Posts made by chad-bisd

I have successfully run Fog 0.32 on Ubuntu Server 10.04 LTS, both a test server for 4 months, and now a production server for the last 9 months or so. I tried newer versions of Ubuntu server, but once I install gnome-desktop-environment, they all seem to have problems with my hardware. So for now, I just stick with what works.

Yes. The definition of the images is separate from the storage of the images. You can have one without the other, which is the situation such as when you create a new image in the web ui and before you upload the actual image. Conversely, if you delete the definition from the database, the file(s) can still exist if they are not deleted also. This means you can copy or move the image files to an external HD if they are no longer in use, but still leave the definitions in FOG. If you ever need to use the image again, just copy it back to the images folder.

To recover your “lost” image:

Create a new image in the Web UI by giving the new image definition the same type (single partition, mult-partition, etc) and specifying the correct storage location.

Example:
Image Name: Image1
Image Description: First Image, Win XP Pro 32 bit
Storage Group: Default
Image file: /images/[Name of file or folder that holds the image or image files]
Image type: [select correct image type]

If the image is stored in 1 file under /images/, such as a single parition-resizable image type, then put the filename as the value for the Image file: /images/[image file]

If the image is stored in a folder under /images/, such as a multiple-partition-single disk or multiple disk, then put the folder name that contains the image files.

They are telling you to update the TFTP uesrname and password in the fog web UI settings to match

Try reading through this. It sounds long and complicated, but I just put every detail I could into it. You won’t have the same selections, but it will lead you to the right places to make your selections for your hardware.

[url]http://fogproject.org/forum/threads/compiling-a-custom-kernel-for-acer-iconia-tab-copied-from-old-forums.216/[/url]

[LEFT]No, sorry. been super busy with state mandated testing and getting student laptops ready. It’s still on my to-do list though.[/LEFT]

This topic should not be in developer help, but I’ll give you some advice anyways.

You want to run your server with ProxyDHCP enabled. This detects PXE boot requests on the network, allows the computers to get an IP address from the real DHCP server but get PXE boot details from your fog server.

I have heavily modified the articles on the wiki with my experience using Ubuntu 10.04 LTS and Fog 0.32. The article title is
[SIZE=6][B][URL=‘http://www.fogproject.org/wiki/index.php?title=Using_FOG_with_an_unmodifiable_DHCP_server/_Using_FOG_with_no_DHCP_server’][SIZE=4]Using FOG with an unmodifiable DHCP server/ Using FOG with no DHCP server[/SIZE][/URL][/B][/SIZE]

[url]http://www.fogproject.org/wiki/index.php?title=Support_FOG[/url]

very bottom.

Our print server is Windows Server 2008 R2. Some drivers are Kyocera, some are Brother.

I will start experimenting with FOG printers in Windows 7 Pro 32 bit using FOG 0.32.

My plan is to add printers to FOG server.
[LIST=1]
[]Record status of interactive services detection service
[]Assign printers to hosts
[]Verify printers on the host
[]Remove printers from host
[]Add same printers back to host
[]Report my findings.
[/LIST]

Use code tags when posting log files or anything you don’t want to be interpreted by the forum parser.

please use the CODE tags.

[quote=“Kevin, post: 1645, member: 3”]So I have to ask, if you’ve never used sysprep, how do you image the computers with a standard image (software/updates etc) without the image having the same name for everything?[/quote]

I let the fog client rename the machines after imaging and then join the clients to the domain after they’ve been renamed.

There are two ways: complete, and partial.

To do a complete restore back to the point you took the database backup, you simply:
[CODE]user@computer:~$ mysql -u root fog < [backupfile].sql[/CODE] where [backupfile].sql is the name of your .sql file that was made by mysqldump. It should have all the commands in it to restore all tables and records in the FOG database.

To do a partial restore, say of just the images and hosts tables, make a copy of the .sql file for backup purposes. Then you’ll need to edit the .sql file and remove all the statements that change the other tables.

Leave the statements at the top that set the parameters and all statements related to the hosts and images tables.

Usually there is a statement block to drop and rebuild the table structure, then a block below it that inserts all the data back into the table, so make sure you delete both blocks for all tables you do NOT want to restore, and leave both blocks for all tables you DO want to restore.

Once you have the file like you want, save it, and run it with the same way you would for the complete database backup, just with the modified file.

You can safely ignore 95% of the messages when fog is loading up on a host. The default fog kernel tries to load just about every driver and fail safe workaround for hundreds (if not thousands) of devices. the AcerHDF, tps6510, etc messages are all just letting you know that you don’t have the hardware that the driver is for, but it tried anyways.

If all that stuff bugs you and you know you’ll never use that device, you can custom compile a kernel and remove most of the non-essential drivers it tries to load to work on the widest possible range of devices.

Was VM on esxi in an intel blade server setup but the FOG server was running into resource issues when imaging multiple computers. Started out great, but would stall at 200 MiB/min when imaging lately.

Now it’s on a Dell PowerEdge 2600 with 2 x 2.4 Ghz Dual Core Xeons, 8GB RAM, and 6x73GB 10k drives in raid 5. Ubuntu 10.04 LTS, with LTSP enhancements for ProxyDHCP functionality.

I also moved the server from the main switch which also had the rest of the servers and stuff, to a decidated switch in my workroom. Now I can image laptops and desktops that are in for service without impacting the rest of the network. 3.2 GiB/min with the new setup.

We’re a school, so the workload is usually 1 to 10 computers images per week, except in July/August when we image every laptop and desktop in the district, which is now up to 800 computers.

I think I understand what you mean.

Load up the old machine the way you want the first image. Register it with the FOG server. Go into the FOG webui, create the image definition, go to the hosts and associate the host record with the image record. Perform and upload task.

