WAAS hardware is OE612, but Rescue Disk detects model as OE512

Question

I've had this issue with 2 RMA'd OE612s already. They originally shipped with 4.0.19, which worked fine, but failed on upgrade to 4.2.1.
Recovery attempt #1 with 4.0.19 Rescue disk displayed :
/* Style Definitions */
table.MsoNormalTable
{mso-style-name:"Table Normal";
mso-tstyle-rowband-size:0;
mso-tstyle-colband-size:0;
mso-style-noshow:yes;
mso-style-priority:99;
mso-style-qformat:yes;
mso-style-parent:"";
mso-padding-alt:0in 5.4pt 0in 5.4pt;
mso-para-margin:0in;
mso-para-margin-bottom:.0001pt;
mso-pagination:widow-orphan;
font-size:11.0pt;
font-family:"Calibri","sans-serif";
mso-ascii-font-family:Calibri;
mso-ascii-theme-font:minor-latin;
mso-fareast-font-family:"Times New Roman";
mso-fareast-theme-font:minor-fareast;
mso-hansi-font-family:Calibri;
mso-hansi-theme-font:minor-latin;
mso-bidi-font-family:"Times New Roman";
mso-bidi-theme-font:minor-bidi;}
   MODEL: UNKNOWN
   FLASH: found, directory validated
  COOKIE: invalid
   IMAGE: NONE
FLASHDEV: /dev/hda
Installer Main Menu:
    1. Configure Network
    2. Manufacture flash
    3. Install flash cookie
    4. Install flash image from network
    5. Install flash image from cdrom
    6. Install flash image from disk
    7. Wipe out disks and install .bin image
    8. Exit (and reboot)
Choice [0]: 3
Unknown model, cannot create cookie.
Subsequent Recovery procedures using Rescue disks from 4.1.7 and 4.2.1 detected the Model as OE512.  Option 2 creates a cookie as a MODEL OE512, which I wrote with option 3. Option 5 works, but trying to write the bin image on 4.1.7 displays this
/* Style Definitions */
table.MsoNormalTable
{mso-style-name:"Table Normal";
mso-tstyle-rowband-size:0;
mso-tstyle-colband-size:0;
mso-style-noshow:yes;
mso-style-priority:99;
mso-style-qformat:yes;
mso-style-parent:"";
mso-padding-alt:0in 5.4pt 0in 5.4pt;
mso-para-margin:0in;
mso-para-margin-bottom:.0001pt;
mso-pagination:widow-orphan;
font-size:11.0pt;
font-family:"Calibri","sans-serif";
mso-ascii-font-family:Calibri;
mso-ascii-theme-font:minor-latin;
mso-fareast-font-family:"Times New Roman";
mso-fareast-theme-font:minor-fareast;
mso-hansi-font-family:Calibri;
mso-hansi-theme-font:minor-latin;
mso-bidi-font-family:"Times New Roman";
mso-bidi-theme-font:minor-bidi;}
Choice [0]: 8
Enter full URL of .bin image to install.
   ftp://[user:pass@]ip_addr/path/to/file
  http://[user:pass@]ip_addr/path/to/file
  file:/local/path/to/to/file
URL for .bin image [file:/cdrom/images/WAAS41.bin]:
Continue?  This will wipe out all disks! [n]: y
Scrogging done
Saving random seed...
dd: opening `/state/random-seed': No such file or directory
  No volume groups found
/ruby/bin/ruby_disk: First disk is missing, bad, marked bad, disabled via shutdown, or it has unknown data.
/ruby/bin/ruby_disk: Second disk is bad, missing, or shutdown; bailing out
/ruby/bin/ruby_disk: First disk is missing, bad, marked bad, disabled via shutdown, or it has unknown data.
