Hi everyone, I have searched but could not find anyone with similar situation.
I’m experiencing a really annoying and a bit disturbing issue with my HDDs to store the chia plots. I would like to ask for help or comments about all this. Or simply share your experience.
Basically I bought 10 disks from Seagate Iron Wolf Pro Nas (ST14000NE0008). After 3 months 3 of them have failed, which despite the low statistical sample, mean a 30% failure rate. This is shocking to me and was unexpected.
To make things worse, once one disk fails in seems chia guy gets lost and if the PC is set as harvester the entire machine becomes disconnected from the main node. I have now learned ho to identify the defective disk and after removing from the folder list at least other disks are farming OK.
But the real problem is the failing disks. Everytime this happens I lose about 13TB of plots!
I wonder if this is happening because I fill up the disks too much and once bad sectors arise they cannot be relocated?
Also is there any way I can recover the files?
The HDD once it stars failing is not recoverable, it starts making strange sounds and I cannot coy anything from it….
I share some images of the error crystal disk program show regarding the disk with issues.
Any comments on how to proceed will be appreciated. The disk is covered by the store warranty and the iron wolf recovery (is this worth the process?)
I have bought five internal hdds for chia farming in total. Note that all drives were advertised as new by the shop were I bought them from. 1x seagate iron wolf 16tb and 4x seagate exos 16tb. Three of the exos drives and the iron wolf started making the clicking noise of death (Hard Drive Click of Death Explanation and Live Demonstration - YouTube) and failed, either at the first start or after some plotting/farming. I sent them back and only use external drives now (with the exception of two internal exos drives). The interesting thing is that many of my external drives are in fact seagate exos 16tb in a external enclosure (I have around 15 such external drives). None of these have failed, so maybe the internal ones that I bought were less tested by seagate before shipment, or maybe I got some previously broken/repaired ones (the price was indeed very good).
As for your practical issue with data recovery: I did it the easy was - simply replotted. Not worth it to mess with endless recovery processes that might result in corrupted plots anyway. Much easier to just replot if you have a fairly ok plotting machine IMO.
I leave like 50-60Gb in each disk. bu the smallest file is 106GB. so if the disks needs to change of location of a currupted file, ther will not be enough space. I’m not sure if that can be a problem in case of some bad sectors…
Hi indeed I should have shared the pictures! here they are. I suppose this is fine in terms of vibration insulation. but let me know if it looks reasonable. temperatures are monitored and tipicaly sit below 40-42ºC.
more over the PSU is a750 corsair and the adapter is LSI® SAS 9201-16i PCI Express® to 6Gb/s SAS Host Bus Adapter. I wonder if the adapter is causing some damage? Is that possible?
I run Max of HDD 18 disks in each setup. I did Have 2 power failure from the grid, but I mean that should not just kill the hdds…
Another theory: sata lack of power/overload.
How many drives per each outlet on the sata/accessory rail?
The voltage may drop and your PSU may not detect it, especially likely during initial spin-up when you power on the system (consumes ~ double the normal power)
Some of your hdd’s has a power convertor connector consist of 4 wire strings which is not right. You shold have 5 wire strings on each connector to get an extra 3 volts power for electronic controllers board on the hdd. This is definitely the problem you are having. Also your power supply should have enough watts to handle all those drives.
I was not aware about this! interesting. You think that really can be causing the problem?
Overal the PSU is more than enough (750w) for 15 HDDs, but I had no idea about these 5 wires…
many thanks for the tip. I will try to read a bit to understand how these convertors work…
I used these connectors becuase the PSU only have a very limited amount. Maybe I should just try to get from other PSU. I had no idea that this could lead to a HDD failure! note that theses HDD worked fine for 3 months on these connectors. so I should eliminate the conversors and connectors all together in all setups. that is crucial right?
It might worked for a while but even then it was not working the way it should and because of that it caused all those problems afterward. Anyway, try to use the correct convertor connectors and you should be fine.
Thanks a lot Behrooz. I will go to the setup and cross check everything. I will look to see if there is a correlation between failing HDD and those adaptors…once again many thanks for your eye attention!