I recently saw my logs start to fill up and see that i'm having a repeated "SATA Link Down" error (pic attached). They cycle through for ata3, ata4, ata5, and ata6. I have my hard drives attached to a SAS expander and I re-seated all SAS to SATA cables on both the expander side and hard drive sides and still get the errors. I have 3 cables total splitting out to 9 drives.

 

First I thought it was a bad cable, that might be connected to the four drives. I checked and each of the errored drives are on different cables. ata 3 is on one cable, ata 4 is on a different cable and ata 5 and 6 are both on the third cable.

 

I thought it could be the PCI slot and so tried a different one but still receiving an error. I had other PCI adapters which I removed thinking that I was taking up too many lanes but that didn't fix it.

 

I'm still able to access all of my disks just fine through CLI and nothing is being emulated. 

 

Could the SAS expander be going bad? I'm not too sure what else to check.

 

 

afterdark-diagnostics-20220509-1603.zip

Strange but there are no devices on those ports anyway, only on ATA1 and 2, so it's not a problem other than the log spam, you could use those 4 ports to connect some devices, instead of for example using so many USB, more reliable and the spam should stop.

 

 

I had thought that these were the drives correlated to the error but maybe I'm wrong.

 

[9:0:3:0]    disk    ATA      WDC WD100EZAZ-11 0A83  /dev/sdq   /dev/sg17
  state=running queue_depth=32 scsi_level=7 type=0 device_blocked=0 timeout=30
  dir: /sys/bus/scsi/devices/9:0:3:0  [/sys/devices/pci0000:00/0000:00:01.1/0000:02:00.0/host9/port-9:0/expander-9:0/port-9:0:3/end_device-9:0:3/target9:0:3/9:0:3:0]
[9:0:4:0]    disk    ATA      WDC WD100EMAZ-00 0A83  /dev/sdr   /dev/sg18
  state=running queue_depth=32 scsi_level=7 type=0 device_blocked=0 timeout=30
  dir: /sys/bus/scsi/devices/9:0:4:0  [/sys/devices/pci0000:00/0000:00:01.1/0000:02:00.0/host9/port-9:0/expander-9:0/port-9:0:4/end_device-9:0:4/target9:0:4/9:0:4:0]
[9:0:5:0]    disk    ATA      WDC WD100EMAZ-00 0A83  /dev/sds   /dev/sg19
  state=running queue_depth=32 scsi_level=7 type=0 device_blocked=0 timeout=30
  dir: /sys/bus/scsi/devices/9:0:5:0  [/sys/devices/pci0000:00/0000:00:01.1/0000:02:00.0/host9/port-9:0/expander-9:0/port-9:0:5/end_device-9:0:5/target9:0:5/9:0:5:0]
[9:0:6:0]    disk    ATA      WDC WD161KRYZ-01 1H01  /dev/sdt   /dev/sg20
  state=running queue_depth=32 scsi_level=7 type=0 device_blocked=0 timeout=30
  dir: /sys/bus/scsi/devices/9:0:6:0  [/sys/devices/pci0000:00/0000:00:01.1/0000:02:00.0/host9/port-9:0/expander-9:0/port-9:0:6/end_device-9:0:6/target9:0:6/9:0:6:0]

 

Another thing I wanted to note is that mover is going really slow at around 25 - 50 KB/s. Not sure how that would be related or not. I only noticed this because i was going to change out my cache drive and saw that it was getting pretty full so I set the appdata share to CACHE:YES and invoked the mover.

 

I had about 22GB on it since yesterday and it's still got 17GB left after letting it run overnight.

Edited May 10, 20224 yr by insomniak

58 minutes ago, insomniak said:

I had thought that these were the drives correlated to the error but maybe I'm wrong.

 

No, there are no drives connected to those ports, those are the onboard SATA ports, only the first two have devices connected:

May  9 09:43:30 afterdark kernel: ata1.00: ATA-9: SanDisk SDSSDHII240G, 150366401046, X31200RL, max UDMA/133
May  9 09:43:30 afterdark kernel: ata2.00: ATA-9: Samsung SSD 850 EVO 500GB, S21HNSAG408848P, EMT01B6Q, max UDMA/133
May  9 09:43:30 afterdark kernel: ata3: SATA link down (SStatus 0 SControl 300)
May  9 09:43:30 afterdark kernel: ata4: SATA link down (SStatus 0 SControl 300)
May  9 09:43:30 afterdark kernel: ata5: SATA link down (SStatus 0 SControl 300)
May  9 09:43:30 afterdark kernel: ata6: SATA link down (SStatus 0 SControl 300)

 

 

Weird, the mother board is a Gigabyte GA-7PESH2 and it only has the two SATA ports which are being used.

 

I tried unplugging ata1 since it's not being used right now and now it booted up fine without errors which is weird.

15 minutes ago, insomniak said:

and it only has the two SATA ports which are being used.

Chipset still supports 6, so the other ones will always be link down.