/ruby/bin/ruby_disk: Second disk is bad, missing, or shutdown; bailing out
ssmgr: open(/state/safestate, O_WRONLY|O_CREAT) failed: No such file or directory
ssmgr: op_open or media_open
ssmgr: open(/state/safestate, O_WRONLY|O_CREAT) failed: No such file or directory
ssmgr: open(/state/safestate, O_WRONLY|O_CREAT) failed: No such file or directory
ssmgr: op_create or media_init
SSMGR RETURNING: 10 (No such file or directory)
Reading and installing image, it may take a few minutes, please wait...
Error, image NOT installed.
   MODEL: OE512
   FLASH: found, directory validated
  COOKIE: valid
   IMAGE: 4.1.7.11
FLASHDEV: /dev/sda
Installer Main Menu:
    1. Configure Network (done)
I suspect that the Installer program on the Rescue CD can't detect the Hardware properly, and thus can't find the proper drivers for the platform.
Is there any manual method of forcing the Rescue CD Installer to use a particular hardware model?
Thanks,
Tom

Answer 1

Hi Zach,
>What led you to change the Adaptec configuration?
I was directed to do so by TAC, although I wasn't sure why, since our other WAE-612-k9s didn't have this option enabled.
>What is the state of the device prior to upgrading?  Does the WAAS software boot?
After disabling the hardware raid, the WAE could run any 4.0.x version, but could not upgrade to 4.1/4.2. I tried both the "rescue CDs" for 4.1/4.2, and the "copy ftp install" methods. "Rescue CDs" would work right up to the option for "wipe disks and install .bin" image, but would report "no disks" as shown above. The "copy ftp install" from version 4.0.27 would install the 4.2.1.38 image, but upon reboot, it would fail during the boot sequence
/* Style Definitions */
table.MsoNormalTable
{mso-style-name:"Table Normal";
mso-tstyle-rowband-size:0;
mso-tstyle-colband-size:0;
mso-style-noshow:yes;
mso-style-priority:99;
mso-style-qformat:yes;
mso-style-parent:"";
mso-padding-alt:0in 5.4pt 0in 5.4pt;
mso-para-margin-top:0in;
mso-para-margin-right:0in;
mso-para-margin-bottom:10.0pt;
mso-para-margin-left:0in;
line-height:115%;
mso-pagination:widow-orphan;
font-size:11.0pt;
font-family:"Calibri","sans-serif";
mso-ascii-font-family:Calibri;
mso-ascii-theme-font:minor-latin;
mso-hansi-font-family:Calibri;
mso-hansi-theme-font:minor-latin;
mso-bidi-font-family:"Times New Roman";
mso-bidi-theme-font:minor-bidi;}
/* Style Definitions */
table.MsoNormalTable
{mso-style-name:"Table Normal";
mso-tstyle-rowband-size:0;
mso-tstyle-colband-size:0;
mso-style-noshow:yes;
mso-style-priority:99;
mso-style-qformat:yes;
mso-style-parent:"";
mso-padding-alt:0in 5.4pt 0in 5.4pt;
mso-para-margin:0in;
mso-para-margin-bottom:.0001pt;
mso-pagination:widow-orphan;
font-size:11.0pt;
font-family:"Calibri","sans-serif";
mso-ascii-font-family:Calibri;
mso-ascii-theme-font:minor-latin;
mso-fareast-font-family:"Times New Roman";
mso-fareast-theme-font:minor-fareast;
mso-hansi-font-family:Calibri;
mso-hansi-theme-font:minor-latin;
mso-bidi-font-family:"Times New Roman";
mso-bidi-theme-font:minor-bidi;}
Frustrated, I looked at the boot logs and might have found an issue. On the "faulty" WAE-612, the boot script detects two SCSI adapters; only one shows up in BIOS POST.
The first adapter (Adaptec AIC79XX) has no drives, but is assigned the device name "scsi0"
Under the second adapter(Adaptec AIC94XX), the Hard disks are detected, and it's assigned a device name of "scsi1"
ACPI: PCI Interrupt 0000:03:03.0[A] -> GSI 17 (level, low) -> IRQ 17
scsi0 : Adaptec AIC79XX PCI-X SCSI HBA DRIVER, Rev 2.0.30
        aic7901: Ultra320 Wide Channel A, SCSI Id=7, PCI-X 67-100Mhz, 512 SCBs
Adaptec aacraid driver 1.1-5[24495]custom-IBM
Loading AIC-94xx Linux SAS/SATA Family Driver, Rev: 1.0.8-12
Probing Adaptec AIC-94xx Controller(s)...
