TimeMachine loses SMB connections constantly

Hi,
These disks are Seagate ST8000DM004, same than the previous I had.
There are no errors reported. I’ll run the diagnostics again as soon as I can.

Thanks for your help

Looks like those are SMR drives, at least according to a quick google search. That could be a part of your issue.

but it did work perfectly with the same drive types/models since 3-4 years !

Not saying this IS your issue, just something you should read about, just in case. It could explain why your previous drives failed and you are having issues with the new ones. Honestly, I don’t know enough about it to say if it applies to you or not. but it seems to fit.

@dosborne Thanks for that drive lesson. Interesting. I would have assumed that all NAS drives are appropriate (ie: CMR) but it appears that is not necessarily the case with WD. Good to know!

Thanks.

This is not the first time I use this serie of disks from Seagate, my previous QNAP was using the same, and they worked around 8 years without any issue…
I’ve launched last night a test on the drives, it’s still ongoing (not sure it is significant, but one is 40 % and the other 10 %).
Is there a tool I can use to test individual drive perfs inside the QNAP ?

I have some Seagate drives that have been running for over 15 years. The manufacturer is not the problem, the SMR technology is (or may be in your case). The issue of CMR vs SMR vs use in RAID is the important factor and I encourage you to investigate this.

I’ve used SEATools (or whatever they call it now) to perform diagnostics on Seagate drives in the past. You may be able to get some information via SMART in the NAS, but it wouldn’t be as comprehensive.

yes, but as my drives are active in the NAS, I would like to check them with some tools in it if that exists.

I tried to install DA Drive Analyzer, but it fails starting…

Like written before I started a complete test on both drives yesterday night. Now HDD2 is 70 %, HDD1 is 10 % of completion. Don’t know if the tests really run in parralel, but if it is the case, something’s going wrong with HDD1…