1 |
On Fri, Feb 26, 2010 at 9:59 AM, Volker Armin Hemmann |
2 |
<volkerarmin@××××××××××.com> wrote: |
3 |
> On Freitag 26 Februar 2010, Mark Knecht wrote: |
4 |
> |
5 |
>> |
6 |
>> The machine _mostly_ crashed while running badblocks. I say mostly |
7 |
>> because the mouse is still alive but I can no longer ssh in and cannot |
8 |
>> open a terminal on my wife's desktop or get to the console. |
9 |
> |
10 |
> because it is not crashed but waiting for the ide timeouts. |
11 |
|
12 |
So if I let it continue running is it going to come back in the next |
13 |
hour or two? I am assuming the IDE timeouts are because the drive is |
14 |
having trouble, correct? That's the theory here? If so then unless the |
15 |
software can mark them bad and somehow create good files out of bad |
16 |
then I'm still left with a machine that is going to need serious work |
17 |
done before it's a happy box again, correct? |
18 |
|
19 |
On the other hand, because I have reasonably good user backups |
20 |
(although no real system backups) right now if I bite the bullet and |
21 |
build the machine then when my wife gets it back it's hopefully going |
22 |
to be more reliable, wouldn't it? |
23 |
|
24 |
I'm thinking that maybe I just copy a little stuff off the box - /etc |
25 |
and the like - and then boot the machine with the Gentoo install CD or |
26 |
System Resuce CD and see what the drive is doing? |
27 |
|
28 |
That doesn't cost me anything to look around, but if SMART won't turn |
29 |
on and badblocks is suggesting the drive is having trouble maybe |
30 |
running something like badblocks and actually __marking__ blocks as |
31 |
bad and then reloading Gentoo would work in the long run? (A lot of |
32 |
work though.) |
33 |
|
34 |
I'm really not interested in buying new drive because the machine is |
35 |
ATA100/133 and if it's not the drive then the money is wasted for a |
36 |
new machine. The cheapest at NewEgg is about $40. Why spend the buck |
37 |
for an old Intel Centrino machine? |
38 |
|
39 |
> |
40 |
>> |
41 |
>> I tried to Ctrl-C out out of badblocks here (this is running shelled |
42 |
>> in) before I figured out it was a total crash which messed up the |
43 |
>> terminal a bit but you can see what it was reporting before the crash |
44 |
>> |
45 |
>> dragonfly ~ # badblocks -sv /dev/hda |
46 |
>> Checking blocks 0 to 156290903 |
47 |
>> Checking for bad blocks (read-only test): 89360960done, 35:00 elapsed |
48 |
>> 89360961done, 35:09 elapsed |
49 |
>> 89360962 |
50 |
>> 89360963 |
51 |
>> ^C^C18% done, 35:27 elapsed |
52 |
>> |
53 |
>> So, there seem to be problems, possibly with the drive, or maybe it's |
54 |
>> some sort of overheating problem on the processor and this was just |
55 |
>> the way the processor failed before the crash? |
56 |
>> |
57 |
>> I ran memtest86 night before last for 8 hours and had no memory |
58 |
>> problems. I'll remove memory and PCI cards, reseat everything, and |
59 |
>> then see what happens. |
60 |
> |
61 |
> protip: if you are running badblocks (or ddrescue) on a probably damaged |
62 |
> device - attach it with an usb adapter. That way your box is still usable. |
63 |
> |
64 |
> /me hates linux kernel for making processes in D unkillable and sucking very |
65 |
> much on diskio. |
66 |
> |
67 |
> |
68 |
|
69 |
Good inputs. Thanks! |
70 |
|
71 |
Cheers, |
72 |
Mark |