ACPI: PCI Interrupt 0000:03:04.0[A] -> GSI 19 (level, low) -> IRQ 19
scsi1 : Adaptec AIC-9405W SAS/SATA Host Adapter
scsi 1:0:128:0: Direct-Access     IBM-ESXS ST3146755SS      BA33 PQ: 0 ANSI: 5
adp94xx:0:128:0: Tagged Queuing enabled.  Depth 32
scsi 1:1:128:0: Direct-Access     IBM-ESXS ST3146755SS      BA33 PQ: 0 ANSI: 5
adp94xx:1:128:0: Tagged Queuing enabled.  Depth 32
AIC-94xx controller(s) attached = 1.
To contrast, our functioning WAE-612s do not have this second detected Adaptec host adapter, and are properly assigned a "scsi0" for their device name.
Adaptec aacraid driver 1.1-5[24495]custom-IBM
Loading AIC-94xx Linux SAS/SATA Family Driver, Rev: 1.0.8-12
Probing Adaptec AIC-94xx Controller(s)...
ACPI: PCI Interrupt 0000:03:04.0[A] -> GSI 19 (level, low) -> IRQ 19
scsi0 : Adaptec AIC-9405W SAS/SATA Host Adapter
scsi 0:0:128:0: Direct-Access     IBM-ESXS MBA3300RC        SA06 PQ: 0 ANSI: 5
adp94xx:0:128:0: Tagged Queuing enabled.  Depth 32
scsi 0:1:128:0: Direct-Access     IBM-ESXS MBA3300RC        SA06 PQ: 0 ANSI: 5
adp94xx:1:128:0: Tagged Queuing enabled.  Depth 32
AIC-94xx controller(s) attached = 1.
On our faulty WAE-612,  in the section that tries to mount the detected drives, it appears that the script is looking for the physical drives under the "scsi0" device name, using the AIC94XX driver.. .
2010 Jul  8 08:58:41 (none) kernel: %WAAS-SYS-3-900000: Attached scsi disk sda a
t scsi1, channel 0, id 128, lun 0
2010 Jul  8 08:58:41 (none) kernel: %WAAS-SYS-3-900000: Attached scsi disk sdb a
t scsi1, channel 1, id 128, lun 0
2010 Jul  8 08:58:41 (none) kernel: %WAAS-SYS-3-900000: Attached scsi removable
disk sdc at scsi2, channel 0, id 0, lun 0
BOOT-100: disk apply
scan_612_disk_map: Can't open . Errno=2
change_disknum_612 failed!
/ruby/bin/ruby_disk: First disk is missing, bad, marked bad, disabled via shutdo
wn, or it has unknown data.
/ruby/bin/ruby_disk: Second disk is bad, missing, or shutdown; bailing out
I'm guessing this fails because there are no physical drives on "scsi0", they are assigned on "scsi1". If the disk mapping script  would've tried , I think the drives would've been mounted properly.
I was hoping there was a modified script out there somewhere, but after our 3rd RMA'd WAE exhibited the same characteristics (these were all in Chennai India), I directed the onsite tech to remove the Adaptec 29320ALP UltrasSCSI adapter.
With the adapter removed, the 4.2.1.38 code installed and booted properly.
When I forwarded my results to TAC, our Rep mentioned that the WAE-612-k9 ships with both ACNS and WAE capabilities. He said that ACNS uses that additonal SCSI card.
I'm fairly sure our configuration is not unique; our environment consists of  32 WAEs, and this is the first time that I've encountered this issue.
I'm just happy it's resolved now.
Thanks for your insight, Zach
Tom