Centos Help
-
@george1421 Yes
-
@mwright will you post a picture of where it freezes? The context of the freeze is important.
Also what hardware (make and mfg) are you trying to pxe boot?
-
@george1421 I am using a Dell T3400 and the machine I am trying to register the host of is an HP Compaq Pro 4300 AiO PC. As for a picture of the frozen screen, It looks like but all the information like client ID reflects the correct information. This is just a picture that I found on the internet as best representation. The computer will never leave this screen though and FOG won’t ever see it.
-
@mwright OK great this tells us that a good portion of your setup IS working. What is hanging at the moment is the iPXE kernel, it never finishes initializing. This does happen on occasion for certain hardware. This screen shot also tells us that you are pxe booting a bios/legacy mode device.
On that compaq pro 4300, can you ensure the firmware is up to date. iPXE rarely hangs for bios mode systems.
-
@george1421 This was a test I set up that was previously decommissioned, so it is possible that the firmware is not up to date. I will have to check on it. When you say firmware do you mean is the bios up to date?
-
@mwright Yes bios. With a uefi based computer, calling it bios seems a bit wrong since its a different type if firmware. Grab another computer and see if you can pxe boot it. You don’t have to image it, just see if it pxe boots. Your FOG server is working so far.
-
@george1421 I will check on that when I get back to work, yeah it is wrong to call it that. Feels like it cheapens it.
-
@george1421 The firmware is up to date. So it is not that, and we normally use Clonezilla and it was able to find the issue, I think I did something wrong and I do not know what.
-
@george1421 You were right, it is something with the computer. I tried a windows 10 machine and it worked instantly.
-
@mwright There are different ipxe kernels you can try to see if they work better. You are using the isc-dhcp server for boot information. In that file (first picture you provided) there is arch 0000 line it lists undionly.kkpxe as the boot file. You could test by replacing undionly.kkpxe with ipxe.kpxe then reboot the fog server. See if that boot kernel works better for that hardware.
The undionly.kpxe uses the undi driver to speak to the network adapter, where it uses bits of the built in network driver, where the ipxe.kpxe boot kernel has all of its own built in network drivers. The undionly.kpxe has wider support for more hardware because it uses parts of the nic’s built in driver, where ipxe.kpxe has its own drivers.
If the ipxe.kpxe driver doesn’t work for this hardware, then switch back to the undionly.kpxe. If it does work then we need to see how we can make isc-dhcp to dynamically adjust according to the boot device.
-
We have gotten further than you have, with using the undionly file, however, our image has 3 partitions, and only the 1st one will write. After it finished, it says cloned successfully and stops. We are also using CentOS with Fog version 1.4.4
-
@hvaransky Are you trying to image a HP Compaq Pro 4300 too?
-
@george1421 I will make the change that you suggested, but I ran it through setup again, on the windows 10 and I got a different error, It seems like fog is not pointing at its new IP address, I ran the installer but it doesn’t seem to want to change.
-
@mwright OK this is moving in the right direction. Do I understand that 65.50.x.x is the wrong address for the fog server?
If so you need to update it in 4 spots.
- In /opt/fog/.fogsettings file. That needs to have the current IP address of the fog server.
- In /tftpboot/default.ipxe
- In the webui FOG Settings->FOG Configuration, Just hit the expand all and search for “65.50.”
- Storage node Master node default. The IP address is in there too.
Once you get all of those places cleaned up then try to pxe boot again. You are sooooooo close.
-
@george1421 Where is the storage node master node default?
-
@mwright Storage Management. If you only have one fog server, then there should only be one storage node configuration in there called “default”
-
@george1421 Thank you so much for the help! You are a life saver!
-
@george1421 We were trying an HP 8200 Elite, however, we also tried a 4300 and got the same result.
-
@hvaransky said in Centos Help:
got the same result
Sorry I read too many threads here, same results as what? The OP had several issues here.
-
@hvaransky I also recommend you open a new thread unless your fog server and configurations are exactly the same as the OP’s I don’t want to confuse posted information because we are talking about different setups.
