@Joe-Gill said in PXE Boot Problems:

I can boot the same machine type from another segment of our network fine as well.

For the first picture: So why should it be an iPXE issue then? This sounds really weird. Is this issue reproducible?

For the second picture: I suppose this model does not properly PXE boot in a different network segment, right? As George said, try different iPXE binaries (find them /tftpboot on your server) to see if one of them works…

@quinniedid There is a .msi version of the client that you could use to install the newer client with the correct address. Unsure if old client would need uninstalled first.

If you have a way to push out a PowerShell script, something like this might help.

$FOG = Get-WmiObject -Class Win32_Product | Where-Object -Property Name -EQ 'FOG Service' $FOG.Uninstall() $url = "http://<ip address>/fog/client/download.php?newclient" $outfile = "FOGService.msi" (New-Object System.Net.WebClient).DownloadFile($url, $outfile) $ArgumentList = "/i $outfile /quiet USETRAY=""0"" WEBADDRESS=""<ip address>"" WEBROOT=""/fog"" /norestart" Start-Process -FilePath MSIEXEC.exe -ArgumentList $ArgumentList -Wait Start-Service -Name FOGService Start-Sleep -Seconds 2 Restart-Computer -Force

Of course first you need to create the group itself.

From a php script this could be done with:

<?php require('/var/www/fog/commons/base.inc.php'); FOGCore::getClass( 'Group' )->set('name', 'SomeNameToNameThisGroup') ->addHost(array(1,2,3,4,5,6)) ->save();

Of course change the array(1,2,3,4,5,6) with the list of ID’s you need to associate to the group.

It sounds like the drives are on separate virtual machines in which case you need to install fog on the second one as a storage node. Not sure how current this is but here is a video about it https://youtu.be/X72WthDGwsw.

@Tom-Elliott @george1421
Ah, so simple! Thanks very much gents!

@Ben-Warfield

Thank you Ben, just updating, simply comment de bind address on my.cnf solves the issue.

Hey george1421,
Thanks for clearing that up mate

@zingaro No, we’ve been putting together documentation where we feel it’s needed, and we consider including any submitted documentation. Is there something in particular you want to know about? Links to the wiki are in my signature.

sorry

network team had changed the scope !!!
there was no network connection

Rename .fogsettings to .fogsettingsold, in /opt/fog

@callcenter good, thanks!

Restarted FOGMulticastManager and it is working properly now

@Wayne-Workman : I will try as per your suggestion and update it

@Tom-Elliott Thanks Tom. I went ahead and deleted all the columns and just left the <mac> and <hostname> and they imported successfully.

@Quazz

Update to 1.4.0 seems to have fixed the issue with imaging.

So found the issue. Individual taskings would work, but group taskings wouldn’t. This is due to some checks added to ensure only valid taskings can be generated. The tests were looking for “Imaging tasks”, then “Snapin Tasks”, then “init needed taskings”. The memtest tasking is not considered an init needed tasking.

I’ve made the checks simpler with:

“Imaging tasks” -> “Snapin Tasks” -> All other and tasks appear to be working properly from the group page. Thanks for informing.

@Tom-Elliott

Well it does appear that it finished… I just didn’t see the play by play like the full wipe… SO that’s my fault!!

Feel free to close this ticket! Thanks!

I think Quazz is right. I was trying “sudo dnsmasq” to troubleshoot. Sorry - am very new to Linux. I think this issue is resolved, though PXE is not working, but I’ll ask about that in a new thread. For anyone reading about this issue, the steps in here will help set up dnsmasq, and answered my question.

@Thiago I think that’s what I needed to know.

If you need the directory listing, you might be better creating an index.php file that will generate the list of files for you.

You can even make them links so you could test downloads.

The idea is we use an API now and if the index file doesn’t exist (which it’s attempting to locate) it will attempt to redirect through the API. I just needed clarity on whether there was an actual issue of which you just let us know there wasn’t. It was just your “testing” that was failing (which is helpful, but not requried).

@safari2010 On screen where it’s giving the long warning, please say yes.