Latest FOG 0.33b

[Tom Elliott]

It isn’t a problem at all.

[fabritrento]

[quote=“Tom Elliott, post: 22140, member: 7271”][php] if ($ftp->rename($dest,$src)||$ftp->put($dest,$src))
($_REQUEST[‘osid’] == ‘1’ ? $ftp->delete($StorageNode->get(‘path’).‘/dev/’.$macftp) : null);[/php]

And these functions are in /var/www/fog/lib/fog/FOGFTP.class.php[/quote]

in fog script, /images is mounted on the client via nfs. so i don’t understand why is needed create pgfile (what is?) and use ftp.

is not more simple to:

1. mount the /images as nfs
2. partclone to /images/dev/mac/d1p1,d1p2…
3. simply do a mv /images/dev/mac /images/imagename

?? why to use ftp?

[fabritrento]

ok i removed /images/labinfociro SOLVED. i don’t understand the reason… this directory was chmodded to 777 so i don’t understand…

[Fernando Gietz]

FOG uses FTP for different reasons (i think

a) is a PHP script who moves the files from /images/dev/MAC/. to /images/imagename. A PHP script can run shell commands, but is not very safe. FOG uses VSFTP, is more safe.
b) is the FOG master server who runs mv commands, and if you have storage node, you need a FTP command.

[fabritrento]

now when i clich on the link to see all machines, ( [url]http://192.168.0.3/fog/management/index.php?node=host&sub=list[/url] ) i get this error
FOG DEBUG: Image: Database Load Failed: ID: 0, Error: Row not found
FOG DEBUG: Image: Database Load Failed: ID: 0, Error: Row not found

[Tom Elliott]

Make sure the system has the image id set.

[Fernando Gietz]

Tom, in the 0.33 version, is the server who makes the compression of the image files, no?

Is not better compress on the client, you can use the client cores, and upload a compressed image ?

[Tom Elliott]

Multicast decompression happens on the server, otherwise, the clients, during upload, compress the image.

[fabritrento]

i have a problem with multicast group deploy:

i addes 25 pc to a multicast deploy but:

-the first powered on started the image restore without waiting for other 24 with a very poor speed of 330mbyte/minute instead of 1000mbyte/minute when restore only one pc.
-all others has a gray window of partclone that waits…

i expect that all 25 pc waits for udp broadcast data to start restore, and start restore all togheter receiving the unique multicast udp connection, and not that the first one starts withlout waiting all others… in 0.32 this was worked.

now i go to home because 300 mbyte/minute means hours of deploy… tomorrow i pray the good that all pc works and are added to domain!

bye and thanks, good night

[ArchFan]

Tom, there’s a small typo in the install script. At the first step when selecting the distro type, it’s spelled “installtion”. Doesn’t cause any issues, but thought I’d pass it on

[Tom Elliott]

I’ll fix that, easy enough to do.

[Tom Elliott]

Already corrected now.

[Tom Elliott]

r1162 released. Should resize SDR disks after imaging completes. It wasn’t before and I’m sorry I missed that part.

Spelling issues have been addressed as well.

[rado]

Moved to: [URL=‘http://fogproject.org/forum/threads/support-for-multiple-nics-subnets.675/#post-22319’][URL=“http://fogproject.org/forum/threads/support-for-multiple-nics-subnets.675/#post-22319”]Support for multiple nics / subnets[/URL][/URL]

[fabritrento]

[quote=“Tom Elliott, post: 22173, member: 7271”]r1162 released. Should resize SDR disks after imaging completes. It wasn’t before and I’m sorry I missed that part.

Spelling issues have been addressed as well.[/quote]

with this release new pc inventory (quick inventory from grub fog menu) dont’ work.
pc says: No Host found for mac address xx:xx:xx:xx:xx (is normal… is a new pc!!) 2 times, one at start of inventory and another at end of inventory

no errors in error.log pf apache.

here access.log:
172.24.77.119 - - [30/Jan/2014:12:43:15 +0100] “POST /fog/service/auto.register.php HTTP/1.1” 200 318 “-” “Wget”
172.24.77.119 - - [30/Jan/2014:12:43:23 +0100] “POST /fog/service/inventory.php HTTP/1.1” 200 318 “-” “Wget”

i cannot find another errors…

[Albatros]

Hey Tom,

Updated to new revision and tested windows xp upload/download (single partition). After I downloaded the image to the client and started the os I get a blue screen something with unmountable boot volume.
The upload with the newest revision works fine but something goes wrong with the download. I switched to a older revision and the download and starting the os works fine, looks like that the error occurs with revision 1142 and newer.

Good luck by finding and fixing the problem,
Lukas

[Tom Elliott]

Albatros,

You updated to r1162 and are still getting the same issue?

I forgot to add a resize for single disk resize, and corrected that.

You state it only happens on deploy after 1142? I didn’t change the source in a way to modify windows, just the methods it uses for the files. Basically, I just separated the Upload/Download/Checkin Bits so they’re on their own files for easier troubleshooting/debugging.

[jbsclm]

I’m seeing the same thing on r1162, namely xp single partition resizable, failing on the reboot after deployment with unmountable boot volume.
A quick check with fdisk in debug mode suggests that it still isn’t resizing after deployment, as sda1 is much smaller than before the upload and deploy.
xp multi seems to work OK.

[Tom Elliott]

Alright, that leads me down a good path to troubleshoot.

[Tom Elliott]

Can you guys try with 1163.

I may have made a change without realizing, but I’ve tested resize for this and all seems to work fine. Then again, I could be wrong.