Posts made by madskillz23

Also it will try installing package mysql-server every time the installation script is run, instead of skipping it. (I installed it after the first failure with yum)

During the install script “Checking Packages…mysql-server” fails. yum install mysql-server now installs mariadb-galera-server-5.5. Is it possible the checking script isn’t handling mariadb properly?

Its an issue with your current network configuration. You might be able to filter that MAC address in DHCP from PXE booting. Also why does it need PXE boot enabled if its not PXE booting from fog? Do you have multiple PXE servers on that subnet?

I setup dban (Derrick’s boot and nuke) in the PXE advanced menu in order to erase hdd’s.
[CODE]:dban
kernel ${boot_url}/dban/dban.bzi nuke=“dwipe” silent vga=785
boot[/CODE]

You can also configure it to autonuke, which will wipe a hard drive as soon as that item is selected from the advanced PXE menu. Its been useful.

It is typically client cpu bound. FOG_PIGZ_COMP is by defaut set to 9 on Fog 1.x.x whereas it was set to 3 on Fog 0.32. You can lower that setting and you should see faster upload times, but the image will be larger.

Check the config on the web interface in FOG Configuration > Web Server > FOG_WEB_HOST and see if that IP adress is correct.

Otherwise check through the config file in /var/www/fog/commons/config.php

Those are good places to start.

[quote=“need2, post: 30550, member: 21891”]A Gigabit network could transmit in the absolute best case scenario 125 gigaBYTES per second. [SIZE=2][/SIZE][/quote][SIZE=2] I think you meant to say 125 megaBYTES. So if everything was network limited and no compression, speeds would approach 7.5 Gigabytes per minute. [/SIZE]

[SIZE=2]1.2 gigabytes is reasonable for old clients (P4, IDE drives) on 0.32, I had a fully gigabit network a[/SIZE]s well[SIZE=2]. Fog in my experience is usually client CPU limited, because PIGZ is running client side. On FOG 1.1.1 with PIGZ_COMPRESSION set to 9, 6.9 gigabytes per minute on Haswell i5’s is standard for us (Dell Optiplex 3020’s). Going to experiment with the compression ratio to see if there is a speed up to be gained there. Hope that helps.[/SIZE]

[quote=“PomonaIT, post: 30489, member: 24658”]Looks like that worked. So in the future, I can use that command with the latest version number to get the official build? Another question; What directories are the server files stored? Am ok to delete any folder starting with Fog_xxx or x.x.x under the foguser home folder? I assume these are just installation folders?[/quote]

Yes, that will get you the official version. It depends on how you installed it, but I typically download fog to /usr/fog_NEW and I upgrade fog, then I delete the /usr/fog_OLD directory. So if you just downloaded fog to the fog user’s home directory, you should be able to just delete the old files.

So I think you are actually ahead of the tagged 1.1.1 revision. If you install from trunk, you get the latest commit. Download the stable branches using the following code:
[CODE]svn co https://svn.code.sf.net/p/freeghost/code/tags/1.1.1 /INSTALL_DIR[/CODE]

[quote=“Joseph Waldfogel, post: 28648, member: 23569”]I get “iptables: unrecognized service.”[/quote]

Try: [CODE]sudo service ufw stop[/CODE]

That’s ubuntu’s Universal Firewall Service

That node is the master storage node.

I updated storage node to 1.0.1, still having the exact same same issue. Any ideas?

Ok I will try that. I didn’t think it would be an issue because 0.32 is running and working on all of the other storage nodes in the other storage groups flawlessly. I will upgrade and report back. Thanks for the advice

When I try to deploy an image to a client on one of my storage groups, I get the following error in the web interface:

Download task failed to create for MyHost with image MyImage
FOGFTP: Failed to connect. Host: xxx.xxx.xxx.xxx. Error: Undefined index: optimalStorageNode

Host refers to the correct IP address for the storage node.

Any ideas on solutions to this problem or how to troubleshoot further?

Basic Info: Fog 1.0.1, storage node is still Fog 0.32. Centos 6.4 on both nodes

Awesome, that got me most of the way there. Now its just messing around with kernel arguments for the two Ubuntu distros. But, I did get Xubuntu 13.04 working with the following code in case anyone else finds it useful:

[CODE]:Xubuntu 13.04
kernel ${boot_url}/xubuntu/casper/vmlinuz root=/dev/nfs/ boot=casper netboot=nfs nfsroot=xxx.xxx.xxx.xxx:/images/xubuntu/
initrd http://xxx.xxx.xxx.xxx/fog/service/ipxe/xubuntu/casper/initrd.lz
boot
[/CODE]

Thanks for all the help!

Thanks, the imgargs line kills it. I’ve been trying various tutorials and combinations for that line but I have yet to find one that works.

Thanks, that got me further. Now it goes back to the first menu (same error) after initrd loads. I’m thinking it has to do with the arguments passed to Ubuntu, but I am unsure. Any ideas?

Hello, I am using Fog 1.0.1 on CentOS 6.5 and attempting to get Ubuntu 12.04 or 14.04 to live boot. Does anyone have a working iPXE menu configuration, or have any idea why mine is failing (screen flashes and it goes back to first iPXE menu). I had a working configuration for both using fog 0.32. My menu code is below:
[CODE]set boot_url http://${fog-ip}/var/www/html/fog/service/ipxe/
set boot_nfs ${fog-ip}:/images/

:MENU
menu
item Ubuntu12 Ubuntu 12.04 x64 Live
item Ubuntu14 Ubuntu 14.04 x64 Live
item return return to previous menu

:Ubuntu12
kernel http://xxx.xxx.xxx.xxx/var/www/html/fog/service/ipxe/ubuntu12/casper/vmlinuz.efi
initrd http://xxx.xxx.xxx.xxx/var/www/html/fog/service/ipxe/ubuntu12/casper/initrd.lz
imgargs boot=casper netboot=nfs nfsroot=xxx.xxx.xxx.xxx:/images/ubuntu12.04/
shell
boot

:Ubuntu14
kernel ${boot_url}/ubuntu/casper/vmlinuz.efi
initrd ${boot_url}/ubuntu/casper/initrd.lz
imgargs boot=casper netboot=nfs nfsroot=xxx.xxx.xxx.xxx:/images/ubuntu14.04/
boot

:return
:return
chain http://${boot_url}/boot.php?mac=${net0/mac} ||
prompt
goto MENU

autoboot[/CODE]