PE T610 & Perc6/i Integrated

Oh, silly me forgot to also ask if a Dell SAS Raid Storage Manages information message of "Controller ID: ) Unexpected sence: PD = :32, CDB = 0x12 0x00 0x00 0x00 0x04 0x00, Sence = 0x70, 0x00 0x06 .... 0x29 0x00 ..." is of concern? What does it indicate?

It is possible to change between RAID levels ... another name for it is Online Capacity Expansion, meaning you can expand the capacity of your RAID array by ADDING a disk to the array (or removing redundant disk space).  So, you can convert a 2-disk RAID 1 to a 3-disk RAID 5, a 3-disk RAID 5 to a 4 (or more)- disk RAID 5, etc., while the array is live and online.  Another supported migration would be a 2-disk RAID 1 to a 1 or 2-disk RAID 0 (as you are removing a disk used for redundancy to make the array bigger) ... thus, a 2-disk RAID 0 cannot be converted back to a 2-disk RAID 1. 

<ADMIN NOTE: Broken link has been removed from this post by Dell>

I do not believe this can be done in CTRL-R or in the limited workstation Storage Manager (although I could be wrong).  OpenManage Server Administrator (OMSA) is the software to do this, but it will not run on Windows 7.  You could do this with OMSA Live! (bootable Linux CD with OMSA integrated).

http://linux.dell.com/files/openmanage-contributions/omsa-64-live/

You might be able to use OpenManage/controller command-line to do this too - IF Storage Manager installed the necessary tools in Windows 7.

I haven't used Storage Manager enough to know what information you will have available to you, but I suspect there would be a way to look at the controller log for SMART and other drive information.  If the drive fails SMART, it will kick off a predictive failure flag on the drive, indicated in the drive status (I would think even in Storage Manager) and blinking status lights on the drive (amber for failure, alternating green/amber for pred fail).

The entry 6:29:00 for the sense key/code indicates a device reset.  That could indicate a bad drive, but is more likely a firmware issue.  If this is not a "certified" (Dell co-branded), that could explain it (non-certified drives do not always know how to respond to all the commands from the controller).  If a certified drive, update your drive firmware (after updating BIOS, ESM, and RAID controller).  If not a certified drive, consider "upgrading" to a certified drive (drives should be at a MINIMUM Enterprise-class drives).

The DRAC has usually updated inventory automatically for me.

Thanks theflash1932 :)

I'll check out OMSA live and see what options exist and cross check against the workstation solutions manager options to see if it is indeed limited in any way.

As for SMART, this has drive info i'm interested in seeing. Though abstraction is good, i still want access to the details:) Having a HDD backplane flashing amber/green light for predicted drive failure is a great abstraction, intuitive and much better than a windows event log that i'm used to seeing :) Pity the T610 bezel doesn't allow me to easily see the flashing leds :( though i know its a server and everything is built for server grade os with remote management,  a bezel with clear plastic view strips would be a great improvement :)'

With the HDD, they are all Dell certified drives so i'm not sure what to make of the log entry. I'll check  the firmware versions again but i'm not 100% across the upgrade process. The last t610 f/w upgrade i did resulted in the fans going spastic and me panicking and shutting the system down. (The panic was due to my previous spastic fan experience on a Dell Optiplex during normal power up; the machine never booted again.)

Should the firmware upgrade process be motherboard BIOS update 1st, followed by the DRAC firmware, followed by the backplane firmware, followed by the power supply firmware, followed by the raid card firmware, followed by the hdd firmware? Is that correct? What is ESM firmware? Can all this be done via windows exe files?

Just so you know ... all system firmware can be done via Windows update files.  The BIOS firmware should be first and simply requires a reboot.  The ESM/BMC should be next (on newer systems, this is the same as the iDRAC ... the update may take up to 10 minutes and the fans will speed up during the duration of the update (don't turn it off! :)).  The backplane (if applicable) would be next, then the RAID driver, then RAID firmware, then drive firmware.  Drive firmware can be easily updated from the bootable utility on the Drivers/Downloads page.  Other firmware updates (NIC, PS, etc.) would follow.

ESM = Embedded Server Management ... it is the device that monitors and manages hardware devices (temps, volts, error logging, etc.).

Here is the log from Storage Manager.
They are all [Information,0] messages, the last in the list is from the previous shutdown while the next 18 are during reboot process and the top 11 are to present time.

