New Pool with new hardware Unhealthy

[andreaconfa]

Hi all,
2 days ago I replaced my old disk with 6 brand new Saegate Barracuda 4TB drives.
I connected it with a new sata controller to my nas systems and settled up a new pool, then transferred all the data.
Now the alla the file is copied to the new pool I disconnected the old pool.

To check if all is working fine I issued via shell command a scrub and to my surprise there are errors.
Can someone help me to understand what can be the problem?

I’m running TrueNAS scale on VMware ESXi. The VM is provisioned with 16GB of ram dedicated only to truenas and the sata controller is passed to the va via PCIE passtrought. Used this configuration for over 4 years without any problem.

EDIT: The drivers are Saegate Ironwolf 4TB and not barracuda

Attached some screenshots:

Thanks in advance :)

 

[NugentS]

CHKSUM errors are usually (but not always) the results of bad cabling. First reseat all the cables, then replace as nessesary
Given that you haven't posted your hardware we can't go much further than that........ hint hint

 

[Alecmascot]

Baracudas are SMR , will not work properly with TrueNAS.

 

[NugentS]

shouldn't be producing chksum errors though - although if they are SMR thats another issue

 

[Alecmascot]

shouldn't be producing chksum errors though - although if they are SMR thats another issue

The SMR drives drop off the buss when they are overloaded with Writes and this causes all kinds of grief.
If they are new maybe the OP can get them replaced.

 

[NugentS]

yeah - either.
1. Don't use TrueNAS
2. Return the SMR drives and use CMR

 

[andreaconfa]

Thanks all for the replies.
Can the issue be related also to memory issue? I ram I memtest and found some problems to it.
I can replace with other drives the barracuda…. What is the best option in the same price range ?
What alternative to truenas I can use?

EDIT: I made a mistake. My drives are Saegate IronWolf 4TB and it appears to be CMR (https://www.amazon.it/gp/product/B01LOOJBQY/ref=ppx_yo_dt_b_asin_title_o06_s00?ie=UTF8&psc=1)

Talking about system specs i'm running Vmware ESXi on a AMD Ryzen 5 2600x (not ideal but working fine) with 64GB of ram.
Then in vmware i created a VM for TrueNAS with dedicated 16GB of ram and PCI Passtrought for the SATA Controller (a cheap one from amazon based on ASM1166 https://www.amazon.it/gp/product/B09K4WKHKK/ref=ppx_yo_dt_b_asin_title_o03_s00?ie=UTF8&th=1)

Thanks @NugentS @Alecmascot

 

[andreaconfa]

now i fixed the ram and changed sata cables. what is the next steps? Delete the corrupted file and scrub again?

 

[NugentS]

Yes, or restore the corrupt files from your backup. YOu need to know if the error is still there or not

 

[andreaconfa]

Yes, or restore the corrupt files from your backup. YOu need to know if the error is still there or not

i deleted the file and i'm running again the scrub.
Is the controller compatible or is better to have a sas HBA like this Dell H20 with IT firmware? https://www.ebay.it/itm/234202819061

 

[NugentS]

Those SATA controller are generally not reccomended. Most are utter shite with the occaisonal OK one.
Much much MUCH better to use a decent HBA.
The one you mention (Dell HBA) is in theory a good one - very well supported. The only issue is that its a bit slow IF (note the IF) you want to run significant numbers of SSD's on the HBA. For HDD's its absolutely fine.

 

[andreaconfa]

ok so maybe i return back the sata controller from amazon and i buy this used HBA Sas controller. Hoping that this scrub will end without error..... now i'm at 18% without errors.....