Unable to extend the datadisk?

Help Automation Suite
1

Greetings, everyone.

I’m trying to run the configureUiPathDisks SH script as documented and have set up all required prereq’s according to the calculator. I am running into the following error for my etcd disk when I run the script:

Physical volume ‘/dev/nvme1n1’ is already in volume group ‘uipathdatavg’
Unable to add physical volume ‘/dev/nvme1n1’ to volume group ‘uipathdatavg’
/dev/nvme1n1: physical volume not initialized.
Physical volume “/dev/nvme1n1” changed
1 physical volume(s) resized or updated / 0 physical volume(s) not resized
New size (131071 extents) matches existing size (131071 extents).
No size change.
[ERROR][2023-09-12T19:25:28+0000]: unable to extend the Datadisk volume: /dev/uipathdatavg/longhornlv

Any insight would be helpful.

Thanks,

Kevin.

2

Hi Kevin,

Did you find a solution ?
I am trying to run the script but have a similar error.
I am looking into the issue.

Kind regards
Julien

[SUCCESS] [2023-09-15T13:29:48+0000]: Data disk is configured successfully at dev location /dev/sdc
  Physical volume '/dev/sdc' is already in volume group 'uipathdatavg'
  Unable to add physical volume '/dev/sdc' to volume group 'uipathdatavg'
  /dev/sdc: physical volume not initialized.
  Physical volume "/dev/sdc" changed
  1 physical volume(s) resized or updated / 0 physical volume(s) not resized
  New size (131071 extents) matches existing size (131071 extents).
  No size change.
[ERROR][2023-09-15T13:29:49+0000]: unable to extend the Datadisk volume: /dev/uipathdatavg/longhornlv

Using the command with --debug parameter leads me to understand that the command that causes this error is

lvextend -r -l +100%FREE /dev/uipathdatavg/longhornlv

Details of the debug here

[SUCCESS] [2023-09-15T13:26:46+0000]: Data disk is configured successfully at dev location /dev/sdc
+ extend_data_disk /dev/sdc
+ local dev_name=/dev/sdc
+ local pvList
+ local iter
+ [[ 0 == \0 ]]
+ extend_vg uipathdatavg /dev/sdc
+ local vg_name=uipathdatavg
+ local dev_name=/dev/sdc
+ vgdisplay -c
+ cut -d : -f1
+ grep -q uipathdatavg
+ vgextend uipathdatavg /dev/sdc
  Physical volume '/dev/sdc' is already in volume group 'uipathdatavg'
  Unable to add physical volume '/dev/sdc' to volume group 'uipathdatavg'
  /dev/sdc: physical volume not initialized.
+ lv_dev_name=/dev/uipathdatavg/longhornlv
++ pvdisplay --select vg_name=uipathdatavg --columns --options pv_name --reportformat basic
++ awk 'NR!=1'
+ pvList='  /dev/sdc  '
+ for iter in $pvList
+ pvresize --yes --quiet /dev/sdc
  Physical volume "/dev/sdc" changed
  1 physical volume(s) resized or updated / 0 physical volume(s) not resized
+ lvextend -r -l +100%FREE /dev/uipathdatavg/longhornlv
  New size (131071 extents) matches existing size (131071 extents).
  No size change.
+ error 'unable to extend the Datadisk volume: /dev/uipathdatavg/longhornlv'
++ date +%Y-%m-%dT%H:%M:%S%z
+ echo -e '\033[0;31m[ERROR][2023-09-15T13:26:47+0000]: unable to extend the Datadisk volume: /dev/uipathdatavg/longhornlv\033[0m\n'
[ERROR][2023-09-15T13:26:47+0000]: unable to extend the Datadisk volume: /dev/uipathdatavg/longhornlv
3

Hi Julien,

I have not yet found a solution. I’m not even sure why the script is attempting to extend the volume if it’s already the right size? Perhaps I’m misunderstanding something.

If you come across something, PLEASE let me know!

4

By any chance are you reusing the same physical disk to be configured as data disks / etcd disks / bundle disks?

Could you post the full command that you ran? Just to see what parameters were provided to the configureUiPathDisks.sh script.

5

Hi,

Here is the command we’re running

configureUiPathDisks.sh --cluster-disk-name /dev/sdd --etcd-disk-name /dev/sde --data-disk-name /dev/sdc --node-type server --install-type online

And here the truncated result of fdisk -l

Disk /dev/sdd: 512 GiB
Disk /dev/sda: 80 GiB
Disk /dev/sdc: 256 GiB
Disk /dev/sde: 512 GiB
Disk /dev/sdb: 16 GiB
Disk /dev/mapper/uipathdatavg-longhornlv: 512 GiB
Disk /dev/mapper/uipathetcdvg-etcdlv: 16 GiB
Disk /dev/mapper/uipathvg-rancherlv: 190 GiB
Disk /dev/mapper/uipathvg-kubeletlv: 56 GiB
Disk /dev/mapper/uipathvg-installerlv: 10 GiB
6

Other than mounting the disks, was there anything else done to the disks prior to running the script?

I had previously run the script for v2022.10 of Automation Suite without issues. Perhaps it’s worth a try to see if it works, to narrow down the issue with the 2023.4 script?
docs.uipath.com/automation-suite/automation-suite/2022.10/installation-guide/downloading-installation-packages

7

So, this problem only seems to exist (for me, anyways) in RHEL 9. In RHEL 8, I can configure the disks using the script properly.

Are there any plans in the near future to support RHEL 9?

8

Hi Kevin,

Thanks for raising your issue! We currently only support RHEL 8 (up to 8.8 today) and we will evaluate supporting RHEL 9 in the future. Can you provide some information on what you currently have or support in your environment? Do you only have RHEL 9 or you can still provision RHEL 8 and install Automation Suite?

Thank you,
Alex

9

We can still provision RHEL 8 in AWS to deploy Automation Suite, so all is well!

Thank you!