1 |
On Mon, Jul 1, 2013 at 2:10 PM, Paul Hartman |
2 |
<paul.hartman+gentoo@×××××.com> wrote: |
3 |
> On Tue, Jun 25, 2013 at 5:51 PM, Mark Knecht <markknecht@×××××.com> wrote: |
4 |
>> This is related to my thread from a few days ago about the |
5 |
>> disappointing speed of my RAID6 root partition. The goal here is to |
6 |
>> get the machine booting from an SSD so that I can free up my five hard |
7 |
>> drives to play with. |
8 |
>> |
9 |
>> SHORT SUMMATION: I've tried noninitrd and noraid in the kernel line of |
10 |
>> grub.conf but I keep booting from old RAID instead of the new SSD. |
11 |
>> What am I doing wrong? |
12 |
> |
13 |
> Here are some things I would try to narrow it down: |
14 |
> |
15 |
> Put raid=noautodetect on your kernel commandline to prevent the kernel |
16 |
> from auto-assembling the array |
17 |
> |
18 |
> It sounds like you are pretty sure it is at least using the boot |
19 |
> sector of the new drive, so I am thinking it is possible there is some |
20 |
> weird combination of using a boot sector from one drive to get you |
21 |
> into the boot partition of another drive. If the old boot drive is |
22 |
> still attached, you could try moving/renaming the grub config or whole |
23 |
> grub folder on the old boot drive to make sure it's not getting used. |
24 |
> |
25 |
> If that doesn't give any clues, I would physically unplug the cable of |
26 |
> every drive other than the SSD (if that is realistic based on your |
27 |
> filesystem layout) and see how far it gets. Maybe if it fails you can |
28 |
> figure out what it's trying to access on the other disks. |
29 |
> |
30 |
> As far as the RAID I think there are at least a few different ways an |
31 |
> mdraid array comes to be assembled: |
32 |
> - your initramfs does it |
33 |
> - your kernel does it (only for a RAID with v0.90 superblock) |
34 |
> - init script does it (/etc/init.d/mdraid) |
35 |
> - udev does it (/lib64/udev/rules.d/64-md-raid.rules) |
36 |
> - you manually do it later on using mdadm |
37 |
> |
38 |
> Viewing dmesg output from around the point where boot begins and the |
39 |
> RAID is assembled might give you some clues about who's doing what. |
40 |
> |
41 |
> I recently upgraded my machine and disks and am using UUID and labels |
42 |
> for everything, I can literally boot from either the old HDD or new |
43 |
> HDD from my BIOS boot menu, plugging them into the motherboard in any |
44 |
> order, and either one will work properly, even though the /dev/sdX |
45 |
> assignments might change from boot to boot. You can use the blkid |
46 |
> command (as root) to see the labels and UUIDs for all of your drives |
47 |
> and partitions. |
48 |
> |
49 |
> Good luck, |
50 |
> Paul |
51 |
> |
52 |
|
53 |
Hi Paul, |
54 |
Thanks for the interest and sorry for the delay in my response. |
55 |
I've ended up going in a slightly different direction in the process |
56 |
which as of yet hasn't yielded much except more work for me. No |
57 |
response is necessary to this post, although I'm always interested in |
58 |
what folks are doing and thinking. |
59 |
|
60 |
This post is nothing more than a status report. |
61 |
|
62 |
First, the purpose of this work: my existing machine has 5 HDDs |
63 |
hooked to the internal SATA controller, an unused SDD and a number of |
64 |
external USB drives holding video & Windows VMs. The RAID6 is slow (my |
65 |
opinion) and I want to investigate other RAID architectures in the |
66 |
machine with the goal of eventually using one (RAID1, RAID5, RAID6, |
67 |
RAID10) in a better optimized and tested setup. |
68 |
|
69 |
The first path I went down was to rsync the exiting Gentoo / to the |
70 |
SDD. For whatever reason that never booted correctly. |
71 |
|
72 |
To get past the issue above I just created a new Gentoo install on |
73 |
the SDD from scratch. This was easy to do in a chroot while I do my |
74 |
trading work throughout the day. As per other posts I decided to |
75 |
create a new boot partition on the SDD with an eye toward possibly |
76 |
using grub2 later. (The SDD is sda, at least in the RAID6 environment. |
77 |
/dev/sda1 is the new boot, /dev/sda2 is the new SDD root) However in |
78 |
the beginning my intention was only to place the SDD kernel on the new |
79 |
boot partition. The HDD RAID6 kernel remains on the HDD boot partition |
80 |
and I continue to use grub on that boot partition to get the machine |
81 |
going. For the SDD boot I just point the root & kernel commands to the |
82 |
first partition on the SDD. |
83 |
|
84 |
This has worked and the SDD boots reasonably well, but it is |
85 |
fragile. It appears that drive ordering changes from boot to boot |
86 |
depending on which USB drives are attached so I probably need to move |
87 |
to something more like your UUID methods. |
88 |
|
89 |
Lastly, because I am still running on the original RAID6 setup I do |
90 |
not want to touch the internal SATA cabling at all. It's important to |
91 |
me that the currently working setup continue to work. I'll just have |
92 |
to struggle along with making the setup more reliable using both the |
93 |
HDD & SDD setups. |
94 |
|
95 |
Due to everything above I've not yet done anything with testing of |
96 |
new RAIDs. Maybe later this week. Who knows? |
97 |
|
98 |
Cheers, |
99 |
Mark |