Posts made by scottmcmullan8

Ah ok I understand now - brilliant this has resolved the issue I wasn’t aware that specifying the default OU would have a knock on effect.

Thanks again for your help

I have previously had fog set up to join different OUs depending on the department the machine will be located. Separated by the | symbol and the default OU being set by the ;

It use to work a treat but I have since removed all this to try and fix the current issue. Still the same message in the log file and it is just not joining the domain. Not sure what it is failing on as it says HostnameChanger Domain Error! (‘Unknown Error’ Code: 2)

Hi,

I am having trouble with getting fog to add machines to the domain. This was working fine and has recently just stopped and I cannot figure out what has changed.

I get as far as the image being deployed and in the fog log it says 20/04/2015 09:51 FOG::HostnameChanger Domain Error! (‘Unknown Error’ Code: 2)

I am not sure what the error code 2 means.

I have tried changing he password and rebuilt the dll and used fog crypt. When I do this it brings up a different error message which I searched for and it states it is something to do with the password. So I am sure that error code 2 is not password related? I have also checked the hostname changer service is running and that the OU directory listed is correct.

Help?

Thanks chad-bisd for the reply.

How do we add a proxydhcp server on fog?

Also after the update to fog 1.1.2 which no longer supports pxelinux.0 we are now using undionly.kpxe or ipxe.pxe.

Thanks
Scott

Already tried a bios update and same issue.

Network card in the machine is as follows: Intel 82578DM Gigabit Network Connection

they are receiving no DHCP offers received or else get to the deploy image stage and freeze on a black screen with cursor flashing. It appears when the machines that aren’t working are trying to PXE boot they are using Intel Boot Agent Version 1.3.24 however the machines that are working and can deploy the image or get to PXE menu without any issues are using a Intel PXE boot agent?

These machines are all on same IP range, under same scope with same boot options applied.

Another update - it appears when the machines that aren’t working are trying to PXE boot they are using Intel Boot Agent Version 1.3.24 however the machines that are working are using a Intel PXE boot agent?

Also as a follow up on this.

Another machine that we have tested that is not working (black screen blinking cursor or reopening network devices or no DHCP offers received) is the Fujitsu ESPRIMO Q530E. So now we have the Fujitsu and Viglen machines not able to deploy image.

However we have Shuttle PCS and Dell OptiPlex machines that we are able to send an image to using same network port and same configurations.

Really not sure of any way around this after changing the boot file from ipxe.pxe, undionly.kpxe and the other options nothing seems to work.

Any one else experienced this?

Have changed the options from Ipxe.pxe and nothing being received at all for the Viglen however the Shuttle and Dell get to Fog menu stage and can image fine. Changed back to Undionly.kpxe and the Viglen then gets to the fog menu and when I select perform full host registration I then get the loading bzimage etc. before screen changes to black with cursor blinking on the screen and that’s as far as it gets.

Do you mean changing the options to boot from IPXE.PXE or UNDIONLY.KPXE etc.?

I have changed this and tried a few different ones (all that came with the install of fog) and same result unfortunately. Last working method was the pxelinux.0 for these machines. Are there some older machines which these files are not compatible?

Hi,

Upgraded from fog 0.3.x to fog 1.1.2 - so far any issues have been solvable. Everything has now been set up and configured as I would want in my environment. The final issue that is happening now is that certain machines are unable to PXE boot that previously had no issues. Testing in same lab environment with 3 different machines:
Dell OptiPlex, Viglen and a Shuttle.

The Dell and Shuttle machines PXE boot without issue. I can then either perform full host registration or manually add in the mac address in fog to get the image to deploy.

However the Viglen machines (which previously PXE booted on fog 0.3.x using pxelinux.0) no longer are working as should. I have received various different error messages. Sometimes it will get to the fog menu but then when trying to send the image it will display only a blinking cursor and not get any further just every couple of seconds will have another _ displayed and move further down the screen. Sometimes it will then come up saying no dhcp offers received or reopening network devices. Then I have also seen the error message displayed DHCP boot not for us. I am at a loss to why this is happening apart from the fact that maybe the network card or boot options are too far outdated for the new method being use for PXE?

