I’ve been having an issue getting machines to image. I’m getting the Checking Queue problem when I try to image and freezing when attempting the host registration. I’ve already been digging through the forums and cannot find a situation that matches mine. I’ve tried everything and cannot make this work.
First about the environment.
We are running fog .28 on Ubuntu server 09.04 (It’s been working for 2 years now)
Windows is providing DHCP with the proper configuration options set
We have one master node and 8 storage nodes connected over a WAN.
The master node is configured like a normal site with the exception of /tftpboot/pxelinux.cfg/ set as a NFS share.
The storage nodes follow the normal storage configuration except for the following:
[I]configureTFTPandPXE; Added to line 214 of installfog.sh[/I]
[I]“tftpd-hpa tftp-hpa” added to line 25 of config.sh[/I]
[I]/tftpboot/pxelinux.cfg/ set as a NFS share back to the same directory on the main server[/I]
[I]The rest of /tftpboot/ set to rsync back to the main server every hour[/I]
each site has it’s own storage node and storage group in the console as well as a set of images created for each of these.
When I try to image or add some machines in 2 of my sites I get the checking queue or the lock-up when adding host. Now what makes this weird. I can take another machine of the same type (Compaq d510 in most cases) and it will work just fine. To further complicate things, I can take one of the machines that DOES NOT work at that location and bring it to another location and have it work just fine. Even though they are pulling the same kernel and init.gz file. I’ve even gone as far as to try imaging a non-working machine on a known working port and vice versa to no avail.
Is there anywhere else I should be looking?