Hard Disk ATA status: 41 (DRDY ERR), error: 10 (IDNF )

Status
Not open for further replies.

Kris Heslop

Dabbler
Joined
Feb 7, 2016
Messages
42
I searched and googled, but could not find a solved/answered situation like this.

If I could, I would replace this disk, just not able to do that right now. It is a refurb WD Red 4TB that I have had running for about 1.3 years. I was not able to pull things off the prior disk in time and ended up "buying refurb w/30-day warranty."

I thought I tested it, but am wondering now if it was enough.

The max read speed I am able to get is 153.6 MB/s with an avg about 128.3 MB/s AVG when running badblocks via the reporting in 11.2, whereas my other Reds are able to get above 165 MB/s read/write peak and avg about 150 MB/s when testing.

The disk is now passing all SMART and Badblock tests-before failed a short and then passed a long (fixed it or the WD tool did), yet still giving me two apparently interrelated problems that affect performance. I am hoping to add this into a RAIDz1 pool to increase my media storage (I have other back-ups, but am running close to 80% space...) for about 6-months, and then replace the drive and transition to Raid Z2 when I can afford it:

A. Error Messages

Jul 23 13:40:07 HeslopNAS (ada2:ahcich2:0:0:0): WRITE_FPDMA_QUEUED. ACB: 61 00 00 82 3e 40 01 00 00 01 00 00
Jul 23 13:40:07 HeslopNAS (ada2:ahcich2:0:0:0): CAM status: ATA Status Error
Jul 23 13:40:07 HeslopNAS (ada2:ahcich2:0:0:0): ATA status: 41 (DRDY ERR), error: 10 (IDNF )
Jul 23 13:40:07 HeslopNAS (ada2:ahcich2:0:0:0): RES: 41 10 00 82 3e 40 01 00 00 00 00
Jul 23 13:40:07 HeslopNAS (ada2:ahcich2:0:0:0): Retrying command

This error code has repeated on a few occasions, maybe under 1 out of 5 times for about 5-6 times total, yet the slow down is consistent with every badblocks pass...

As shown here (B.) Bad read times:
upload_2018-7-23_16-21-7.png

Time covered 1 hour above
The results of the last part of a series of 8 badblocks passes over the first 75,000,000 blocks on the disk.

No errors with badblocks reported this time.

I ran mke2fs once that I think formatted it plus one straight badblocks on this disc for 4 passes that flagged one bad block one time and either fixed it or marked it (and the read times for other portions of the disk are up to speed.)

SMART says it passes, yet it seems to me that I have at least one, although looks more like 2 bad areas with bad blocks are not getting flagged, since it slows reads to almost zero. (I did run a WD Diagnostic test with the disk in a windows machine that showed and a utility concurrently running that cofirmed the read rate dropping down to under 0.3MB/s, but no way to tell what LBA.)


So the questions are:
1. Is there a way to find out what blocks/LBAs are causing the slow down and manually mark them bad (I seem to recall seeing directions on marking the block #).
2. Any other ideas, or just toss the disk? (I was also thinking I could block out the bad area and use it as a mirror for my single 3TB disk.... in the alternative.)

I am using a Lenovo TS140
I3-4130 3.4 Ghz with 12 GB factory memory (DDR3 1600 ECC UDIMM)
1 WD Red 3TB stand alone
2 WD Red 4TB + 1 HGST Deskstar NAS 4TB in Radi Z1
Boot is a Sandisk flash drive.

Running PMS and Time Machine on the RaidZ1 and the 3TB drive is a backup for scans and other files. (Plus have external back-ups)
 
Joined
May 10, 2017
Messages
838
The max read speed I am able to get is 153.6 MB/s with an avg about 128.3 MB/s AVG when running badblocks via the reporting in 11.2, whereas my other Reds are able to get above 165 MB/s read/write peak and avg about 150 MB/s when testing.

This could be explained by the refurb disk being the older 4 platter model (WD40EFRX-68WT0N0) while your other ones are the newer and faster 3 platter model (WD40EFRX-68N32N0).

You should post a SMART report of the disk so we can better see what you're working with, but in my experience refurb disks should be avoided as they are much more likely to fail.
 

Kris Heslop

Dabbler
Joined
Feb 7, 2016
Messages
42
Thank you for your input and sorry for the delayed upload, I thought they were on a different machine (but they were on the server instead.)

OK, the slower speed makes some sense if the model is different.

What I am not understanding is how the read speed can drop to under 0.33 MB/s and it says it is reading OK. I ran some tests and it looks like about 100,000 blocks are impacted and reading super slow.

The blocks that are reading slow are not in the range of what was previously reporting as having an issue, and I show no reallocated sectors.

Like I said, I would like to replace the disk, but am not in a position to right now.

Given the issues, I am wondering if it is safe to use in any way.
 

Attachments

  • WDRedLogs070218a.txt
    41.8 KB · Views: 384
  • ADA2 072618.txt
    20.1 KB · Views: 393
Joined
May 10, 2017
Messages
838
It doesn't look good, the very slow reading speed is most likely caused by slow sectors, visible in the SMART report as the Raw_Read_Error_Rate attribute, on WD drives you want this to be 0 or at worst low double digits, that disk went from 1294 to 12918 in just a few hours

Code:
ID# ATTRIBUTE_NAME		  FLAGS	VALUE WORST THRESH FAIL RAW_VALUE
  1 Raw_Read_Error_Rate	 POSR-K   200   200   051	-	1294
  ...
  9 Power_On_Hours		  -O--CK   084   084   000	-	11952


Code:
ID# ATTRIBUTE_NAME		  FLAGS	VALUE WORST THRESH FAIL RAW_VALUE
  1 Raw_Read_Error_Rate	 POSR-K   200   200   051	-	12918
  ...
  9 Power_On_Hours		  -O--CK   084   084   000	-	12178



And in my experience these slow sectors turn to bad sectors sooner rather than later.

This is also not a good sign:

200 Multi_Zone_Error_Rate ---R-- 200 200 000 - 110

This value should also be 0 on an healthy disk, if the disk was mine I wouldn't use it for anything of importance.
 
Status
Not open for further replies.
Top