DOKK / manpages / debian 12 / bpfcc-tools / nfsslower-bpfcc.8.en
nfsslower(8) System Manager's Manual nfsslower(8)

nfsslower - Trace slow NFS file operations, with per-event details.

nfsslower [-h] [-j] [-p PID] [min_ms]

This tool traces common NFSv3 & NFSv4 file operations: reads, writes, opens, and getattrs. It measures the time spent in these operations, and prints details for each that exceeded a threshold.

WARNING: See the OVERHEAD section.

By default, a minimum millisecond threshold of 10 is used. If a threshold of 0 is used, all events are printed (warning: verbose).

Since this works by tracing the nfs_file_operations interface functions, it will need updating to match any changes to these functions.

This tool uses kprobes to instrument the kernel for entry and exit information, in the future a preferred way would be to use tracepoints. Currently there aren't any tracepoints available for nfs_read_file, nfs_write_file and nfs_open_file, nfs_getattr does have entry and exit tracepoints but we chose to use kprobes for consistency

Since this uses BPF, only the root user can use this tool.

CONFIG_BPF and bcc.

-p PID Trace this PID only.

Trace output in CSV format.
Minimum I/O latency (duration) to trace, in milliseconds. Default is 10 ms.

# nfsslower
# nfsslower 1
# nfsslower -j 1
# nfsslower 0
# nfsslower -p 181 1

Time of I/O completion since the first I/O seen, in seconds.
Process name.
Process ID.
Type of operation. R == read, W == write, O == open, G == getattr.
File offset for the I/O, in Kbytes.
Size of I/O, in bytes.
Latency (duration) of I/O, measured from when it was issued by VFS to the filesystem, to when it completed. This time is inclusive of RPC latency, network latency, cache lookup, remote fileserver processing latency, etc. Its a more accurate measure of the latency suffered by applications performing NFS read/write calls to a fileserver.
A cached kernel file name (comes from dentry->d_name.name).
Completion timestamp, microseconds (-j only).
File offset, bytes (-j only).
Latency (duration) of the I/O, in microseconds (-j only).

This adds low-overhead instrumentation to NFS operations, including reads and writes from the file system cache. Such read, writes and particularly getattrs can be very frequent (depending on the workload; eg, 1M/sec), at which point the overhead of this tool (even if it prints no "slower" events) can begin to become significant. Measure and quantify before use. If this continues to be a problem, consider switching to a tool that prints in-kernel summaries only. This tool has been tested with NFSv3 & NVSv4, but it might work with NFSv{1,2}, since it is tracing the generic functions from nfs_file_operations.

Note that the overhead of this tool should be less than fileslower(8), as this tool targets NFS functions only, and not all file read/write paths.

This is from bcc.

https://github.com/iovisor/bcc

Also look in the bcc distribution for a companion nfsslower_examples.txt file containing example usage, output, and commentary for this tool.

Linux

Unstable - in development.

Samuel Nair

biosnoop(8), funccount(8), fileslower(8)

2017-09-01 USER COMMANDS