Controller ID:  0   Time established since power on:   Time   2011-12-13,17:00:46      11092  Seconds
Successful log on to the server User:  xxxx, Client:  192.168.1.100,    Access Mode:  Full, Client Time:  2011-12-13,16:19:50
Controller ID:  0   Time established since power on:   Time   2011-12-13,16:00:46      7492  Seconds
Controller ID:  0   Time established since power on:   Time   2011-12-13,15:00:46      3892  Seconds
Controller ID:  0   Battery charge complete
Controller ID:  0   Time established since power on:   Time   2011-12-13,14:00:46      292  Seconds
Controller ID:  0   Time established since power on:   Time   2011-12-13,13:57:57      112  Seconds
Controller ID:  0   Battery started charging
Controller ID:  0   Current capacity of the battery is above threshold
Controller ID:  0   Battery temperature is normal
Controller ID:  0   Time established since power on:   Time   2011-12-13,13:57:56      50  Seconds

Controller ID:  0  Device inserted   Device Type:      Disk  Device Id:   --:--:1
Controller ID:  0   PD inserted:      --:--:1
Controller ID:  0  Device inserted   Device Type:      Disk  Device Id:   --:--:0
Controller ID:  0   PD inserted:      --:--:0
Controller ID:  0  Device inserted   Device Type:      Unknown  Device Id:   :0
Controller ID:  0   PD inserted:      :32
Controller ID:  0  Enclosure element status changed Enclosure     :  :255  SES code   23
Controller ID:  0  Enclosure element status changed Enclosure     :  :255  SES code   23
Controller ID:  0   Communication restored on enclosure:        :  :255
Controller ID:  0   Unexpected sense:   PD      =   :32,   CDB   =    0x12  0x00  0x00  0x00  0x04  0x00     ,   Sense   =    0x70  0x00  0x06  0x00  0x00  0x00  0x00  0x0a  0x00  0x00  0x00  0x00  0x29  0x00  0x00  0x00  0x00  0x00
Controller ID:  0  Board Revision:   
Controller ID:  0  Package version      6.3.1-0003
Controller ID:  0  Controller hardware revision id is:  (0x0)
Controller ID:  0   Battery Present
Controller ID:  0   MegaRAID firmware version:      1.22.32-1371
Controller ID:  0   MegaRAID firmware initialization started:      ( PCI ID   0x1000/ 0x0060/ 0x1028    / 0x1f0c)
Controller ID:  0   MegaRAID firmware version:      1.22.32-1371
Controller ID:  0   MegaRAID firmware initialization started:      ( PCI ID   0x1000/ 0x0060/ 0x1028    / 0x1f0c)

Controller ID:  0   Hibernate command received from host

Only 1 perc6/i exists, 2x 300GB 15K dell SAS hdd in slot 0 & 1 are inserted.
The enclosure has device id 32.
I assume the enclosure equates to the backplane.
Are information/warning/error codes listed anywhere?
Are the meaning of the words within the codes listed anywhere?

Cheers.

Thanks very much theflash.

I assume that if during boot, listed firmware version are indeed an acknowledgement that a firmware flash has 'taken' and the loaded image is indeed correct, that is, the firmware does a image checksum/hash check to ensure integrity. Is this correct? If not can i do a reflash or will the utility just tell me the versions are the same and abort?

I guess next step for me is to check all firmware is at the latest :) Tried yesterday but the dell support site was having issues :(

Yes, the update will do a check before committing the update.  If it is the same version, it will warn you that it is the same, but it will usually let you proceed in reflashing that version anyway.

Updated or reapplied all available firmware DUP file i could find for T610 (server2008R2)so now have:

BIOS: 6.1.7
iDRAC Firmware: 1.80 (build 17)
PERC6/i Integrated Controller firmware: 6.3.1-0003
Dell OS driver pack: 6.5.0.12, A00 6.5.0.12
Dell 32 Bit Diagnostics 5154A0
Dell Lifecycle Controller, v.1.5.0.671, A00 1.5.0.671
Broadcom NetExtreme II Gigabit Ethernet firmware: 6.2.12

BIOS and PERC were updated along with OS driver pack & diagnostics while iDRAC was reapplied.

Unfortunately i am still getting the following shown via "Dell SAS Storage Manager v2.66-00".

[Information,0]    48seconds from boot   Controller ID:  0  Unexpected sense:  PD  =  :32,  CDB  =  0x12  0x00  0x00  0x00  0x04  0x00 , Sense  =  0x70  0x00  0x06  0x00  0x00  0x00  0x00  0x0a  0x00  0x00  0x00  0x00  0x29  0x00  0x00  0x00  0x00  0x0

I have looked but can't find any document that describes what the bytes within the info message description mean. Nearest i could get was from the sas storage managers list of event messages under the tools help menu but this referenced message 0x0071 which is listed as a warning but gives no breakdown of what the %S means within the description "Unexpected sense: %s, CDB%s, Sense: %s".

I am using Windows 7 and the SAS Storage Manager meant for Precision workstations, since the T7500 has a PERC6/i Integrated and Windows7 OS). Am i worng to assume this should work and the app is not limited in some way? Surely Precision workstation users that have paid for a PERC6/i Integrated deserve a fully capable working solution under Windows 7?

Where to from here (yet to try OMSA live)

Sense  =  0x70  0x00  0x06  0x00  0x00  0x00  0x00  0x0a  0x00  0x00  0x00  0x00  0x29  0x00  0x00  0x00  0x00  0x00

http://en.wikipedia.org/wiki/Key_Code_Qualifier

06 - 29 - 00 indicates "Unit Attention - POR or device reset occurred" ... a device reset is normal in some circumstances, but if you continue to get this constantly while the drive is in use, it can indicate a bad drive - or a drive that didn't respond in a timely manner or did not understand the command sent to it.  Updating the drive firmware can often help the drive understand what is being asked of it or eliminate unnecessary resets.  Make sure you have also updated the drive firmware.  If not Dell drives, then you will need to obtain the firmware from the drive manufacturer.  If not a Dell drive, certified for use on a PERC 6, then even with new firmware, it may not act like the controller expects it to.  Are you getting this message for only one drive?  If so, how many do you have and what models are they?  (Sorry if you told me already.)

