Sometime I’m in a hurry and delete a VM’s disk image before removing it from libvirt/virsh, which leaves a lingering, dead reference to that VM.
The following fixes this:
sudo virsh –connect qemu:///system undefine vm_to_be_removed
Replace vm_to_be_removed with your VM name.
I had to make a few changes to /etc/default/grub to get Ubuntu 14.04 to boot with a working console. Initially, my monitor just went dark.
Changes with reasons:
GRUB_HIDDEN_TIMEOUT: if you want to see the Grub menu, this must be commented out.
GRUB_CMDLINE_LINUX: allows passing parameters to the linux startup command. Newer kernels (>3.5? or so) reinitialize the graphics card. My AMD Radeon something-or-other didn’t like this at all. nomodeset forces linux to use/keep BIOS graphics settings.
GRUB_GFXMODE: select resolution for Grub graphical starup. 1280×1024 works on my aging ViewSonic. This setting is for the Grub menu.
GRUB_GFXPAYLOAD_LINUX: specifying keep, causes Grub to keep using the previous resolution (from GRUB_GFXMODE) for the rest of the boot.
NB: after making any changes, remember to update-grub (or nothing will happen).
Contents of my /etc/default/grub after making required changes:
# If you change this file, run 'update-grub' afterwards to update
# /boot/grub/grub.cfg.
# For full documentation of the options in this file, see:
# info -f grub -n 'Simple configuration'GRUB_DEFAULT=0
#GRUB_HIDDEN_TIMEOUT=0
#GRUB_HIDDEN_TIMEOUT_QUIET=true
GRUB_TIMEOUT=5
GRUB_DISTRIBUTOR=`lsb_release -i -s 2> /dev/null || echo Debian`
GRUB_CMDLINE_LINUX_DEFAULT=”splash”
GRUB_CMDLINE_LINUX=”nomodeset”
# Uncomment to enable BadRAM filtering, modify to suit your needs
# This works with Linux (no patch required) and with any kernel that obtains
# the memory map information from GRUB (GNU Mach, kernel of FreeBSD …)
#GRUB_BADRAM=”0x01234567,0xfefefefe,0x89abcdef,0xefefefef”# Uncomment to disable graphical terminal (grub-pc only)
#GRUB_TERMINAL=console# The resolution used on graphical terminal
# note that you can use only modes which your graphic card supports via VBE
# you can see them in real GRUB with the command `vbeinfo’
#GRUB_GFXMODE=640×480
GRUB_GFXMODE=1280×1024
GRUB_GFXPAYLOAD_LINUX=keep
# Uncomment if you don’t want GRUB to pass “root=UUID=xxx” parameter to Linux
#GRUB_DISABLE_LINUX_UUID=true# Uncomment to disable generation of recovery mode menu entries
#GRUB_DISABLE_RECOVERY=”true”# Uncomment to get a beep at grub start
#GRUB_INIT_TUNE=”480 440 1″
My ISP (ding, dong) is stupid, so I need to disable IPv6.
sudo vim /etc/sysctl.conf
Add these four lines:
# No IPv6
net.ipv6.conf.all.disable_ipv6 = 1
net.ipv6.conf.default.disable_ipv6 = 1
net.ipv6.conf.lo.disable_ipv6 = 1
Adding existing logical volumes to the list of virtual disks is non-obvious.
First add the VG to Proxmox using the “Storage|Add” UI. Once done, adding existing LV to Proxmox must be done via LVM command line tools.
A logical volume needs to have the correct naming, and (surprise!) an LVM tag set. For the correct tag (and LV name) format, use the UI to create a sample volume (VM|Hardware|Add|Hard Disk).
Note: LV names must fit a certain pattern. However, names like this seem to work: vm-110-timemachine.
The example below shows a pre-existing LV (vg1/vm-110-disk-2) without the “pve-vm-110” tag Proxmox requires.
Subsequent steps add the tag, after which the volume appears in Proxmox.
# lvs -o vg_name,lv_name,lv_tags
VG LV LV Tags
pve data
pve root
pve swap
vg0 vm-110-disk-1 pve-vm-110
vg0 vm-110-disk-2 pve-vm-110
vg1 vm-110-disk-1 pve-vm-110
vg1 vm-110-disk-2
To see LV tags:
# lvs @pve-vm-110
LV VG Attr LSize Pool Origin Data% Move Log Copy% Convert
vm-110-disk-1 vg0 -wi-ao--- 500.00g
vm-110-disk-2 vg0 -wi-ao--- 750.00g
vm-110-disk-1 vg1 -wi-ao--- 750.00g
To add missing tags:
# lvchange --addtag pve-vm-110 /dev/vg1/vm-110-disk-2
Refresh the Proxmox Storage UI to see the new LV.
