@rogalskij From the log we see that the client keeps receiving Middleware::Response Invalid host. This means the MAC is somehow already known to FOG so it doesn’t wanna register this “new” machine through the client.
Probably best to take a look at the database directly. Take a look at the SQL queries in this topic: https://forums.fogproject.org/topic/12078/i-don-t-see-some-hosts
@3mu Thanks for the update. I will take a look at this when I have more time.
@3mu said in FOGSnapinReplicator.service failing on fresh install:
Install FOG with no router, DNS server = the FOG server IP (with no DNS server running)
Thanks for the details. Though I still need to clarify this point. Maybe I am able to figure out the root cause of this and even prevent from it in our setup.
So when you say DNS server, do you mean the FOG installer option/question that is asking for DNS server or just standard Linux DNS server setup though /etc/resolv.conf?
@icescout27 As I said, it used to be fog but now fogproject is used. You are not meant to play with this. As well you shall not use this account to login to your FOG server.
Please run ls -alR /media/fog/External ; mount and post output here.
After changing the username field to fogproject in both the .fogsettings file and Default Storage node I am still receiving the same error as specified in the original post.
That is impossible. At least you should see the username changed in the error output. Saying you get the same error message is not of much help because you might overlook something important. Please take another picture of the new error screen and post here.
@zyzhang @george1421 Nice we seem to have finally figured it out!
@Oleg60 Beside forcing an error like George said there are other ways to get to the shell. Either you edit /tftpboot/default.ipxe and make the first two lines read like this:
#!ipxe
shell
The rest of the original script can stay as is. If you exit the shell it will execute the rest of the commands. Be aware that all your PXE booting machines will run into the shell no matter what!
The other simple way is to create a new iPXE menu entry (web UI -> FOG Settings) and use the shell command with that. This way you have a new menu entry that you can choose to get to the iPXE shell on PXE boot up.
@lakk Well let’s pick this up now that you’re back from your holidays. Please run the database maintenance commands to see if that helps: https://wiki.fogproject.org/wiki/index.php/Troubleshoot_MySQL#Database_Maintenance_Commands
@george1421 @zyzhang I put in the download links for init and kernel in my last post.
@3mu Thanks for letting us know you solved this and DNS server was the issue. Can you please be more specific on how this conflicted and why?
Really sounds like other people could have run into this as well and would embrace your helpful information on this.
@abulhol Give it a go and post information here or on github to further discuss things. Or you might send in a pull request on the fos repo right away.
@icescout27 said in Image Capture Error: "Failed to open stream: Permission Denied":
When I elevate to root and run the command suggested I receive the following message
No. Looks like this is not a SElinux enabled server. Ubuntu uses another security framework called AppArmor but I don’t think this is causing the issue.
Run df -h as root and post output here.
Can you please tell us more about this FOG sever? Sounds like you haven’t set it up yourself but took it over from someone. Do you know when it properly captured an image last?
Just to check the basics: Is 192.168.1.1 the IP of your FOG server?
… FOG version 1.5.9 RC2
Ahhhhh, now I see. Something has gone wrong when updating from an earlier version. In the picture we see is state the username fog. This was used in earlier versions of FOG but not with the one you are using. Please login to the web UI, go to Storage settings, edit the “Default Storage” node and set the Management Username to fogproject. As well open a root command shell on your FOG server, edit file /opt/fog/.fogsettings and adjust the line starting with username=... accordingly.
@george1421 @zyzhang I would try using latest init and kernel. Hope switching to those is not causing other issues.
@Jeffson said in PXE-E78 some computers:
What I don’t really understand is why does it work for the same model of computer with the same BIOS settings etc and for some not…
Do those that work, reliably PXE boot every time? If you answer this with yes I am sure there must be a difference between those that work and those that don’t! Or it’s some specialty within your network setup (different subnets, VLANs or firewall settings).
@Hartman4820 said in fog capture stuck at starting cmain:
starting cmain
This message is only being displayed by GRUB chainloader as far as I know. What this means is that the host your are PXE booting is using GRUB to chainload the OS from your hard drive and therefore I am sure there is no task even scheduled for this host!
You are saying “I check the Tasks and it just stops the task.”. Can you please let us know how exactly you scheduled the task and if you get to see it in the tasks list at the very beginning?
Just started a new fog server.
Which version of FOG do you use?
@icescout27 What Linux OS is this? Any chance SELinux is/got enabled?
Run getenforce as root on the command shell and post output here.
@mkeane Using USB boot. Start looking into this here: https://forums.fogproject.org/topic/6350/usb-boot-uefi-client-into-fog-menu-easy-way
@george1421 @zyzhang This issue might be fixed in a later init version when we added the flock calls.
@mkeane FOG does use the MAC address to differentiate between clients. So if you use the same USB adapter for all your laptops you can’t have a separate inventory for those.
This topic has been discussed many times and we have worked towards a different solution but have not found a proper solution yet: https://github.com/FOGProject/fogproject/issues/198
@SDToone Why do you want to update to version 1.5.7 - not the latest?
Sure we can give you instructions but at first we better talk about where you wanna go.
Do you have a full backup of the server (at least database dump and contents of /images directory)??
@zyzhang Do you still get the exact same error on screen? Please take a new picture of the errors on screen and post here!