Proliant 6400r with Smart Array 3200 Controller 1

[k3v1n]

I have 2 servers getting the same error message.
1783-slot 1 Drive Array Controller Failure
(command line cmd=11 err=020h)
Any help solving this problem would be appreciated.

 

[TheLad]

If you look at the server itself, are all hard disk lights green or is one orange/red? If one is orange/red then you need to swap it out as the disk has failed. -----------------------------------------------------
"It's true, its damn true!"
-----------------------------------------------------

 

[k3v1n]

Replaced drives in Proliant storage system rso/u. Have changed out the controller in the rso/u also. The error occurs when you first turn on the servers. I have also installed a known good 3200 controller and still get the error. Have reinstalled rom paqs on server.

 

[TheLad]

Try looking at the following Compaq Doc. It is not specifically for the 3200 controller buy it might be in the right direction:

http://wwss1pro.compaq.com/support/reference_library/viewdocument.asp?source=EX010306_CW01.xml&dt=3

Basically it details problems when installing the RAID controller into a hot plug PCI expansion slot -----------------------------------------------------
"It's true, its damn true!"
-----------------------------------------------------

 

[k3v1n]

Thanks, Slot 1 is not a hot plug PCI expansion slot. I have however tried the controller in all the other slots. The 3200 manual says to install the Option rom paq to clear the error. The only problem is since the server sees it in a failed condition it will not install the paq.

 

[kd1s]

We just had a drive go orange on our Proliant 1600R with the embedded RAID controller.

Took the drive out, snapped another one in and just for laughs put that drive we took out into a test server. It's RAID controller had no problem with it.

What causes this?

 

[TheLad]

The server detects I/O errors reading and writing to the disk. After the server detects multiple read/write errors, it automatically puts the disk out of action. This is basically prompting you do replace the disk because the server thinks there is an error with it. If you were to pull and push the disk back into the server, you would find it would work ok as the counters are reset.

You need to make sure that the disk does not do the same eventually in your test box - going orange means a problem and you should treat it as a problem. -----------------------------------------------------
"It's true, its damn true!"
-----------------------------------------------------

 

[gizbourn]

Just had the same happen with us, raid 5, 3200 controlle

took the drives out to clean, as it had not been done for quite some time and we are upgrading it to w2k for a non critical server.

placed all drives back, in the correct order. switch on and for a milli second , then all lights on the drives die.

and the error message comes up eventually 1783 error and the cmd error.

Any suggestions gladly welcome

 

[aquaboy1976]

I just had this happen. Got the 1783 error, replaced the controller and reboot and still get the 1783 error. We don't have any of the lights on the hard drives turn on though. Any suggestions?

 

[Wizputer]

I have recently aquired a Proliant 7000, (4) 450Mhz Pentium 2 Xeons 1MB Cache, 2.25Gb PC100 EDO RAM and Smart Array 3100ES w/ 56Mb Accelerator Cache.

I had to replace the Internal CPU Fans to get it running.

When I powered it up and got past the CPU Fan Failure error and I now get the 1783 Array Controller failure that also says Self Test ErrCode=400.300Ch, the hard drive lights turn on and then off when power first turns off and they do not return. The 0-10 LEDs are green, 11-12 are off and 13-14 are orange on the 3100ES.

I have tried erasing the configuration and using scu to rebuld the configuration, it still does not work.

Anyone have some suggestions?

 

[Royster42]

This issue is caused by the battery in the array accelerator being flat. We just had the same problem with 1783 - Slot 3 Array Controller Error
[Command failure (cmd=11h, err020h)]
1850R with a 3200 ctrl. We checked our live backup and that went the same way. We then found an array controller that was new out the bag and it came to life straight away. Until you power the server off you cannot tell if the battery is bad. We managed to get the old one to respond to array diags and it said as much as the battery was flat. We had also tried several (supposedly) good controllers before we found a good one. Try the drives on the internal controller and see if they come to life.