There are plenty of broken jsfiddles for this, so to save time, here is some code that works: boostrap-scrollspy
or preview (or download) here
This jsbin demonstrates how to use a CoffeeScript class to implement an Angular Service, using Service, Factory and Provider styles. You may have to click “Run with JS” to update Output pane. Download gist from github.
Update: by itself this is not enough, since services usually have to inherit other resources, and look something like this:
angular.module("app")
.service('thumbService', ['$log', '$q', ($log, $q) ->
...
A working example (using .factory instead of .service) from the angular-grunt-coffeescript seed app:
###
Example of a service shared across views.
Wrapper around the data layer for the app.
###
name = 'common.services.dataSvc'
class DataSvc
constructor: (@$log, @$http, @env) ->
_get: (relPath)->
return @$http.get("#{@env.serverUrl}/#{relPath}")
getPeople: () ->
return @_get('people')
getPerson: (id) ->
return @_get("person/#{id}")
angular.module(name, []).factory(name, ['$log','$http', 'common.services.env', ($log, $http, env) ->
new DataSvc($log, $http, env)
])
Some C functions accept more than one type of struct, so I wanted a Go wrapper able to pass a uintptr derived from an interface{}. Go’s reflect package makes this easy:
package main
import (
"fmt"
"reflect"
)
type USN_JOURNAL_DATA struct {
UsnJournalID int
FirstUsn uint32
NextUsn uint32
LowestValidUsn uint32
MaxUsn uint32
MaximumSize uint32
AllocationDelta uint32
}
func doitA(ujd *USN_JOURNAL_DATA) (size uintptr, align int, p uintptr) {
v := reflect.ValueOf(ujd)
t := v.Elem().Type()
size = t.Size()
align = t.Align()
p = v.Pointer()
return
}
func doitB(i interface{}) (size uintptr, align int, p uintptr) {
v := reflect.ValueOf(i)
t := v.Elem().Type()
size = t.Size()
align = t.Align()
p = v.Pointer()
return
}
func main() {
a := USN_JOURNAL_DATA{1, 2, 3, 4, 5, 6, 7}
size, align, p := doitA(&a)
fmt.Println(size, align, p)
size, align, p = doitB(&a)
fmt.Println(size, align, p)
}
Interestingly, as demonstrated above, the same implementation can be used for a func accepting either a struct pointer or an interface{}.
Go is a wonderful language, able to build complex applications with very minimal configuration information. For many apps, only import statements are required.
However, there are a couple of special import cases worth knowing about:
import . “github.com/user/repo”
From golang.org: “If a program imports a standard package using import . “path”, additional names defined in the imported package in future releases may conflict with other names defined in the program. We do not recommend the use of import . outside of tests, and using it may cause a program to fail to compile in future releases.”
import _ “github.com/user/repo”
From Andrew Gerrand (Google): “Prefixing an import
with an underscore causes the package to be imported for its
side-effects only. (that is, it’s init functions are executed and
global variables initialized)”
O/S |
Test |
Raspberry Pi |
Cubieboard |
android (busybox to nvram) |
4.8 M/B/s | ||
berryboot (raspbian squashfs) 4G(c4) |
|||
| 200M | 16 +/- .5MB/s | 7.8 +/- .1 MB/s | |
| 1G | 11 +/- .5MB/s | 8.3 +/- 1 MB/s | |
| 200M to SATA HD | 63 +/- 2MB/s | ||
| 1G to SATA HDD | 45 +/- 2MB/s | ||
| 13,727 files (/usr) from SD -> HDD (ext4) | 9.2 sec | ||
| 13,727 files (/usr) from HDD -> SD | 38 sec | ||
debian (drazbian) 16G(c10) |
|||
| 200M | 22 +/- 2 MB/s | ||
| 1G | 7.6 +/- .5 MB/s |
Raspberry Pi (and Cubieboard) SD disk images typically have two (or three) partitions: the first is Fat32 (50-100MB) and includes the boot files, while the second (>1GB) contains the main Linux image. With compression, the combined image can be reduced from several GB (of mostly empty space) to a couple hundred MB.
With a regular Linux desktop computer that has kpartx and mksquashfs installed, you can convert the second partition to SquashFS like this:
$ sudo kpartx -av image_you_want_to_convert.img
add map loop0p1 (252:5): 0 117187 linear /dev/loop0 1
add map loop0p2 (252:6): 0 3493888 linear /dev/loop0 118784
$ sudo mount /dev/mapper/loop0p2 /mnt
$ sudo mksquashfs /mnt converted_image.img -comp lzo -e lib/modules
$ sudo umount /mnt
$ sudo kpartx -d image_you_want_to_convert.img
Recent Comments