Create a new image record for your second image. Setup the computer the way you want for the second image. Go into FOG and set the upload task so it creates the second image.

Create a new image definition in fog, change the host record to associate to the new image, setup the computer the way you want, upload the image.

One computer, multiple images.

I use phpmyadmin as my web based mysql admin tool. Be careful of installing any web accessible admin tool if you did not set a root password for your mysql instance, as it leaves you open to “curious” people who find the website.

[B]Kill all tasks that are still in the queue. You do not want to do this procedure with active or scheduled tasks.[/B]

[B]Backup your database in case you mess it up.[/B]
from a terminal:
[CODE]user@machine>mysqldump:~$ mysqldump --user=root fog > fog.date +%Y%m%d.sql[/CODE]

note the use of [B]backticks[/B], not apostrophes or single quotations. If you don’t have mysqldump, get it using the package manager of your choice. On Ubuntu:[CODE]user@machine:~$ sudo apt-get install mysqldump[/CODE]

[B]Let’s get started[/B]
So I’m assuming you are deleting the image definitions for old images you no longer use and you want to consolidate the existing images into the 1 to X id range.

I’ll show you how to do it from the command line, because even through the webui, you still have to write the SQL code out OR do each update individually.

from a terminal:

[CODE]user@machine:~$ mysql -u root

mysql>use fog;

mysql>select imageid, imagename from images;[/CODE]

At this point, you should get a listing of all the image definitions showing the ID and the Name. I’m assuming you can tell the images apart based on the name. Make note of the existing image ID’s and names, we’ll need them later. I use putty, so a simple copy and paste to Notepad on my workstation is easy, but if you just have a few, write them down on paper.

[B]Note:[/B] If you deleted the old image definitions for outdated images, you either need to delete the host records associated with those entries if the hosts are no longer valid, or change their image value to a zero(0) value so they are not associated with the wrong image after you update the image ids. Just a bit of house-keeping so your FOG data stays sane. If you want to keep your database clean, skip to the end and run that script before you proceed here.

So now we have a list of the “good” images and their ID’s, it’s time to change them. Let’s say our existing data looked like:
[CODE]±--------±--------------------+
| imageID | imageName |
±--------±--------------------+
| 8 | StudentSpare-HL91 |
| 4 | BSELAB |
| 11 | MS16372Teacher2 |
| 10 | MS163KTeacher2 |
±--------±--------------------+[/CODE]

You want to change this to be:
[CODE]±--------±--------------------+
| imageID | imageName |
±--------±--------------------+
| 1 | StudentSpare-HL91 |
| 2 | BSELAB |
| 3 | MS16372Teacher2 |
| 4 | MS163KTeacher2 |
±--------±--------------------+[/CODE]

You’ll execute the following lines at the mysql> prompt.
[CODE]mysql>update images set imageID=1 where imageID=8 limit 1;
mysql>update images set imageID=2 where imageID=4 limit 1;
mysql>update images set imageID=3 where imageID=11 limit 1;
mysql>update images set imageID=4 where imageID=10 limit 1;[/CODE]

This changes the imageID value for each image definition. You can reorder them however you want, just adjust the imageID values in each statement. the “limit 1” on the end of each statement makes sure you change just 1 record, because logically there can be only 1.

What this has done is broken the link between the host records and the image records. At this point, the hosts no longer have a valid image associated with them, because they know the image by it’s previous ID, which is no longer valid.

Now you have to update the hosts table so that any hosts which used the old image ID, now uses it’s new image ID.
[CODE]mysql>update hosts set hostImage=1 where hostImage=8;
mysql>update hosts set hostImage=2 where hostImage=4;
mysql>update hosts set hostImage=3 where hostImage=11;
mysql>update hosts set hostImage=4 where hostImage=10;[/CODE]

The host records should now know the new image id for each image. If you’ve got host records that still reference images that are no longer valid, they may associate with the new ID’s you’ve given the images. You can clean them out individually through the webUI, or just leave them if you don’t care about your database being clean and sane.

[B]Keeping your database clean and sane by fixing host records that point to non-existent images[/B]
If you have removed old image definitions from FOG using the webUI, can you run the following statement from the mysql> prompt to disassocated any hosts records from those image ID’s that no longer exist. You want to do this before you change the “good” image id’s.

[CODE] update hosts set hostImage=0 where hostImage NOT IN (Select imageID from images);[/CODE]

Alternatively, you can delete all hosts from FOG which no longer have a valid image associated with them. Again, run this before you change the image ID’s. This is a bit more risky as there is no way to recover if you mistype or decide you want something back later.

[CODE] delete from hosts where hostImage NOT IN (Select imageID from images);[/CODE]

[quote=“Darren George, post: 2249, member: 375”]Good day all,

Ive been running fog for about 6 months now and think the system is absolutly brilliant, One thing im having an issue with is adding more storage for the images folder, ive tried a set of instructions from an old post but i keep getting mount issues, really all im looking to do is expand the image folders, Any help would be great, also any news on 0.33 looking forward to testing the new system out?[/quote]

[url]http://www.fogproject.org/wiki/index.php?title=Adding_Storage_to_a_FOG_Server[/url]

If the disk is already installed, partitioned, and formated, just follow the steps starting with renaming the existing /images folder.

FOG’s main job is to image computers quickly when doing mass deployments. It can do full system backups and restores for important computers if you have the storage space. It has a built-in memtest, and is extensible enough to allow you to PXE boot into other applications for disk checks and boot helpers. You have to be able to PXE boot though, very important as the majority of consumer grade desktops don’t PXE boot or don’t do it easily.

Clonezilla, Ghost, and others have boot CD’s and can backup/restore images from an external HD.