vastnfs-ctl
This helper script is installed with the package, and can assist in various situations.
The usage of this script is optional in most cases.
Status
Observe the status of the loaded NFS services and kernel modules
vastnfs-ctl status
Example output:
version: v4.0-pre4-11-ge520dd8eb9cb
kernel modules: sunrpc rpcrdma compat_nfs_ssc lockd nfs_acl auth_rpcgss nfsd rpcsec_gss_krb5 nfs nfsv3 nfsv4
services: rpcbind.socket rpcbind
rpc_pipefs: /var/lib/nfs/rpc_pipefs
Reload
The following command will try to reload the NFS stack. It will first try to unmount all NFS file systems, and then reload all the necessary kernel modules.
vastnfs-ctl reload
Example output:
vastnfs-ctl: stopping service rpcbind.socket
vastnfs-ctl: umounting fs /var/lib/nfs/rpc_pipefs
vastnfs-ctl: unloading kmod nfsv4
vastnfs-ctl: unloading kmod nfsv3
vastnfs-ctl: unloading kmod nfs
vastnfs-ctl: unloading kmod rpcsec_gss_krb5
vastnfs-ctl: unloading kmod nfsd
vastnfs-ctl: unloading kmod auth_rpcgss
vastnfs-ctl: unloading kmod nfs_acl
vastnfs-ctl: unloading kmod lockd
vastnfs-ctl: unloading kmod compat_nfs_ssc
vastnfs-ctl: unloading kmod rpcrdma
vastnfs-ctl: unloading kmod sunrpc
vastnfs-ctl: loading kmod sunrpc
vastnfs-ctl: loading kmod rpcsec_gss_krb5
vastnfs-ctl: loading kmod rpcrdma
vastnfs-ctl: loading kmod nfsv4
vastnfs-ctl: loading kmod nfsv3
vastnfs-ctl: loading kmod nfsd
vastnfs-ctl: loading kmod nfs_acl
vastnfs-ctl: loading kmod nfs
vastnfs-ctl: loading kmod lockd
vastnfs-ctl: loading kmod compat_nfs_ssc
vastnfs-ctl: loading kmod auth_rpcgss
vastnfs-ctl: mounting fs /var/lib/nfs/rpc_pipefs
vastnfs-ctl: starting service rpcbind.socket
Tracing
The vastnfs-ctl utility comes with a helper command for tracing the NFS
stack. Please see a short tutorial on using vastnfs-ctl trace.
Show NFS Clients
Use vastnfs-ctl nfs-clients path to show the status of NFS client used by specific path.
Use vastnfs-ctl nfs-clients to show status of all active NFS clients.
Example output:
NFS client 3000
state: 0
hostname: server.nfs-testing.com
rpc_clnt: 3
hostname: The used NFS server
rpc_clnt: The RPC client's ID used by the NFS client
Show RPC Clients Used by the NFS Protocol
Use vastnfs-ctl rpc-clients <path> to display the status of RPC clients for a specific path.
Use vastnfs-ctl rpc-clients to display the status of all active RPC clients.
Note: This command shows RPC clients used by the NFS protocol only. Clients used by other protocols aren't included.
Example output:
rpc_client: 3 switch-2002
sunrpc-id: 3
RPC iostats version: 1.1 p/v: 100003/3 (nfs)
xprt: tcp 1012 1 2 0 0 6474 6473 0 77192 3579 128 26235 66718
192.168.23.11, state: CONNECTED BOUND
remote_port_idx: 0
role: REG
pci device:
hardware gpus:
remoteports_offset: 0 0
spread reads: 0 writes: 0
mdconnect: 0
per-op statistics
NULL: 1 1 0 44 24 0 0 0 0
GETATTR: 0 0 0 0 0 0 0 0 0
SETATTR: 0 0 0 0 0 0 0 0 0
LOOKUP: 0 0 0 0 0 0 0 0 0
ACCESS: 0 0 0 0 0 0 0 0 0
READLINK: 0 0 0 0 0 0 0 0 0
READ: 0 0 0 0 0 0 0 0 0
WRITE: 0 0 0 0 0 0 0 0 0
CREATE: 0 0 0 0 0 0 0 0 0
MKDIR: 0 0 0 0 0 0 0 0 0
SYMLINK: 0 0 0 0 0 0 0 0 0
MKNOD: 0 0 0 0 0 0 0 0 0
REMOVE: 0 0 0 0 0 0 0 0 0
RMDIR: 0 0 0 0 0 0 0 0 0
RENAME: 0 0 0 0 0 0 0 0 0
LINK: 0 0 0 0 0 0 0 0 0
READDIR: 0 0 0 0 0 0 0 0 0
READDIRPLUS: 0 0 0 0 0 0 0 0 0
FSSTAT: 0 0 0 0 0 0 0 0 0
FSINFO: 0 0 0 0 0 0 0 0 0
PATHCONF: 0 0 0 0 0 0 0 0 0
COMMIT: 0 0 0 0 0 0 0 0 0
rpc_client: The unique ID of the RPC client and the associated transport switch.
xprt: The brief status of transport/transports.
