1 |
On Sun, Oct 20, 2013 at 1:29 PM, Mark Knecht <markknecht@×××××.com> wrote: |
2 |
> That said I'm not clear about the virtual console point. I thought the |
3 |
> virtual consoles were Alt-Ctl-F[1,2,3,..] When this even occurs my |
4 |
> keyboard isn't working so I don't know how to get there. You must mean |
5 |
> something else? |
6 |
|
7 |
It will only be helpful if the console is displayed when the panic |
8 |
occurs. This is helpful when the panics tend to happen when you're |
9 |
away from the system. |
10 |
|
11 |
> To make progress with /etc/local.d/kdump.start it turns out I also |
12 |
> needed to enable |
13 |
> |
14 |
> File Systems -> Pseudo File Systems -> /proc/kcore |
15 |
> |
16 |
> The Gentoo wiki only talked about vmcore. |
17 |
|
18 |
Feel free to update it, after you're sure you have everything figured out. |
19 |
|
20 |
> At this point I'm a bit beyond my depth. If the hang created by |
21 |
> Virtualbox isn't a panic, but my keyboard is completely locked up, |
22 |
> then I don't know how I'm going to issue the magic sysrq to get the |
23 |
> dump process going. |
24 |
|
25 |
Are you SURE that it is COMPLETELY locked up? As in alt-sysrq-b |
26 |
doesn't reboot? I've found that this almost always works, even if |
27 |
sysrq otherwise appears to not work (most of the other options won't |
28 |
appear do anything in a panic with the display not on a virtual |
29 |
console while in a graphics mode). |
30 |
|
31 |
> I get a error screen and the system reboots. The first time I did it I |
32 |
> had a bunch of disk activity - presumably stuff being copied to either |
33 |
> kcore or vmcore - and then much later X & KDE came up running a single |
34 |
> processor. This seemed like a positive result. |
35 |
|
36 |
Nothing gets "copied" to kcore/vmcore. The state of the previous |
37 |
system is already in RAM, or at least it was until KDE launched and |
38 |
overwrote everything. You need to have it boot to a crash kernel and |
39 |
rescue shell for it to be of any use, unless you just want it to |
40 |
auto-reboot to minimize downtime. |
41 |
|
42 |
> QUESTION 1: This machine has 24GB DRAM. I've set crashkernel=256M and |
43 |
> hoped for the best but don't know if that's a good setting. |
44 |
|
45 |
It is probably fine - it just needs enough space to hold the kernel |
46 |
and initramfs. |
47 |
|
48 |
> |
49 |
> QUESTION 2: Am I correct that the captured dump output is going to be |
50 |
> a file that's roughly 24GB? Maybe this takes hours or something being |
51 |
> that it's that big and I'm presumably saving it to a RAID6 which is |
52 |
> doing a lot more parity calcs all in single processor mode. Is there a |
53 |
> way to estimate how long I'd have to wait to even get to a login |
54 |
> prompt? |
55 |
|
56 |
You won't get a login prompt unless you have some script set to |
57 |
auto-save the dump file. Typically you'll set things up to just |
58 |
launch a root shell. Saving a 24GB file shouldn't take more than a |
59 |
few minutes with nothing else touching the hard drive. |
60 |
|
61 |
Per the wiki you should be running: |
62 |
kexec -p /[path-to-kernel] --append="root=[root-device] single irqpoll |
63 |
maxcpus=1 reset_devices" |
64 |
|
65 |
Note the single option in there - you're not going to get a login |
66 |
prompt. It will just dump you at a shell. If you've booted into that |
67 |
kernel then /proc/vmcore will contain a core file from the paniced |
68 |
kernel. |
69 |
|
70 |
If you just rebooted normally then I don't think /proc/vmcore will |
71 |
even exist. That was the problem I was having a while back - kexec |
72 |
wasn't actually having any effect. I have no idea if whatever broke |
73 |
it has been fixed. |
74 |
|
75 |
Rich |