Diskfull RAID1

[jw-26-sudo]

Hello everybody,

I am using TrinityX version 15.3 for my Proxmox VM.

I am trying out the diskfull RAID1 scripts via https://docs.clustervision.com/recipes/disk-full-node/.

I am currently encountering these errors:

Do I need to create hardware/software RAID before using the diskfull RAID1 scripts? What are the other things to look out for?

Thank you.

Regards,
Jing Wei

[aphmschonewille]

It looks like you did a double /dev definition?
Did you have a look at https://docs.clustervision.com/howtos/disk-full-node/#using-the-included-diskfull-plugin-script ?

The raid1 script sets up a software raid on 2 disks. More info can be found here: https://docs.clustervision.com/howtos/disk-full-node/#using-the-included-raid1-plugin-script

By default the suggested disks are basic. These should be configured to match your disk(s)

[jw-26-sudo]

Hi Antoine,

I followed the doc and tried removing the double dev definition. But the issue still persists.

Regards,
Jing Wei

[jw-26-sudo]

Hi Antoine,

Just to update.

I have used the following in my TrinityX portal:

SCRIPTS

diskfull-raid1

PRE SCRIPT

cat << EOF >> /tmp/my-local-disk.sh
export MY_LOCAL_DISK1_NAME=/dev/sda # <--- make sure this is correct!
export MY_LOCAL_DISK2_NAME=/dev/sdb # <--- this one too
export MY_LOCAL_DISK1_SECTORS= # nr of sectors, optional. if defined, then verified.
export MY_LOCAL_DISK2_SECTORS= # nr of sectors, optional. if defined, then verified.
export PARTITION_MY_DISK=yes
export FORMAT_MY_DISK=yes
export MAKE_BOOT=yes # configures and installs grub/shim for standalone boots
EOF

The end results I have achieved is this:

And whenever I want to make any changes, especially when turning off NET BOOT, it shows this error:

Please advise if I should create a new group in my TrinityX cluster.

Thank you.

Regards,
Jing Wei

[jw-26-sudo]

Hi Antoine,

In addition, please advise if I should use TrinityX release version 15.3u1 instead. Understand that there are no solutions yet for the Ubuntu diskfull setup based on #505

Thank you.

Regards,
Jing Wei

[jw-26-sudo]

Hi Antoine,

I should use this instead:

SCRIPTS

raid1

And I am able to do RAID1 mount as per below screenshot:

Note that I can turn NET BOOT both on and off.

Finally, can I conclude that the Ubuntu diskfull setup solutions will be likely available in the coming years?

Thank you.

Regards,
Jing Wei

[jw-26-sudo]

Hi Antoine,

Based on the initial comment, can I confirm if the RAID1 partscript and postscript from the below link:

https://docs.clustervision.com/recipes/disk-full-node/

are still pending fixes?

Seems like the double /dev definition was from the RAID1 partscript I have followed previously.

Appreciate the soonest updates.

Thank you.

Regards,
Jing Wei

[aphmschonewille]

Hi Jing,
No pending fixes for the current implementation. These should work as proposed. However....
We're working on desigs for a whole new approach towards diskless configs with the goal to make this process easier. Implementation to follow, first for EL, later for ubuntu as well, but it's not certain when it will be available as it's not an easy feat. Current target is early Q3 but likely to change.
-Antoine

[jw-26-sudo]

Hi Antoine,

So for RAID1 diskfull script(s), we should follow this instead?

https://docs.clustervision.com/howtos/disk-full-node/#using-the-included-diskfull-plugin-script

Seems like the double /dev definition was from this partscript previously.

DISK1=/dev/sda
DISK2=/dev/sdb

for DISK in $DISK1 $DISK2; do
parted /dev/$DISK -s 'rm 1; rm 2; rm 3'
parted /dev/$DISK -s 'mklabel gpt'
parted /dev/$DISK -s 'mkpart efi fat32 1 512m'
parted /dev/$DISK -s 'mkpart boot ext2 512m 1g'
parted /dev/$DISK -s 'mkpart root ext4 1g 100%'
parted /dev/$DISK -s 'set 1 boot on'
parted /dev/$DISK -s 'set 2 raid on'
parted /dev/$DISK -s 'set 3 raid on'
parted /dev/$DISK -s 'name 1 "EFI System Partition"'
done
while [[ ! -b /dev/sdb3 ]]; do sleep 1; done

echo y | mdadm --create /dev/md0 --metadata 1.0 --force --assume-clean --level=mirror --raid-devices=2 /dev/${DISK1}p1 /dev/${DISK2}p1
mkfs.fat -F 16 /dev/md0
echo y | mdadm --create /dev/md1 --metadata 1.0 --force --assume-clean --level=mirror --raid-devices=2 /dev/${DISK1}p2 /dev/${DISK2}p2
mkfs.ext4 /dev/md1
echo y | mdadm --create /dev/md2 --metadata 1.2 --force --assume-clean --level=mirror --raid-devices=2 /dev/${DISK1}p3 /dev/${DISK2}p3
mkfs.ext4 /dev/md2

mount /dev/md2 /sysroot
mkdir -p /sysroot/boot
mount /dev/md1 /sysroot/boot
mkdir -p /sysroot/boot/efi
mount /dev/md0 /sysroot/boot/efi

Thank you.

Regards,
Jing Wei

[aphmschonewille]

Hi Jing,

Not sure where you have found the (old) part script, but if you set (just) the below block in -pre for the node or group (*):

cat << EOF > /tmp/my-local-disk.sh
export MY_LOCAL_DISK1_NAME=/dev/sda
export MY_LOCAL_DISK2_NAME=/dev/sdb
export MY_LOCAL_DISK1_SECTORS=   # nr of sectors, optional. if defined, then verified.
export MY_LOCAL_DISK2_SECTORS=   # nr of sectors, optional. if defined, then verified.
export PARTITION_MY_DISK=yes
export FORMAT_MY_DISK=yes
export MAKE_BOOT=yes             # configures and installs grub/shim for standalone boots
EOF

While using the right disk names, clear -part and just use the raid1 script, it should work.

*) luna group change -qpart "" -pre -s raid1 <group>
And/or
luna node change -qpart "" -pre -s raid1 <node>
Would open an editor where above block can be copied, pasted and modified to one's needs

The disks could be any /dev device as well as any /dev/disk/by-* device incl nvmes.

Antoine