far does it get with device registering on the scsi chain?

Cheers,

Finegan

In looking in the scsi bios it has a option called host adapter scsi termination and it is set to enabled on both a "a" and "b" channels, everything else seems fine on that.

What is the dmesg? and how do I see about the device registering on scsi chain?

also here is some more info I forgot to include in my first post:
There is only one device connected to the scsi controller and that is a 18 gig scsi drive it is at scsi id#1 on the "a" channel and the scsi host adapter is at id#7

also I am running lilo as the boot manager.

dmesg is a command to see the kernel boot record, which is what scrolls past as the machine is booting. By the scsi device chain, I meant what gets reported into dmesg when the scsi bus gets scanned, usually looks something like this:

scsi1 : sym53c8xx-1.7.3c-20010512
Vendor: SONY Model: CD-R CDU948S Rev: 1.0h
Type: CD-ROM ANSI SCSI revision: 02
Attached scsi CD-ROM sr0 at scsi1, channel 0, id 0, lun 0
sym53c810a-0-<0,*>: FAST-10 SCSI 10.0 MB/s (100.0 ns, offset 8)
sr0: scsi-1 drive

or:

i91u: PCI Base=0xB400, IRQ=11, BIOS=0xFF000, SCSI ID=7
i91u: Reset SCSI Bus ...
scsi0 : Initio INI-9X00U/UW SCSI device driver; Revision: 1.03g
Vendor: Model: CD-R/RW RW7060S Rev: 1.30
Type: CD-ROM ANSI SCSI revision: 02
Attached scsi CD-ROM sr0 at scsi0, channel 0, id 0, lun 0
sr0: scsi3-mmc drive: 24x/24x writer cd/rw xa/form2 cdda tray


Go figure, I have a symantec and an initio, the only two NOT adaptec scsi card manufacturers.

These are with the drives hooked straight up to the SCSI card, no terminators, but the cards can self terminate. Honestly, SCSI voodoo, just try different configurations and booting one of the RH kernels that would reset and when you get a configuration that doesn't reset you're good.

Cheers,

Finegan

So, same problem here. trying to install RH 8.0 on Dual P3 server w/Adaptec XX96 SCSI board. Goes in to loop looking at each scsi hd to find boot. i have only one hd. Finally locks after device #15. Very much a newbie here but need the following info. What should the scsi settings be? Should I install the smp kernel?


info: Premio 2ul Phoenix Dual P3
adaptec 7896 Dual Channel SCSI Board
1 GIG ram
1 IBM 18.6GB HD
RH 8.0 full retail version.

Any help would be appreciated.

Never mind. Got it working.

how did you get it working - I have the same prob.

You need to run the newest stable kernel, it was a kernel bug where it didn't reset the controller right. I used Redhat 9.0 and it picked it up correctly right out of the install.