Proliant ML350 E200i SmartArray Controller Fails To Initialise

[gentlefoot]

Dear All,

Last week our Proliant ML350 fans started running flat out. On inspection I noticed the CPU heatsink fan had stopped working. So the server was shutdown and the fan was replaced. However when attempting to boot the server the array controller failed to initialise. I got the message "1783 - Slot 0 Drive Array Controller Failure! [Controller Self-test failure]"

The E200i array controller is integrated into the system board so the system board was replaced. However, after replacement the array controller would still not initialise. Same error.

I checked the BBWC battery pack voltage with a multimeter and this was down around 2.3 volts so I replaced the battery with a new one. I also checked the new battery voltage with the multimeter before fitting it and the output was a healthy 3.8 volts.

However, the array controller is still failing to initialise with the same error and I am competely out of ideas as to what else could be the cause.

Anyone else had this problem or are there any further diagnostics I could be undertaking?

Thankyou kindly for your replies.

James

[gentlefoot]

Been doing some reading and I'm now thinking BBCW module. What do you think?

[Med-H]

Regarding the issue reported on the server ML350 G5 with a SmartArray Controller E200i  :

HP Smart Array Controller - 1783 Slot x Drive Array Controller Failure!

ISSUE:

System boots up slowly during drive initialization.

slot x HP smart array controller
1783-slot x Drive array controller failure!

• To isolate the 1783 issue related to a Smart Array controller cache module, complete the following troubleshooting steps:

1. Reseat the array accelerator module..

2. Turn off the server.

3. Update the server BIOS  and the controller firmware to the latest version .

4.Troubleshoot this issue by isolating the failed assembly if not a problem of controller cache :

• Disconnect the sas data cable(s) from the controller and post the server.

NOTE: If the 1783 error does not occur, then the fault is either cable(s), backplane(s) or disk(s). Isolate one item at a time until the faulty assembly is identified.

• Reconnect the cable(s)
• Remove all disks from the backplane by unplugging, but not fully removing from the disk slots.
• Power up the server, if the 1783 controller failure still occurs, isolate the cable(s) and backplane(s). If the controller passes self test then a disk(s) is causing the controller failure. Install half of the disks and power up the server. Continue adding or removing disks until problem disk(s) are isolated.

NOTE: Always press F1 after the controller displays 1789-Slot X drive disks not responding. Do not press F2 or data will be lost.

• If the 1783 controller failure still occurs with cables disconnected, the Battery Backed Write Cache (BBWC) or the controller is defective. Replace the BBWC and then the controller to isolate the failure.

  NOTE: some controllers will display a 1783 controller failure if the BBWC ( array accelerator module ) is either defective or not installed.

Please let me know if you need any further information.

thanks.

[Srini_Raghavan]

Hello James,

128 MB cache module can actually cause the controller not to initialize and thereby marking it as failed. You may try 64MB cache module to isolate the issue.  If your server is still covered by warranty then i would suggest you to HP for more qualified answer.

Regards

Srini.

[TVVJ]

Update the server’s BIOS and the Smart Array E200i Controller and the firmware of the hard drives. To do so, boot the server with the latest version of the Firmware DVD from the ML350 G5 server’s driver download link:

http://h20000.www2.hp.com/bizsupport/TechSupport/ProdSearch.jsp?lang=en&cc=us&taskId=&prod=ML350 g5

If the issue persists, then boot the server from the latest version of SmartStart CD image available in the above link and run Maintenance > Array  Diagnostics > Array Diagnostics Utility and generate the Array Diagnostics Utility Report and save it in a USB drive. Review the report for any hard drive failures, cache module failures, battery failures.

If the issue persists after updating the firmware, then remove the cache memory module and check if the issue is resolved.

Regards,

Vijayasarathy

[gentlefoot]

Hi and thanks for all the replies.

Srini, I took note of your mention that the 128Mb module can cause problems and the 64Mb module could be used to isolate the problem.

I am about to order a new 128Mb module and wanted to order a 64Mb module at the same time just in case replacing the 128Mb module does not solve the problem and same day delivery is expensive. Problem is I don't have a part number and cannot find any mention of a 64Mb module for the E200i anywhere.

