Download stops with blinking underline after reboot.
-
Hopefully you can help me out. I have several Dell R410 servers that I am testing with FOG as a replacement for System Imager. One of the servers is loaded with CentOS 6 and another with CentOS 5. FOG has be installed and configured without any issues (that I know of) on an admin box. Both servers have uploaded images without problem. I have also restored the same systems with the images I took from them and they work fine. When I try to use the CentOS 6 image on the previous CentOS 5 unit, (and others like it), FOG goes through all the motions populating all the partitions, running a post install script I wrote to set the network configuration, reboots to pxe again, chooses “Boot from harddisk” and then hangs right after that with a blinking underline.
I have tried several things so far:- changing the BIOS to use SATA and AHCI - does the same thing for both
- tried to use debug and run fixparts /dev/sda - dumps me into a mbr utility prompt.
- set the FOG PXE boot Exit to Hard Drive Type to Exit and Grub - same deal.
I like the way FOG works and really want to use it but I must be missing something. Ideas?
Thank you. -
You’re on the right track, what you’ve tried is what I would have suggested doing, good job in that area.
So, why is it not exiting to the HDD properly…
Will the server boot normally if you don’t PXE boot?
-
@Wayne-Workman Tried to boot directly from the hard drive and same thing. Let me boot with a live linux version and look at what the disk configuration looks like. In the meantime, I thought I saw where someone else was having issues with Dell equipment? Most of the FOG imaging seems to be Windows oriented so maybe not too many folks going down my road?
-
@bmyers It’s geared towards windows, but FOG images Linux great. Your issue is likely hardware related.
What filesystem type is these OSs setup with? Did you use a resizable image or not?
Also - you’d likely have much better luck with FOG Trunk if you’re not already using it.
-
@bmyers Live booted Linuxmint and looked at disk partitions with fdisk and parted. The output looked consistent with a good system. I examined the mbr with dd for 512 bytes and it looked reasonable compared to another system. Any other suggestions?
-
@Wayne-Workman Sorry didn’t see your post… The file system is ext4. I did use resizable image since we have had some systems replaced with a different size hard drive in the past. Let me look into FOG Trunk and see what that is about.
-
@bmyers Definitely give FOG trunk a try (see in the wiki on how to install). If you still see the issue could you please create a new image definition - non-resizable and try deploying this. In case this does not help we need more information. Please check the image directory on the server
ls -al /images/<imagename>and post the contents of all the small text files you have in there. As well uploadd1.mbr. -
@Sebastian-Roth I was in the middle of upgrading to trunk and of course… I have been told to back burner this for another project (problem). I’ll get back to this next week. Thanks for everyone’s help so far.
-
@Sebastian-Roth I’m Back! Upgrade to FOG trunk 8020. Created new image with non-resizeable option. Deployed and get the same thing. (blinking underline cursor). Changed host Bios Exit Type to Grub, pxe boot, imaged, reboot, pxe boot again, boot from hard drive…then hung up as before but on the screen says: Begin PXE Scan Starting Cmain() (blinking underline cursor). It seems like it’s not even getting to the MBR??
[root@linuxbuild1 CentOS6htpvncNRHD]# pwd /home/images/CentOS6htpvncNRHD [root@linuxbuild1 CentOS6htpvncNRHD]# ls -al total 2173608 drwxrwxrwx 2 root root 4096 Jun 15 15:52 . drwxrwxrwx 8 root root 4096 Jun 15 15:52 .. -rwxrwxrwx 1 root root 0 Jun 15 15:49 d1.has_grub -rwxrwxrwx 1 root root 1048576 Jun 15 15:49 d1.mbr -rwxrwxrwx 1 root root 47 Jun 15 15:51 d1.original.swapuuids -rwxrwxrwx 1 root root 52941424 Jun 15 15:49 d1p1.img -rwxrwxrwx 1 root root 2106038315 Jun 15 15:51 d1p2.img -rwxrwxrwx 1 root root 512 Jun 15 15:51 d1p4.ebr -rwxrwxrwx 1 root root 512 Jun 15 15:49 d1p5.ebr -rwxrwxrwx 1 root root 65710547 Jun 15 15:52 d1p5.img -rwxrwxrwx 1 root root 368 Jun 15 15:49 d1.partitionsd1.has_grub (empty)
cat d1.original.swapuuids /dev/sda3 7a7a7ae8-9b81-49bd-8c3a-ca447dbb6c60cat d1.partitions label: dos label-id: 0x000c1095 device: /dev/sda unit: sectors /dev/sda1 : start= 2048, size= 2048000, type=83, bootable /dev/sda2 : start= 2050048, size= 102400000, type=83 /dev/sda3 : start= 104450048, size= 102400000, type=82 /dev/sda4 : start= 206850048, size= 281430016, type=5 /dev/sda5 : start= 206852096, size= 281427968, type=83 -
@bmyers I just tried to load the same image I took from the same machine (CentOS 5) and it went through the whole imaging process then on reboot…dhcp…boot from disk…
Chainloading failed, Hit ‘s’ for the iPXE shell; reboot in 10 seconds. -
@bmyers change your exit type to, I’m just guessing, grub.
-
@Tom-Elliott Thank you Tom. That fixed the CentOS 5 version but the CentOS 6 imaging is still a problem. I did try the Grub with that one also as stated above.
-
@bmyers said:
I did try the Grub with that one also as stated above.
Is that one an UEFI machine? Have you tried all the other exit types? Ah right… You also changed boot option in the BIOS to boot straight from disk and still didn’t work, right?
I did use resizable image since we have had some systems replaced with a different size hard drive in the past. Let me look into FOG Trunk and see what that is about.
The file listing and stuff you posted shows a non-resizable image. Maybe you changed the image type somewhere along the way? Please create a complete new image (whichever type you like), fresh upload, then download and see if you still have the issue. Other than that you better make sure you have the IDE/AHCI settings all matching on your systems!