Any help would or advice would be appreciated.

Thanks

Hi there,

This is a quick question as I’m sure one of you will know the answer to this. I had upgraded from fog 0.3.x to fog 1.1.2. Uploaded a new image using fog 1.1.2 so that the image could be pushed out through part clone to machines. However I have also storage nodes set up in different offices on different ip scopes. The storage nodes are still on fog 0.3.x and I haven’t got around to updating yet. I have noticed when pushing out the image it still works using part clone however there is no progress bar. Is this because the versions of fog are different ? as the progress bar works fine when using master node running 1.1.2?

Thanks
Scott

I have uploaded the new files and the issue has now been resolved fog is recognising the correct mac address. Thank you for taking the time to look at this.

Thank you - will give it a go. Appreciate your help and will let you know the results.

Are these commands just overwriting the files:

/var/www/fog/service/Pre_Stage1.php
/var/www/fog/service/Post_Stage2.php
/var/www/fog/service/Post_Stage3.php
/var/www/fog/service/Post_Wipe.php

If so can I download new files from:
[URL=‘https://svn.code.sf.net/p/freeghost/code/trunk/packages/web/service/Pre_Stage1.php’][U][COLOR=#0066cc]https://svn.code.sf.net/p/freeghost/code/trunk/packages/web/service/Pre_Stage1.php[/COLOR][/U][/URL]
[url]https://svn.code.sf.net/p/freeghost/code/trunk/packages/web/service/Post_Stage2.php[/url]
[url]https://svn.code.sf.net/p/freeghost/code/trunk/packages/web/service/Post_Stage3.php[/url]
[URL=‘https://svn.code.sf.net/p/freeghost/code/trunk/packages/web/service/Post_Wipe.php’][U][COLOR=#0066cc]https://svn.code.sf.net/p/freeghost/code/trunk/packages/web/service/Post_Wipe.php[/COLOR][/U][/URL]

and then upload to /var/www/fog/service/ myself?

Ok thanks - going to give this a go. Will the commands mean that only a single mac address will be added as the host as opposed to both mac addresses?

Also not 100% how to make the change to SVN - where can I find this to edit? Is it easy enough to complete?

Ah right ok I see - thanks that would be great appreciate it

Another update (Sorry lol). Have now managed to get the machines to send the image by disabling one of the NICs in the BIOS menu. This means that the only mac address being picked up in the logs is for the one which is enabled. However I have previously sent images to the same machines, all the same make and model, with two NICS with both being enabled in the bios. Is it possible that IPXE doesn’t like having two NICS enabled on a machine and since I have only been using this method this week is why I have been having problems?

No changes have been made to the boot.menu.class.php

Logs under apache2 show the correct mac address of the machine as below:

10.0.4.67 - - [09/Jul/2014:11:32:31 +0100] “GET /fog/service/Pre_Stage1.php?mac=80:ee:73:96:71:27|80:ee:73:96:71:26&type=down HTTP/1.1” 200 314 “-” “Wget”

Only thing I am noticing from above log is that it has two mac address ? It does have two NICS but only one is connected. Also checking the logs from the same make and model of machine with two NICS from yesterday it only shows one mac address

10.0.4.24 - - [08/Jul/2014:14:27:56 +0100] “GET /fog/service/Pre_Stage1.php?mac=80:ee:73:83:a0:46&type=down HTTP/1.1” 200 299 “-” “Wget”

Quick update also - whenever I add the host in and select the image and download it adds to the database. When I boot the machine it looks as if it is about to send the image and goes through the process. However it then gets so far and comes back with the error “No host found for this mac address” when trying to send to inventory. Even thought the host has been added through the fog console using that mac address.