On Sun, Sep 13, 2020 at 8:48 AM Jeremey Wise <jeremey.wise(a)gmail.com> wrote:
Greetings:
3 x servers each server has 1 x 512GB ssd 2 x 1TB SSD JBOD
Goal: use HCI disk setup wizard to deploy initial structure
Each server has disk scanning in as different /dev/sd# and so trying to
use more clear /dev/mapper/<disk ID>
As such I set this per table below:
# Select each server and set each drive <<<<<<<<<<Double
Check drive
device ID as they do NOT match up per host
# I transitioned to /dev/mapper object to avoid unclearness of /dev/sd#
thor
/dev/sdc
/dev/mapper/Samsung_SSD_850_PRO_512GB_S250NXAGA15787L
odin
/dev/sdb
/dev/mapper/Micron_1100_MTFDDAV512TBN_17401F699137
medusa
/dev/sdb
/dev/mapper/SAMSUNG_SSD_PM851_mSATA_512GB_S1EWNYAF609306
What is /dev/mapper/Samsung_SSD_850_PRO_512GB_S250NXAGA15787L?
It looks like you are using multipath for local disks, which is not a good
idea. First
the additional layer can harm performance, and second, these disks may be
exposed
to engine as FC storage, which can lead to disaster when you try to use the
local disk
as shared storage.
If you want to use stable names instead of /dev/sd{X}, use /dev/disk/by-id/
symlinks, like:
$ ls -lh /dev/disk/by-id/nvme-eui.5cd2e42a81a11f69
lrwxrwxrwx. 1 root root 13 Sep 7 01:21
/dev/disk/by-id/nvme-eui.5cd2e42a81a11f69 -> ../../nvme0n1
You can find the right stabe name of the device using udevadm:
$ udevadm info --query symlink /dev/nvme0n1
disk/by-id/nvme-INTEL_SSDPEKKF256G8L_BTHH827507DP256B
disk/by-path/pci-0000:3e:00.0-nvme-1 disk/by-id/nvme-eui.5cd2e42a81a11f69
Nir
# Note that drives need to be completely clear of any partition or
file
system
[root@thor /]# gdisk /dev/sdc
GPT fdisk (gdisk) version 1.0.3
Partition table scan:
MBR: protective
BSD: not present
APM: not present
GPT: present
Found valid GPT with protective MBR; using GPT.
Command (? for help): x
Expert command (? for help): z
About to wipe out GPT on /dev/sdc. Proceed? (Y/N): y
GPT data structures destroyed! You may now partition the disk using fdisk
or
other utilities.
Blank out MBR? (Y/N): y
[image: image.png]
[image: image.png]
But deployment fails with Error:
<snip>
TASK [gluster.infra/roles/backend_setup : Filter none-existing devices]
********
task path:
/etc/ansible/roles/gluster.infra/roles/backend_setup/tasks/get_vg_groupings.yml:38
ok: [thorst.penguinpages.local] => {"ansible_facts":
{"gluster_volumes_by_groupname": {}}, "changed": false}
ok: [odinst.penguinpages.local] => {"ansible_facts":
{"gluster_volumes_by_groupname": {}}, "changed": false}
ok: [medusast.penguinpages.local] => {"ansible_facts":
{"gluster_volumes_by_groupname": {}}, "changed": false}
TASK [gluster.infra/roles/backend_setup : Make sure thick pvs exists in
volume group] ***
task path:
/etc/ansible/roles/gluster.infra/roles/backend_setup/tasks/thick_lv_create.yml:37
TASK [gluster.infra/roles/backend_setup : update LVM fact's]
*******************
task path:
/etc/ansible/roles/gluster.infra/roles/backend_setup/tasks/thick_lv_create.yml:83
skipping: [thorst.penguinpages.local] => {"changed": false,
"skip_reason":
"Conditional result was False"}
skipping: [odinst.penguinpages.local] => {"changed": false,
"skip_reason":