Show RPC Transport Switches
Use vastnfs-ctl rpc-switches <path> to show the status of RPC transport switches for a specific path.
Use vastnfs-ctl rpc-switches to show the status of all active RPC transport switches.
Example output:
RPC switch-2002
num_xprts=1
num_active=1
num_unique_destaddr=1
queue_len=0
xprt-2-tcp: CONNECTED BOUND
queue_len: The number of currently allocated RPC tasks for this transport switch.
xprt-2-tcp: The transport (or list of transports) in use, along with their current status (e.g., CONNECTED, BOUND).
Show RPC Transports
Use vastnfs-ctl rpc-transports <path> to show RPC transports associated with a specific path.
Use vastnfs-ctl rpc-transports to show all active RPC transports used by the NFS protocol.
This command reads sysfs /sys/kernel/sunrpc/xprt-switches/*/xprt*/xprt_info and
related fields for transports belonging to VastNFS mounts.
Example output:
xprt-2-tcp CONNECTED BOUND
dstaddr=192.168.23.11
binding_q_len=0
sending_q_len=0
pending_q_len=0
backlog_q_len=0
tasks_queuelen=0
remote_port_idx=0
reconnection_attempts=0
transport=tcp
The first line shows the transport name (xprt-2-tcp) and current state flags
(CONNECTED, BOUND, etc.). Queue depth lines come from sysfs xprt_info:
| Field | Meaning |
|---|---|
binding_q_len | Tasks waiting for bind/rpcbind |
sending_q_len | Tasks waiting to send |
pending_q_len | Tasks in flight (awaiting reply) |
backlog_q_len | Tasks waiting for a free request slot |
tasks_queuelen | Total tasks associated with this transport |
When a transport is disconnected, xprt_info may report <closed> and queue
fields are not shown.
For continuous monitoring, the Prometheus exporter
exposes these queue depths as
queue length gauges
with xprt_id, netid, and addr labels. Switch-level queue_len from
rpc-switches is not exported by the exporter.
Show RPC Transport Statistics
Use vastnfs-ctl rpc-transports-stats <path> to show per-operation RPC
statistics for transports associated with a specific mount path.
Use vastnfs-ctl rpc-transports-stats to show statistics for all active RPC
transports used by VastNFS mounts.
This command reads the per-op stats section from debugfs
/sys/kernel/debug/sunrpc/rpc_xprt/*/extra for transports belonging to VastNFS
mounts. Rows where all counters are zero are omitted.
Example output:
xprt-2-tcp addr=192.168.23.11:2049 CONNECTED BOUND
per-op stats
op ops trans t/o sent recv q_ms rtt_ms exe_ms errs
NULL 1 1 0 44 24 0 0 0 0
GETATTR 12 12 0 1234 5678 1 2 1 0
READ 100 100 0 1048576 2097152 10 50 40 0
The header line shows the transport name, remote addr, and current state flags.
The table columns map directly to the Prometheus exporter RPC
counters.
The q_ms column is cumulative queue time (milliseconds), not an
instantaneous queue depth — see
queue length gauges
for live depths from rpc-transports.
For continuous monitoring, use the Prometheus exporter.
It exposes these per-op counters with xprt_id, netid, addr, and op
labels, plus per-transport queue depths and state flags from the same sysfs and
debugfs sources as rpc-transports and rpc-transports-stats.
Show Cached NFS Inode State
Use vastnfs-ctl dump-meta-data <path> to show the inode state using the file's absolute path.
Use vastnfs-ctl dump-meta-data :inode@mount_point to show the inode state by inode number and mount point.
Use vastnfs-ctl dump-meta-data :inode@fsid to show the inode state by inode number and fsid, where fsid is the value from the third column of /proc/self/mountinfo.
Example output:
junk.fsxlog fh 0x010001010100000066f43ebcefa3000000000000
idx 0 mode 0100644 owner 0:0 flags 0x82:0xc nlink 1
size 285 state 0x100 version 1879990373217750775 count 0
atime 1750877260079976788ns mtime 1750877474684478199ns ctime 1750877474684478199ns
nfsi flags 0x4 cache validity 0x0 gencount 4942822 attrtimeo 12000
timestamp 4297125548 unlock_mtime 0ns jiffies 217558
access_cache 1 entryes, cached 0 pages flags 0x0
First line displays the file name and its NFS file handle.
Document generated on 2026.09.14 19:54 for v4.5.9, branch 4.5