I know that the SAS Storage Manager is limited in comparison with OMSA, but having used SSM only very little, I don't know exactly how limited it is.  I would use OMSA Live to access functions of the controller that SSM may not be able to use.  It isn't a matter of deserving fully-functional software, it is a matter of "use".  OMSA does much more than interface with the controller - it interfaces with the ESM, BIOS, and all the other specifically-linked devices in a server to manage the entire box.  Precision workstations do not have the appropriate devices for OMSA to install on it.  SSM may be limited, but it was designed to manage specific hardware - servers and workstations that need RAID monitoring but that lack the system requirements/necessity for running OMSA.

OMSA Live would be your best option for using functions of OMSA (Reconfigure, etc.) - OMSA Live will work on a server independently from the OS.

The OMSA64 live CD provides lots of information and options not available on SAS Storage Manager and i do not see the same 'sence' error from OMSA64 live, maybe i need to look harder :) It's just a pity i can't install the OMSA64 live CD image on VMWARE as the VM install process fails to see a linux install disk :( It would be nice to have access to OMSA via the OS :)

Anyway, the T610 server has the Dell PERC6/i Integrated controller and 3.5" drive backplane with latest firmware on these components nearest i can tell. I also i have 2x Dell 300GB 15K SAS drives installed in slot 0 and slot 1 while slot 2 & 3 contain dell blank plates and slot 4 to 7 contain 4x Dell 300GB 15K SAS drives that are only partly inserted and acting as blanking plates (the noise is too much when these drives are plugged in and i don;t have 4x blank plates handy). 

I did a search on "Unexpected sence: %s, CDB%s, Sense: %s" and this page indicated that sense errors could be due to a controller not disabling the write cache on drive reset, at least on IBM controllers.  Also this intel document describes more detail about decoding unexpected alert codes and i am attempting to digest the information :) Also, this page describes driver/firmware issuesin a BSD environment with LSI controllers. Unfortunately i have not found any Dell documents other than the SAS Storgae manager help file previously mentioned.

Being that i am using a Dell system with a Dell controller and Dell certified drives along with Dell software for T7500, one would expect it should work without error and detail documentation should exists on this informational messages and all other information, warning and error messages. This highlights to me that there is some issue at the least with Dell documentation if not with some firmware...

I haven't rung Dell support as i expect they will just say NOT SUPPORTED CONFIGURATION and not investigate if there is some update that hasn't been propagated between LSI, T610 and T7500 system.

I have only seen the error with the string XXseconds from boot, so i may need to run some disk activity tests over a period to see if the message occurs outside boot. I have what i expected to be a robust system with very capable hardware but am fearful i may suffer data faults, that would be a disappointment.

As for SSM's ability, i would have expected T7500 owners would still like to be able to reconfigure their drives from within the os (despite not having all the idrac and other remote management features).

Cheers.

Well, i had a complete brain fade :emotion-10:
VMWare Player provides abstraction of the hardware and presents a basic system to the installed 'virtual' OS. As such, starting a blank VM with no OS and booting the VM with OMSA64 Live CD results in seeing nothing of any interest, just a basic empty shell of a Dell system (unlike booting the server proper with the live CD) - Doh :emotion-10:

If i want to administer my Win7 T610 'workstation' from a working OS, could i not do a full VM install of CentOS and OMSA (proper) and then loop back via ethernet and in affect remote administer my T610 workstation from the remote virtual machine so to say? What if anything would i have to load on the T610 workstation Win7 OS? What should i look out for and configure (certificates, etc)?

Happy new year to all

theflash1938, sorry i missed your post.

A Dell supported OS like Windows Server 2008 R2 with full remote management features is just too expensive and complex considering how i use the T610 machine. Obviously i am not in Dells target market. :emotion-4: Since i must use a specific windows XP app (not my choice), the only alternative i could accept was (a Dell unsupported) Win7. This meant using management apps from Precision Workstation line, and though they seem to work, they also seem to be missing some features. :emotion-18:

Your suggestion of the CentOS OSMA64 live CD works well enough for planned administration though i would like to avoid having to reboot the system just to manage it. That's why i thought OSMA64Live within a VM was a goer, doh :emotion-10: I just want to be able to administer the T610 raid and hardware locally on the T610. Ideally i would also like to have a local popup when some critical issue occurs (disk failure, power failure, etc). At the moment, i can only expect to get event log messages from MR_MONITOR :emotion-6:.

So the question is how do i fully manage (e.g. change raid types) the raid array locally from within Win7 since the Precision Workstation Storage Manager is less able than that available in OSMA? Dito for hardware/firmware. Is there a way of getting local popups for critical errors? Is a CentOS VM with a full OSMA install the only way to go (seems complex and i still have the issue of how to handle local critical event popups :emotion-42: )?