I wonder if it’s cause I have / instead of . If that’s it, I’m gonna feel silly.

So I have a Fog .32 server that the hosts won’t join AD. Name changer is enabled, AD settings are correct with the encrypted password, set to join after imaging task, etc. I’m getting the host name changed, but not getting joined to AD.

EDIT: Title meant to be Host won’t join AD not Fog, durp.

I’d ass in my.cnf --service-startup-timeout=900 to see if that helps you out.

Sounds like MySQL isn’t waiting to connect and just fails on startup. Ubuntu does weird things in the sake of “speeding up startup”.

Check in your my.cnf for MySQL what “[FONT=Helvetica][COLOR=#000000][SIZE=14px]–service-startup-timeout” is set to.[/SIZE][/COLOR][/FONT]

I’d think deleting the old server (Except for the images folder of course) would allow you to run the install and set it up as a storage node.

[url]http://www.fogproject.org/wiki/index.php/Uninstall_FOG[/url]

Just do not do “[FONT=monospace][COLOR=#000000]sudo rm -rf /images” [/COLOR][/FONT]

The editable text field is over in Fog Configuration > Fog Settings > Active Directory Defaults. You can put the default OU there and it’ll be selectable in the dropdown.

[quote=“Jim Holcomb, post: 31679, member: 15162”]Anyone??? I upgraded to v1.1.2 and still the same issue appears (TFTP invalid password)[/quote]

Did you try/verify anything I said at all? It was a direct response and everything.

If you update does it go away just for that time while it’s up then require an update again after a reboot?

2. What you can do is get into the Database “fog” and the table “users” and inject a line to create a user.
3. Once the user is created, you can login and add/edit/delete users.

SSH into the server
mysql -u root -p
Enter password
connect fog;
INSERT INTO fog.users (uId, uName, uPass, uCreateDate, uCreateBy, uType) VALUES (NULL, ‘newuser’, ‘5f4dcc3b5aa765d61d8327deb882cf99’, CURRENT_TIMESTAMP, ‘’, ‘’);
exit

If you do those steps and past that INSERT INTO line exactly, it will create a user called “newuser” with the password “password”.

Does disk space not report then for either storage node?

Sounds like the same problem as this thread, give those helpful steps a try and see if that helps.

[url]http://fogproject.org/forum/threads/imaging-stuck-on-upload-after-finished.10775/[/url]

When you did the registration, did you indicate to tell it to join AD or not? I see you’re still using 1.1.1 which that was the problem for me at a point.

To get the old images to work, you’ll have to under Fog Configuration > Fog Settings > General Settings you’ll need to enable “[FONT=Ubuntu][COLOR=#555555]FOG_LEGACY_FLAG_IN_GUI”.[/COLOR][/FONT] Now when you create an image, after you initially save it there’s a drop down to choose it to be PartImage or PartClone. Any new image will stay PartClone, but any old image will need to be told it’s PartImage.

You’re not supposed to use pxelinux.0 anymore, you’re supposed to use [B][COLOR=#000000][FONT=sans-serif]undionly.kpxe[/FONT][/COLOR][/B][COLOR=#000000]. You’ll need to update that in your DHCP option settings.[/COLOR]

Here’s my output

[CODE]-bash-4.1# free -m
total used free shared buffers cached
Mem: 7743 7616 126 0 18 7160
-/+ buffers/cache: 437 7305
Swap: 7887 0 7887
-bash-4.1#
[/CODE]

I only have an up time of 1 hour 39 minutes, but I only have 121 free also in free -m for comparison, 7621 used.

I’m betting you probably have a separate server that does the DHCP boot/options. You’d have to find out what that is and make the required changes.

See here: [url]http://www.fogproject.org/wiki/index.php/Modifying_existing_DHCP_server_to_work_with_FOG[/url]

For some reason my Fog server (1.1.1) is just chewing through RAM. Currently it’s CentOS 6.5, and was only leaving with 120M free out of 8GB after being online for a while and receiving images and deploying only one. I went ahead and removed GDM so it’s headless, and it manages to creep up on the RAM still it seems.

Short of the occasional reboot, not sure what to keep an eye on. Top (Sorted by memory) is useless and tells me it’s mostly all free anyway. I used [SIZE=14px][COLOR=#000000][FONT=Consolas]ps -e -o pid,vsz,comm= | sort -n -k 2 [/FONT][/COLOR][/SIZE] to figure out it was GDM as the big culprit, now I’m trying to narrow down the rest.

Have you changed your DHCP boot option to point to the new boot file?

It should be [B][FONT=sans-serif][COLOR=#000000]undionly.kpxe [/COLOR][/FONT][/B]instead of [B][FONT=sans-serif][COLOR=#000000]pxelinux.0[/COLOR][/FONT][/B].