On Friday 25 March 2005 02:54 pm, Anders Norrbring wrote:

Hmm.. Perhaps that could be a thing. The controller is hooked up to a Compaq Proliant Rack with Compaq original disks, and they usually terminate correctly. In any case there are no options to do it in any other way. Perhaps I should try to play around with the termination options on the controller in its setup.

Any more ideas, anyone?

Anders.

I am using one of these cards in a SuperMicro case. I used a terminated cable to connect the backplane and flashed the firmware to the latest version. It has 5 Ultra 320 196G Cheetahs connected to it that have been working well for just over a month now. I believe termination may be the cause of your problems. The card appears to be working well for me (so far) and I wish you luck with yours. I could not get the Storage Manager software to work. I loaded the browser edition listed for the Adaptec Serial ATA RAID 2810SA. I could then use the command line software that came with the card. The following script compares the output from the command to a known good output and emails me in case of trouble (some lines may have wrapped). #!/bin/bash # =====> Change the following line to whom should be notified # mailwatch=louis@ldrit.net # host=`hostname` # cd /usr/local/src/raid rm raid.current.config /usr/sbin/aaccli open aac0 : container list : disk list : task list

/usr/local/src/raid/raid.current.config curdiff=`/usr/bin/diff raid.current.config raid.production.config` # raiderr=`/bin/cat raid.current.config` # if [ "$curdiff" != "" ] ; then # notify mailwatch team - raid 5 may be broken /usr/sbin/sendmail -f devnull@$host -t << RedCatSun To: $mailwatch From: root@$host Subject: Raid 5 may be broken on $host Errors-To: devnull Raid 5 may be broken on $host.

====> A diff between production and current is: $curdiff ====> The current container list and disk list is: $raiderr RedCatSun fi # exit 0 -- Louis Richards

On Saturday 26 March 2005 04:31 am, Anders Norrbring wrote: <snip>

I don't know.. Termination seems to be "odd" in this case since it's an old Compaq Proliant rack case with original hard disks in it. They're not known to be problematic. Even if they're old and slow.. :)

Also, when I go to the controller's disk utility to verify media and low-level format the drives they're reported as 8.4GB, which is correct. But when I go to "Create Array", they're shown as 4GB disks. I'm really confused. The other drives in the rack are 18.2GB drives and they're reported correctly all the way.

Hi Anders! When I was in the RISC/*nix SCSI storage products market, it wasn't uncommon for our customers to get "bit" on occassion by subtle variations in firmware, meaning the OEM had "optimized" the device or controller (sometimes both) that they were selling. Other common problem areas: - mixing single-ended and differentially driven devices (forgetting to set jumpers, when available.) - using passive terminators when active were required and active when forced were required - incompatible, marginal and even new defective cables - inadvertent double-terminations (using a terminator on a last device jumpered to auto-sense and self-terminate) - mixing wide (16 bit) devices with narrow (8 bit) devices and controllers Just thought I'd share these observations with you in hopes they might help you develop a new avenue of inquiry. regards, - Carl -- _______________________________________________________________________ C. E. Hartung Business Development & Support Services http://www.cehartung.com/ carlh@cehartung.com Dover Foxcroft, Maine, USA Public Keys 68396713 & F8207216 Reg. Linux User #350527 http://counter.li.org/ ----->>>>>http://www.fsf.org/campaigns/supportlinuxbios.html <<<<<-----