Is there a way to have my machines boot order to have PXE first, and when they boot, it goes to PXE, there is no job for them so they then boot to hard drive? Tired of pressing F12 on 50 machines.
@george1421 The images are not on root partition. Here is our partition setup. This is in a fully STIG compliant secured centos 7.3
/root
/
/home
/boot
swap
/var/log
/var/log/audit
/tmp
Images are stored in /home
Long story short. We maintain 1 centos 7.3 image. Our home directory is around 1.5 TB of storage. We create the /home/images directory to store images. In the past, after the install I had to go into the settings and simply change the location of the storage location and dev, now maybe, if we need to rebuild, I can just specify the location during the install. Trying to simplify things as much as possible for those that come behind me to attempt.
I was running a fresh install today and I did not see where it asked for storage location.
Danke! Will give it a go. Not too big a deal as I can change the settings but would be nice to just specify during install.
Is there a way during the fog install to change the default /images location? I see the section in the installer.sh where is prompts you to put in the location but when I run the installer, I never see the prompt. I there a command line switch or something. We use /home/images as our centos 7.3 boxes are all built from the same image and we do not wish to maintain another baseline. Our partition where /images defaults to is not large enough.
V/R
I did load 1.3.0 RC23 and the swap issue is there a problem. The UUID is carried over from the imaged to the deployed machine. I must change fstab to put in the correct UUID.
Is the UUID issue fixed when imaging? In my fstab I use UUID instead of /dev names and once an image is captured and deployed to new system, the UUIDs in fstab are for the imaged machine not the deployed machine. SWAP will not come up and I have to run blkid, then copy in the new UUID into the fstab for swap to work. Just wondering if there is a fix for this. I have 100+ machines to image so for now I have changed to using /dev/sda format in the fstab and this should not be an issue as we do not change the setup at all.
This post is solved
@Tom-Elliott
Not sure I understand what you are saying.
Do I only need the /home/images in my /etc/exports?
I think I understand that the /home/images should be rw instead of ro?
Quick question:
The FOG server was able to upload (capture) the image successfully, however when I look in /home/images (my new image path), I see the mac address as the image name instead of centos72IT1103, the image name associated with my hosts? The old fog server created the same name folder in /home/images as the image name in “Image Management”.
V/R
Got it, was missing the .mntcheck and dev folder. Not sure why the dev did not get created but just copied over and maintained perms.
I am IMAGING!!!
A great learning process for me for sure. Thank you for all the help. Now the lab is up. Back to 1.3.0 on CENTOS 7.2. I think I got it this time around no prob.
V/R
/images *(ro,sync,nowdelay…fsid1)
/images/dev *(ro,sync,nowdelay…fsid2)
not seeing /home/images… does it need to be here, if so I believe it would be something like
/home/images/dev *(ro, sync…fsid3)
?
The default of /images works just fine when I image a machine. However the / partition only has 15 GB of space. The /home partition has over 500GB of space for images. I created a /home/images folder and made it the master node, the only node for holding images. I am getting a permission denied error.
NFS is mounting 172.16.61.17:/home/images/dev on /images failed: permissioned denied.
I checked the perms and they appear to be ok at 777 on the /home/images folder recursively.
Any ideas?
Just trying to change the default path from /images to /home/images
@george1421
Would love to have that script. I am not giving up on this one. Will be at this first thing in the morning. Gotta go home and feed the kids.
Let me know how I can get that script you mentioned. I did check the default.ipxe and it was good to go.
V/R
Thanks to all of you for your time. I know this is a rough one, at least for me.
So is reinstalled on a fresh centos 7.2 machine and I am getting tftp timeout issues. I installed using my corporate network of 10.x.x.x and I then went in attempted to change over to 172.x.x.x but there seems to be alot of places the IP gets stored. Think I will build a centos 6.8 and just use 1.2.0 as its all contained and doesnt need to talk back to the net. I can at least control the IPs etc, as I am not an expert in linux, php, apache, fog, etc.
I just need a server running asap, then I can go back and build the updated trunk as I do wish to have the new features, but right now we are down hard in our lab.
As of now, I am scraping the fix on this one. some of the files are owned fog.apache, some are root.root, permissions are wrong all over the place.
Going to attempt to put a clean fog install on a new centos 7 machine.
@Tom-Elliott said in /Default.ipxe.....Operation not supported:
isolated fog servers
Now I am here.
Mounting file system fail.
An error has been detected
could not mount images folder (/bin/fog.download)
args passed:
Reason: mount: mounting 172.16.61.17 ://home/images/ on /images failde: Permissions denied