2.5" external hard drive is going to stumble

[cobito]

Hello. I have two external hard drives of 2.5" and 2Tb, USB 3.0: a Samsung and a Maxtor. The first one I use as data storage and the second as a backup of the first.

The backup one is brand new because I bought it a few months ago but I have only used it a couple of times for a few hours. The thing is that today I decided to make a copy of my data from scratch, that is, to format the backup one and copy everything. In total there are about 1.3TB and when it starts, everything seems to be going well. The rate is between 60-100MB/s. When it has been going for a while, the data drive sometimes stops being active while the backup one remains active. This happens in fits and starts: sometimes it copies at cruising speed and other times it gets stuck for a good while. It does it quite irregularly.

If I put the backup drive to my ear, I don't hear any strange noises but rather normal activity. I don't suspect the data drive because it simply runs out of things to do and sometimes even goes into power-saving mode.

I ran the HDD regenerator on the backup drive and it gives no errors or delays. When measuring the read speed, it gives between 80 and 90MB/s. I suppose that writing is slower than reading and from that I would expect a somewhat lower load on the data drive. But sometimes it takes too much time doing nothing and above all, the stops are very irregular.

Both drives are formatted in EXT4 with partitions encrypted with LUKS. If I look at the system load, I see that it is at 100% but it turns out that 60-70% of the load comes from the WA section (waiting times for input/output transactions), that is, from waiting times for the hard drive.

I don't remember how the copy was the first time I did it, but it seems to me a bit strange behavior. What might be going on?

Thanks.

[whoololon]

The first nonsense that comes to mind is that it may be due to the USB port's power supply. You can try it on another computer or take it out and try it as an internal... if the case is one of those with external power supply, I'll cut it off (as long as it works well, of course).

This has happened to me with miniPC motherboards from schools, where they have several things connected and when I connect mine (a Travelstar from the year of the cough of 250 GB) it starts transferring data but after a minute and a half it starts to decline until it gives an error... while at my house I do all the crap I want and it holds up like a champ.

The second, that although the SMART table does not reflect anything, the disk is physically damaged.

But anyway, don't pay too much attention to me.

[kynes]

That's the same thing that happens to me with some 3.0 hard drives, it seems that there are certain incompatibilities of some chipsets with some controllers, since I have two identical enclosures that present the same problem on my laptop on the 3.0 port, but they have no problem on 2.0 ports and on other computers, and other 3.0 enclosures have not given me any problems on my laptop. I have tried it with magnetic disks, with hybrid disks and with ssd, and in all cases they present the same problem. I initially attributed it to a power issue, but I had the option to connect an external power supply to the enclosure, and it still presented the same problem. The solution I applied was to use a usb 3.0 hub, it stopped presenting that problem using an intermediate device.

[whoololon]

@kynes Well yes, it makes more sense about the issue with the controller and the USB 3.0 port.

In my case, they were B85 equipment (GA-B85M-D3H rev2.1) and the cases that I spent some B-Move as beautiful as they are old.

[cobito]

@whoololon It's a drive that comes pre-assembled and the casing has no screws. I don't want to try to disassemble it, break a tab or anything else and then not have it fit properly.

Regarding having physical problems, supposedly HDDRegenerator checks the entire physical surface of the drive. There should have been something seen there and it found nothing, curiously, without any apparent performance drops during the check.

I had also thought about the power supply, but when the copy gets stuck, the drive doesn't sound like it has that problem but rather that it has normal activity. On the data drive I've had that kind of problem on other PCs and the noise it makes is quite evident.

What @kynes comments about is more like what happens to me. The thing is that I made the copy and when it finished I did a diff to see if the data had been corrupted, but both drives are identical.

The curious thing is that during the diff check, both drives were running at cruising speed. I wasn't watching all the time, but it was going at 90-120MB/s with some drops I suppose due to small files and decreasing as the drive progressed (that is, normal). That is, the cuts occur in writing but not in reading. Since I don't push it too hard, I don't even know if this has been happening to me with the data drive.

The PC I use for these things has an Asus P8B75-V with B75 chipset that has a BIOS from 2012 (I've never updated it). I don't have a PC with a couple of USB 3.x with Linux handy where I can confirm this.

Anyway. The full backup is something I do a couple of times a year and taking 7 or 8 hours instead of 3 or 4 doesn't bother me too much. The relevant thing here is that it seems that the data isn't corrupted, which is what I was most worried about.

Thanks to both of you for the answers.