It shows
0000000000181250-0000000000181824 0000000000184322-0000000000185029 0000000000188025-0000000000188417 db.bin-wal snapshot-4-188417-4741894662.meta
0000000000181825-0000000000182273 0000000000185030-0000000000185345 0000000000188418-0000000000189021 metadata1 snapshot-4-189441-5402297704
0000000000182274-0000000000182849 0000000000185346-0000000000186042 0000000000189022-0000000000189441 metadata2 snapshot-4-189441-5402297704.meta
0000000000182850-0000000000182901 0000000000186043-0000000000186369 0000000000189442-0000000000189766 open-1
0000000000182902-0000000000183297 0000000000186370-0000000000186701 0000000000189767-0000000000189810 open-2
0000000000183298-0000000000184004 0000000000186702-0000000000187393 0000000000189811-0000000000189988 open-3
0000000000184005-0000000000184321 0000000000187394-0000000000188024 db.bin snapshot-4-188417-4741894662
now.
The lxc ls works now. My team had followed it up.
The Solution from NITK Team (Thanks to Sushant Rao) :
LXD has released newer versions where they mandated the use of ZFS > 0.8, but in bionic(4.15 kernel) only ZFS 0.7.5 is there. So added a HWE kernel for bionic(which makes the version 5.4.0 & makes ZFS 0.8.5). This was the issue in both of the systems.
But one server had a problem of DQlite file corruption. There were 4 segments that were complete when compared against the snapshots captured.
So those corrupted segments were identified and delete. Then LXD restored its state from the last snapshot nearest to that segment.
Recommended : Upgrade the OS to 20.04/22.04 (after taking complete backup of the current machines) - 18.04 is now officially out of support.
Our Sincere gratitude to the community who helped us. 