-
|
Hi https://github.com/Sithuk/ubuntu-server-zfsbootmenu and I can't find anywhere to set or limit the size of the rpool/ it's setting itself on part 3 of my disk and taking up a whole 499GB. the efi is taking only 537 and and the swap 534. I would like to give rpool 100GB and use the rest for datapool. How do i do this? Also I have 2 500GB disks and when I set mirror instead of single it results with separate efi partitions, mirrored swap and separate rpools. so of course not surprisingly, the whole thing refuses to boot. Am I choosing the wrong thing or should I use single and then partition and add the 2nd disk after installation completes with zpool add or zpool attach? |
Beta Was this translation helpful? Give feedback.
Replies: 2 comments 11 replies
-
|
Hi bossey, "I can't find anywhere to set or limit the size of the rpool" The script is setup to create the root pool and data pool on different physical disks. The rpool partition takes up the remainder of the capacity of the disk after the swap and EFI partitions are created. If you wanted to have two pools on a single disk then you would need to create a partition for the other pool and then create the other pool in that partition. The script isn't configured to do that automatically, so you would need to amend the script to achieve the configuration you want. "Am I choosing the wrong thing or should I use single and then partition and add the 2nd disk after installation completes with zpool add or zpool attach?" Setting the topology_root variable to mirror and the disks_root variable to 2 should achieve an rpool with a two disk mirror topology. You shouldn't need any manual steps. The separate EFI partitions shouldn't be an issue. rEFInd will auto-select one to boot from, or you can intervene and select which to boot from at the rEFInd menu screen. The contents of both EFI partitions should be identical so it doesn't matter which EFI partition the system boots zfsbootmenu from. There have been other threads about boot issues with particular hardware configurations so the issue may not be the script. |
Beta Was this translation helpful? Give feedback.
-
I have now been able to install with 2 EFI parts, mirrored boot and mirrored rpool, by specifying a size of 102400 MiB for the rpool. |
Beta Was this translation helpful? Give feedback.
-
|
Could you paste the contents of /tmp/data_creation.sh ? Check for any sensitive information before posting the contents. |
Beta Was this translation helpful? Give feedback.
-
|
Is that after initial or postreboot? |
Beta Was this translation helpful? Give feedback.
-
|
It should be created as part of running the datapool creation function. The file shouldn't be created until you run the data pool creation function at least once. I'd like to see what commands are being run to create the data pool. |
Beta Was this translation helpful? Give feedback.
-
|
okay here's the disk layout:
Starting datapool script`$ sudo bash ./ubuntu_server_encrypted_root_zfs.sh datapool Aug 19 15:51:17 poma systemd[1]: Starting systemd-timesyncd.service - Network Time Synchronization... cannot open 'tank1': no such pool Checking script variables for data pool... Carefully enter the ID of the disk(s) YOU WANT TO DESTROY in the next step to ensure no data is accidentally lost. Option number 2 selected: ata-WDC_WDS500G1R0A-25A7W0_24101S800916' Select option: Clearing partition table on disk ata-WDC_WDS500G1R0A-68A4W0_24101S800718. Warning: The kernel is still using the old partition table. I think because of the above error the datapool never gets created so there is no /tmp/data_creation.sh |
Beta Was this translation helpful? Give feedback.
-
|
What happens when you comment out "clear_partition_table "data"" on line 2074? So change: |
Beta Was this translation helpful? Give feedback.
-
|
Okay just rebooting and will try |
Beta Was this translation helpful? Give feedback.
-
It just stops. I assume there is a check to see if the whole disk has been cleared. I've added screenshots of after the datapool script is run.
I also noticed this on reboot as it took a long time to boot.
|
Beta Was this translation helpful? Give feedback.
-
|
There isn't a check that the whole disk has been cleared. Was there a /tmp/data_creation.sh file created? Uncomment "set -x" on line 9. So change "#set -x" to "set -x". That'll provide more information when the script runs to help debug. |
Beta Was this translation helpful? Give feedback.
-
|
Okay thanks. I'll be out all day today so will do that tonight. Thanks for your patience. 1./ 2./ https://github.com/peternickol/script.zfsbootmenu.ubuntu/blob/master/zfsbootmenu.sh |
Beta Was this translation helpful? Give feedback.
-
|
"I was playing around and decided to use the single install last night which went without a hitch. Could it be best to do that then zpool create a datapool on part 4 and then use zpool attach to add mirrors to all the partitions?" I agree that would probably be the easiest way forward. I'm not familiar with the script you linked. The quickest way forward is likely to be manually creating the datapool. You can then work out the bugs with modifying this script to your configuration at your leisure once you have a working system. |
Beta Was this translation helpful? Give feedback.
Hi bossey,
"I can't find anywhere to set or limit the size of the rpool"
The script is setup to create the root pool and data pool on different physical disks. The rpool partition takes up the remainder of the capacity of the disk after the swap and EFI partitions are created. If you wanted to have two pools on a single disk then you would need to create a partition for the other pool and then create the other pool in that partition. The script isn't configured to do that automatically, so you would need to amend the script to achieve the configuration you want.
"Am I choosing the wrong thing or should I use single and then partition and add the 2nd disk after installation completes with …