Commit ae6ec29
authored
feat(disk): add Lustre filesystem support for Azure Managed Lustre and AWS FSx (#1189)
Add support for Lustre filesystem type in DefaultNFSFsTypeFunc(),
enabling
gpud to properly monitor Lustre mounts (Azure Managed Lustre, AWS FSx
for
Lustre) as shared storage volumes.
Previously, when customers configured Lustre paths as NFS volumes, the
NFS
component would fail validation because Lustre's fstype is "lustre", not
"nfs*". This caused nodes to be marked degraded even though the storage
was accessible. As a workaround, customers mounted Lustre as node-local
volumes, but this bypassed health monitoring entirely - when mounts
failed,
nodes weren't marked degraded and jobs were scheduled on broken nodes.
This change:
- Adds "lustre" to DefaultNFSFsTypeFunc() alongside wekafs, virtiofs,
nfs*
- Adds comprehensive test coverage for Lustre mount detection
- Includes real-world lsblk JSON test data from Azure customer
environment
Signed-off-by: Gyuho Lee <gyuhol@nvidia.com>
Signed-off-by: Gyuho Lee <gyuhol@nvidia.com>1 parent 20c7ae2 commit ae6ec29
5 files changed
Lines changed: 651 additions & 1 deletion
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
23 | 23 | | |
24 | 24 | | |
25 | 25 | | |
| 26 | + | |
| 27 | + | |
| 28 | + | |
26 | 29 | | |
27 | 30 | | |
28 | 31 | | |
| |||
0 commit comments