Do you know how I can find out the part number for the 64Mb module?

Thanks again.

James

[Med-H]

Hello,

Do you have a Raid 5 configured on your server ?

Please find bellow all Parts references installed on the server include the 64Mb Module : 

QuickSpecs for the smart array E200i/E200 :

http://h18004.www1.hp.com/products/quickspecs/12460_div/12460_div.pdf

also you can use the Part surfer to get the exact spare part of the memory module cache :

http://partsurfer.hp.com/search.aspx

[Hazem Bahgat]

Dear James,

Kindly use the SMart Start access the Array Diagnostics utility (ADU), from Array Configuraton Utility (ACU), if the arrays are working well, go to the ACU and repair the array relation.

or you can repair it from the begining,  and then check if you may have any other errors

[Srini_Raghavan]

 Good Day James,

Find the part number below for 64MB cache module.

412800-001 Smart Array E200i controller cache module - 64 MB.

Below is what i would recommend to do :

1) Reseat 128MB cache module and check.

2) You need to have a cache module installed on e200 controller. So removing the cache module to isolate the issue is not an option. Try 64 MB cache if you have decided to buy one.

 Make sure that you perform all the required troubleshooting without the HDDs connected. We certainly won’t like to hurt the hard drives. In this way we can also isolate if any of the HDDs are causing this issue.

Do consider upgrading the controller to P400, if the cost of repairing e200 is going to sniff the price of a new controller.

Srini.

[BN_Yosra]

Hi James,

SA-E200/SA-E200i Controllers with 64 MB of memory support only  RAID 0, RAID 1, and RAID 1+0 drive arrays, but do not support  RAID 5 .

So if you're using a RAID 5 you need a 128 MB cache module .

try to reset the cache module and check the result.

[crisol2007]

What was solution? please!!!

[PRPD]

I just had the same issue. I was told one of these three parts was my problem: battery, 128MB battery-backed write cache (BBWC) or system board.

My issues was a bad drive.

[Sbrown]

It's not a bad idea to pickup a spare P400 on ebay for $25 and get a new battery. The older P400 won't work without a cache board. The P410 of course will work without a cache board, just not a wise idea at all.

I've run a P400/p410 in an ML350 G5 and the performance was quite a bit faster with the larger 512/1gb BBWC than the E200i!

[Ruffy]

Hi folks, the post error 1783 on ML350 G5 indicates that the embedded Smart array controller P200i is faulty; the same issue was encountered today with one of the customers. Support end recommended system board replacement. When my user saw the big he thought I came with a new server, when I told him it is system board he said he never requested for sys bd replacement but requested for array cache to be replaced. Anyway I replaced sys bd twice, it never resolved the issue not until I replaced the Memory Cache module of the array controller.

When I discussed with user to know more, he said they are having lots of ML350 G5 on several sites and whenever they face similar issue they only request for memory cahe replacement.

In the nearest future just start by replacing the memory cache, make sure your get the correct part number of the cache by verifying physically, once you repalce this the error will disappear, array controller will be fully initialised and detect your logical drives.

[Dave_Petterson]

Just for the record, I have had this exact fault today (Controller Failure) after replacing the Cache Battery, So then replaced Motherboard...no change...so replaced Cache Module itself and it is fine again...Phew. Try the Module first!

[Alex72]

Hello, i have the same problem but in a DL380 G5, recently add a pair of SAS 300Gb hard disk, the error occur just when i connect this hard disk, with original hard disk, sas 146Gb doesn´t occur this problem.

This is the error:

1783-Slot 3 Drive array controller failure.

I had to many problems to install OS its took about 4 hours to be installed, cause the error, i had to reboot to many times and system load, after if you need to turn off or reboot the server, the system can´t be load cause this error.

When i connect the original hard disc its work fine, and the error does´t appear.

This is the model of the hard disk.

507127-B21 HP 300GB 6G SAS 10K rpm 2.5-inch Dual Port Enterprise Hard Disk Drive

Regards.