"Conditional result was False"}
skipping: [medusast.penguinpages.local] => {"changed": false,
"skip_reason": "Conditional result was False"}
TASK [gluster.infra/roles/backend_setup : Create thick logical volume]
*********
task path:
/etc/ansible/roles/gluster.infra/roles/backend_setup/tasks/thick_lv_create.yml:90
failed: [medusast.penguinpages.local] (item={'vgname': '
gluster_vg_Samsung_SSD_850_PRO_512GB_S250NXAGA15787L', 'lvname':
'gluster_lv_engine', 'size': '1000G'}) =>
{"ansible_index_var": "index",
"ansible_loop_var": "item", "changed": false,
"err": " Volume group
\"gluster_vg_Samsung_SSD_850_PRO_512GB_S250NXAGA15787L\" not found.\n
Cannot process volume group
gluster_vg_Samsung_SSD_850_PRO_512GB_S250NXAGA15787L\n", "index": 0,
"item": {"lvname": "gluster_lv_engine", "size":
"1000G", "vgname":
"gluster_vg_Samsung_SSD_850_PRO_512GB_S250NXAGA15787L"}, "msg":
"Volume
group gluster_vg_Samsung_SSD_850_PRO_512GB_S250NXAGA15787L does not
exist.", "rc": 5}
changed: [thorst.penguinpages.local] => (item={'vgname': '
gluster_vg_Samsung_SSD_850_PRO_512GB_S250NXAGA15787L', 'lvname':
'gluster_lv_engine', 'size': '1000G'}) =>
{"ansible_index_var": "index",
"ansible_loop_var": "item", "changed": true,
"index": 0, "item": {"lvname":
"gluster_lv_engine", "size": "1000G", "vgname":
"gluster_vg_Samsung_SSD_850_PRO_512GB_S250NXAGA15787L"}, "msg":
""}
failed: [odinst.penguinpages.local] (item={'vgname':
'gluster_vg_Samsung_SSD_850_PRO_512GB_S250NXAGA15787L', 'lvname':
'gluster_lv_engine', 'size': '1000G'}) =>
{"ansible_index_var": "index",
"ansible_loop_var": "item", "changed": false,
"err": " Volume group
\"gluster_vg_Samsung_SSD_850_PRO_512GB_S250NXAGA15787L\" not found.\n
Cannot process volume group
gluster_vg_Samsung_SSD_850_PRO_512GB_S250NXAGA15787L\n", "index": 0,
"item": {"lvname": "gluster_lv_engine", "size":
"1000G", "vgname":
"gluster_vg_Samsung_SSD_850_PRO_512GB_S250NXAGA15787L"}, "msg":
"Volume
group gluster_vg_Samsung_SSD_850_PRO_512GB_S250NXAGA15787L does not
exist.", "rc": 5}
NO MORE HOSTS LEFT
*************************************************************
NO MORE HOSTS LEFT
*************************************************************
PLAY RECAP
*********************************************************************
medusast.penguinpages.local : ok=23 changed=5 unreachable=0
failed=1 skipped=34 rescued=0 ignored=0
odinst.penguinpages.local : ok=23 changed=5 unreachable=0
failed=1 skipped=34 rescued=0 ignored=0
thorst.penguinpages.local : ok=30 changed=9 unreachable=0
failed=0 skipped=29 rescued=0 ignored=0
Please check /var/log/cockpit/ovirt-dashboard/gluster-deployment.log for
more informations.
############
Why is oVirt ignoring when I set (and double check explicite device call
for deployment?
Attached is the ansible file it creates and then one I had to edit to
correct what wizard should have built it as.
--
p <jeremey.wise(a)gmail.com>enguinpages
_______________________________________________
Users mailing list -- users(a)ovirt.org
To unsubscribe send an email to users-leave(a)ovirt.org
Privacy Statement:
https://www.ovirt.org/privacy-policy.html
oVirt Code of Conduct:
https://www.ovirt.org/community/about/community-guidelines/
List Archives: