New hard disk. Got errors

I

ilemur

I had two 3,5" drives in a zfs pool. Destroyed the pool. Inserted a new disk (WD 2,5") . Created a new ZFS pool and started to get errors.

Code: 
+ad13: TIMEOUT - READ_DMA retrying (1 retry left) LBA=143235584 
+ad13: TIMEOUT - READ_DMA48 retrying (1 retry left) LBA=353030400 
+ad13: TIMEOUT - READ_DMA48 retrying (0 retries left) LBA=353030400 
+ad13: FAILURE - READ_DMA48 timed out LBA=353030400
or like this

Code: 
kernel: WARNING - ATA_IDENTIFY freeing taskqueue zombie request
kernel: ad13: WARNING - ATA_IDENTIFY taskqueue timeout - completing request directly

This happens under light load (bonnie++, moving files, e.t.c). But if ZFS compression is turned off - i don't seem to get these errors... So can they be related? Or the problem lies somewhere else?

it's a c2d-6750 processor with 4Gb RAM on a Asus P5B Deluxe MB (Intel p965+ICH8R)
 
I have further investigated the problem... I have 2 zfs pools

Code: 
  pool: idata
 state: ONLINE
 scrub: scrub completed after 1h2m with 0 errors
config:

 NAME        STATE     READ WRITE CKSUM
        idata       ONLINE       0     0     0
          ad13      ONLINE       0     0     0

errors: No known data errors

  pool: storage
 state: ONLINE
 scrub: none requested
config:

        NAME           STATE     READ WRITE CKSUM
        storage        ONLINE       0     0     0
          raidz1       ONLINE       0     0     0
            gpt/disk0  ONLINE       0     0     0
            gpt/disk1  ONLINE       0     0     0
            gpt/disk2  ONLINE       0     0     0
            gpt/disk3  ONLINE       0     0     0
errors: No known data errors
Well as i have stated i have received errors in dauly output

Code: 
+ad13: TIMEOUT - READ_DMA retrying (1 retry left) LBA=143235584 
+ad13: TIMEOUT - READ_DMA48 retrying (1 retry left) LBA=353030400 
+ad13: TIMEOUT - READ_DMA48 retrying (0 retries left) LBA=353030400 
+ad13: FAILURE - READ_DMA48 timed out LBA=353030400

After that i have turned off compression and started bonnie++ to induce the load on AD13 during this test my AD12 - GPT/DISK2 also started reporting problems and fell out of storage ARRAY!!!

Code: 
Mar 17 10:00:48 kernel: ad12: WARNING - ATA_IDENTIFY taskqueue timeout - completing request directly
Mar 17 [B]10:00:48[/B] kernel: ad12: WARNING - ATA_IDENTIFY freeing taskqueue zombie request

Plus:

Code: 
Mar 17 [B]10:00:50[/B] pc last message repeated 368 times
Mar 17 10:00:50 pc root: ZFS: zpool I/O failure, zpool=idata error=28
Mar 17 10:00:50 pc last message repeated 17 times
Mar 17 10:00:50 pc root: ZFS: vdev I/O failure, zpool=idata path= offset= size= error=
As AD12 and AD13 are both connected to the same controller

Code: 
ad12: 238475MB <Seagate ST3250824AS 3.AAH> at ata6-master SATA300
ad13: 305245MB <WDC WD3200BEVT-00A0RT0 01.01A01> at ata6-slave SATA300

So where i should dig now? AD12 has 36 reallocated blocks for a long time... Can this be a problem?
 
you could try to test on another controller with different sata-cable or check the cable in the first place.
 
1) Have you checked your sata/power cabling and tried with a different sata cable / different power plug for the disk?
2) If your 2.5" disk is inside a caddie/dock, remove it from the dock and see if the errors persist

In the past, I've seen really weird issues related to bad sata cables / cable connectors and also really really weird issues when I was using a 2,5" disk inside a converter dock for 3,5" formfactor.
 
You must log in or register to reply here.
Back
Top