X58A-UD7 (rev.2) after Bios FC no access to SATA3 Raid anymore

[Zool]

Happy new year !

I have updates Mobo X58A-UD7 rev. 2.0 to the latest Bios FC from the previous release FC6. Now i have big trouble with my SATA3, beacause i cant access my raid 1 in bios anymore. These are two WD RE4 1TB Drives. (Enterprise drives - so you dont have to expect the caviar black problems). The raid was initialy build under Bios revision FB.

The update process was FB -> FC6 -> FC7 -> FC

Now i always get the message:
"SATA3 IS NOW WORKING IN IDE MODE, CHANGE TO RAID MODE FIRST. BESIDES, HDDS SHOULD BE PLUGGED INTO SATA3 PORTS AND REBOOT THE SYSTEM TO CONFIGURE SATA3 RAID IN BIOS SETUP MENU."

In Windows the MRU (Marvell RAID Tool) Reports -> all fine

After that i have FORCED the bios update of the controller.... no luck
Here are the versions - perhaps this is important for someone:

Bios Version: 1.0.0.1027 -> 1.0.0.1029
Firmaware Version: 2.1.0.1504 -> 2.2.0.1113

After that i have done some experiments with all the setting (auto to raid and all that stuff) .... no luck
The speed with the new Bios is also bad. If someone wants to know i will also show you the screens of the différent tools.

Now i want to know if there is somebody out there with ideas and even with some helping hands....

Need help with that - because i dont want to destroy the raid again.

regards

.Zool.

[F5BJR]

*
Why you use not the INTEL ICH10R port ?

Very fine in RAID mode

**

Marvell port is very bad for using in RAID mode

I have tested with my * 2 Port Multiplier enclosures with 4 * disks ( 1TO )

with 1027 disks number are normal 0 , 1,  2,  3 or 4

with 1029 i have anormal numbers 2 , 3 , 4 , 5 ... !!

Pierre

*

I have also big problems withanother Marvell port : 88SE6145 , i cannot use my port multipliers box's ( many time out )

i search a solution > 2 years without success

My 2 *  box's  run perfectly with HighPoint RR620 , Silicon Image SIL3132 ( Raid 0 , 1 , 5 , 10 )

[Lsdmeasap]

The above is correct, Marvell is not very good for RAID at all.

However, I did post a reply for you about this issue here
http://forums.tweaktown.com/gigabyte/28441-gigabyte-latest-bios-398.html#post381037

[Zool]

good morning again ...

we moved to TweakerTown:
http://forums.tweaktown.com/gigabyte/28441-gigabyte-latest-bios-398.html#post381066

but i thought a long time using marvell or the onboard Intekl chip and decided to use the marvell for only the both "WD RE4 1TB SATA3" Drived for one array. The chip is eactly build for that I also tried a raid 0 array for testing and the speed was not so bad ->

http://img718.imageshack.us/i/wd1tbre4raid0.jpg/


I also use a OCZ Vertex2 and it nearly the same speed ....

http://img80.imageshack.us/i/oczvertex2.jpg/


regards
.Zool.

[Dark Mantis]

Obviously you must use whichever controller you feel gives you the best performance, however in our tests the Marvell 9128 controlled ports didn't respond as well as the Intel ICH10R ports controlled by the Southbridge.

I noted that you are using the Enterprise versions of the Western Digitals but that doesn't mean that the Marvell is the better controller for the RAID0 configuration as it just falls down on the throughput.

[F5BJR]

*
A new bug with Marvell 889128 this day ... !!!!!

I install MRU last version Marvell RAID Utility 4.1.0.1909 with WINXP

And ..

Surprise my disks ( in IDE mode ) are always in activity

Similar with Marvell RAID Utility 4.1.0.1905

With old versions < 1700 i have not this problem ...

Pierre

* a new MRU release , a new bug !!!

*

[F5BJR]

*
Excuse me , i have make an big error

--> With MRU 19xx my 2 * disks are not attached to the 88SE9128 port but ...

... ... to the HighPoint RR622 ports and are always in activity with MRU 19xx !!!

Pierre

[Dark Mantis]

So now Marvell are not content to screw up their own controller they want to screw up someone elses too

[Zool]

sorry for beeing so late but i had a lot of work to do ...

i have testet teh downgrade to Bios FC7 release and even without forcing a downgrade of the marvell bios i could access my raid in bios without any problems. So the bug seems to be in the FC Bios of the mainboard. Could someone tell me how i can give these infos to the guys from gigabyte ?

if someone else hase a good idea - write it down ... i need an idea how to go on ...

Someone else wrote, that i should use the Intel as a raid. The prroblem was, taht i have installed the system with an ssd in ahci mode on the intel. And if i switch to raid mode now i get the bluescreen and missing bla bla info now....

regards
.Zool.

[Dark Mantis]

Please just start a dialogue with GGTS about it.

Just enter your email address and click on the language of choice.
GGTS   http://ggts.gigabyte.com.tw/Default.aspx

[F5BJR]

*
For last Marvell problem with MRU 4.1.0.1909 i have contacted giabyte ggts

The response is this in french !!

Bonjour,

Nous n'avons malheureusement aucune influence sur Marvell, il vous faut persévérer dans vos tentatives de les contacter.

Unfortunately we have no influence on Marvell, you must persevere in your attempts to contact them

*******

GigaByte cannot contact Marvell ?

Pierre

[Dark Mantis]

I would have thought that a company the purchases chips in the millions off Marvell would have a little bit of influence with them. I find it hard to believe that this is actually the case and not just an excuse

[F5BJR]

*
I have tested Marvell MU 4.1.0.1905 installed in another motherboard ( ASUS P5E WS )

I have no Marvell port active ( 88SE6145 invalided in BIOS ) , no add-on cards  , only Intel ICH

Only one IDE disk connected in IDE mode

And surprise ..

-> with mvraidsvc.exe CPU activity is 18% , IDE disk run always ( disk LED with many blink )

-> without mvraidsvc.exe CPU activity is some 2 % ( disk LED not many blink )

** MRU run as a virus **

Pierre




 

[Dark Mantis]

It sounds buggy as hell Pierre!

[F5BJR]

*Tested with this MoterBoard's same problem ( WINXP 32 ) :

* Sure MRU 4.1.0.1905 or 1909 run as a virus *

ECS RS400

ASUS P5E WS

ASUS P5B WIFI AP

ASROCK 775DUAL-880PRO

GigaByte GA-X58A-UD5

Pierre