slow SmartArray P440ar server, need help getting vendor to take action based on ADU report

[fgjruiz]

Hi all, I work for a software vendor that has been having issues getting a hardware vendor to take action on a problematic SQL server. 

Whenever:

#1 the SQL server writes heavily to a database on the problematic array

#2 a yearly report is run that hits the tempdb on the drive hard

#3 one of my colleagues runs a db shrink

#4 the drive array gets below 15 percent free

performance doesn't just slow down, it falls off a cliff.

The server is running a Smart Array P440ar (fw: 4.52) on an embedded slot controlling a total of ten drives.

Four 1.2TB and four 1.8TB drives with an additional one of each type running as a hot spare.

My issue is with drive 12, an HP EG1800JEHMD, which seems to be the last 1.8TB drive on the second array.

As you will see in the ADU reports this drive has been throwing an increasing number of Physical Drive Errors in that section of the ADU report.  None of the other drives have even one physical drive error. to this day.

This is how many Physical Drive Errors show up on each ADU report:

September 2017  018390 errors

August 2020 127025 errors

January 2022 219944 errors

The vendor dismisses the issue because SMART or whatever tools HP has are not tripping.

 I would really appreciate one of you experts either telling me to go to heck this says nothing or giving me ammunition to push for action.  It IS a little about being right, but if I'm wrong, I'm okay with just letting it go and putting a huge warning sign on the wallpaper on their server.

Now can someone help me attach an ADU as a text, I've seen the TXT files in other posts

Here's an excerpt from the ADU report I took today:

Smart Array P440ar in Embedded Slot : Internal Drive Cage at Port 1I : Box 1 : Physical Drive (1.8 TB SAS HDD) 1I:1:12 : Serial SCSI Physical Drive Error Log

Entry Size 20 (0x14)
Entry Count 64 (0x0040)
Next Entry Offset 0x28
Errors Logged 219944 (0x00035b28)
Physical Drive Error Log Entries Error Type SCSI Operation Code SCSI Status CAM Status Sense Key ASC ASCQ Block Valid Block Reference Time Additional Information
---------- ------------------- ----------- ---------- --------- ---- ---- ----------- ---------- -------------- ----------------------
0x01 0x2a 0x02 0x04 0x0b 0x47 0x01 0x00 0x401aec38 0x0028fb1a 0x0000
0x01 0x2a 0x02 0x04 0x0b 0x47 0x01 0x00 0x4065ff88 0x0028fb1e 0x0000

[Cali]

Hi,

I would first check this:

Is there a Write Cache Battey and is the Write Cache active (check this in iLO).

Also, I would add SSD for SQL Logfiles or/and add HPE SmartCache License and one SSD as general Read Cache.

This is a small investment and brings a good Performance push.

Cali

[sudhirsingh]

@fgjruiz 

Sense Key
---------
0x0b  >Drive abort commands - indicates that the drive aborted the command.

0x0b  >Drive abort commands- indicates that the drive aborted the command.

You should update drive firmware and controller firmware to latest and see if that helps!

Regards,

Sudhir