1 |
Mick wrote: |
2 |
> On Thursday, 2 January 2020 18:48:03 GMT Dale wrote: |
3 |
>> Ian Zimmerman wrote: |
4 |
>>> On 2020-01-01 18:09, Dale wrote: |
5 |
>>>> As some may recall, I have a 8TB external SATA hard drive that I do |
6 |
>>>> back ups on. Usually, I back up once a day, more often if |
7 |
>>>> needed. Usually I turn the power on, mount it, do the back ups, |
8 |
>>>> unmount and turn the power back off. Usually it is powered up for 5 |
9 |
>>>> minutes or so. When I unmount it tho, I sometimes notice it is still |
10 |
>>>> doing something. I can feel the mechanism for the heads moving. It has |
11 |
>>>> a slight vibration to it. Questions are, what is it doing and should |
12 |
>>>> I let it finish before powering it off? I'd assume that once it in |
13 |
>>>> unmounted, the copy process is done so the files are safe. I guess it |
14 |
>>>> is doing some sort of internal checks or something but I'm not sure. |
15 |
>>> I have observed the same thing. But in my case, I also disconnect the |
16 |
>>> _cable_ from the computer to the enclosure when I am done ... and still |
17 |
>>> the drive activity goes on. From that I conclude that it is the drive |
18 |
>>> circuitry itself doing some kind of internal housekeeping, and there is |
19 |
>>> no point in worrying about it because one would wait forever for it to |
20 |
>>> end. |
21 |
>> That's one thing that makes it unnervey. I'll put my hand around to the |
22 |
>> back and feel those little bumps. I wait until I think it is done but |
23 |
>> just as I'm about to power it off, it bumps again. It's so |
24 |
>> unpredictable, I never know if it is done doing its thing or not. Just |
25 |
>> like now, it's unmounted, did that during last reply, it hasn't did the |
26 |
>> bump thingy while reading your reply or me typing mine in so far. Now |
27 |
>> as soon as I reach around to turn it off, it'll likely do the bump thing |
28 |
>> again. lol |
29 |
> I suspect it is now reading your mind! Is this a case of AI?! |
30 |
> |
31 |
> LOL! |
32 |
> |
33 |
> |
34 |
>> One thing is for sure tho, if you unplug the cable, whatever it is |
35 |
>> doing, it's internal. Sort of hard for the puter to be doing something |
36 |
>> when it isn't connected. That narrows the options down a lot. That's a |
37 |
>> good piece of info there. |
38 |
>> |
39 |
>> Thanks. |
40 |
>> |
41 |
>> Dale |
42 |
>> |
43 |
>> :-) :-) |
44 |
> In case you missed it in a previous post - have you compiled CONFIG_DM_ZONED |
45 |
> in your kernel to see if its performance changes? |
46 |
> |
47 |
> https://zonedstorage.io/linux/config/ |
48 |
> |
49 |
> It's behaviour may not change whatsoever, if the drive only has an internal |
50 |
> (SMR firmware) data write mechanism. However, if the drive is exposing an I/O |
51 |
> API to the OS, then you could well see a difference to how its data storage |
52 |
> bumps, spins and shakes as it flushes its journal and goes about its garbage |
53 |
> collection process when the Linux kernel 'talks' to it. |
54 |
> |
55 |
|
56 |
|
57 |
It seems that is not enabled in my kernel. |
58 |
|
59 |
|
60 |
Symbol: DM_ZONED [=n] |
61 |
|
62 |
|
63 |
It took some effort to find and enable everything to get that but I |
64 |
finally found them all. It seems three things has to be enabled in |
65 |
order for that option to be visible. Then after that, make wouldn't |
66 |
work. I checked the gcc settings, emerged libtool and checked on |
67 |
everything else with no improvement. Then I copied the .config file to |
68 |
a safe spot and did a make clean. When I copied the file back and ran |
69 |
make again, it worked. No clue on what was up with that. You know my |
70 |
luck tho. ;-) |
71 |
|
72 |
I'll try to reboot the new kernel in a bit. It's building at the |
73 |
moment. Thanks for posting about this. I did not see it in other |
74 |
replies. I thought it might be in Rich's but didn't see it. The extra |
75 |
nudge was helpful. |
76 |
|
77 |
Dale |
78 |
|
79 |
:-) :-) |