Kernel ATA Error Messages while using HDD Cache

    Hi everyone,


    I'm new here and just started to work with OMV because I build me a NAS/VM Server for private usage. I'm an embedded software developer from Germany, so I'm able to understand most technical things, however I'm not a Linux-expert. :) I just wanted to give you some background and sorry for my bad English.



    Ok, so let's start to describe my problem. I brought me a DELL T20 and put the hardware into a new case, brought 8 WD Red with 3TB, a SATA Controller, a SSD and some additional RAM and put everything together. Because I wanted to use the T20 for other services then a file server, too and because just out of curiosity, I installed XenServer 7 on it. Then I installed OMV3 into a VM and passed the SATA controller though to the VM. At first everything seemed to be ok, but after I enabled the HDD Write Cache in OMV, the Kernel started to throw ATA Error messages while coping files onto the server:


    dmesg:



    At first I thought one of my disk were defect, so I permord a extended SMART test on every disk:


    smartctl -a /dev/sda:


    smart.txt


    As far as I can tell, the SMART data seems ok. So I started to switch the SATA cabels to other HDDs, traced back the ATA Channel to the actual device and noted the Serial-Number to determine if one special HDD is causing the problem, but I could not see any similarities between them. I got this errors with 6 HDDs. So for me, and I'm hoping that i'm right, not a HDD is causing the problem. I suspect it is the SATA controller or the fact, that I use PCI-Passthrough.


    Anyway the write performance seems to be pretty bad, too:


    dd if=/dev/zero of=testfile bs=1G count=1 (with cache enabled):


    dd if=/dev/zero of=testfile bs=1G count=1 (without cache enabled):


    Here some information about the raid:


    mdadm --examine --brief --scan --config=partitions:


    mdadm --detail /dev/md127:


    I assigned 2 V-Cores to the VM and 6GB RAM. I also disabled the idle3-timeout on the WD Red HDDs. Any ideas what could cause this problem? Can you help me? ;( I'm using OMV 3.0.64.



    Greetings


    m0nKeY

    Thanks, but even though there seems to be different FW versions of the WD30EFRX, I can't find a firmware update. But I disabled the idle3 timeout, too.


    I will try another SATA/SAS controller this week.

    Yesterday I changed the SATA Controller (DeLOCK 89384, PCIe 2.0 x2 ) to an SAS/SATA Controller (HP H220, PCIe 3.0 x8 (650933-B21) ) and everything works fine now. I got write-rates ~500Mb/s. I will send the DeLOCK Controller back as malfunction.

    I am pretty late, but I have 100% same situation.


    10 HDDs, Openmediavault, DeLOCK 89384, all SMART ok, "retrying FLUSH 0xea Emask 0x4" under high load.


    Edit: I contacted DeLOCK, but they dont see any issues with the card. So it would be interesting to know the exact name of the DeLock JMicron chip being used.

    cpu Intel(R) Core(TM) i5-10400 CPU @ 2.90GHz
    omv 6.9.13-1 (Shaitan)

    kernel 6.1.0-0.deb11.11-amd64

    Einmal editiert, zuletzt von godfuture ()

    Just got response from DeLock (very quick response!):

    2x JMB575 and 1x ASM1062 chip is being used on DeLOCK 89384 (current model).

    cpu Intel(R) Core(TM) i5-10400 CPU @ 2.90GHz
    omv 6.9.13-1 (Shaitan)

    kernel 6.1.0-0.deb11.11-amd64

Jetzt mitmachen!

Sie haben noch kein Benutzerkonto auf unserer Seite? Registrieren Sie sich kostenlos und nehmen Sie an unserer Community teil!

Benutzerkonto erstellen Anmelden