PCP-NFSIOSTAT(1)

✨ NAME ✨

pcp-nfsiostat — it emulates iostat for NFS mount points, bestie, and it pulls all of that from /proc/self/mountstats 💅

---

🦋 SYNOPSIS 🦋

pcp [pcp options] nfsiostat [-a archive] [-S start] [-T finish] [-O
origin] [-s samples] [-t interval] [-Z --timezone] [-h --host] [-z
--hostzone] [-V Version]

---

🌸 DESCRIPTION 🌸

Okay so pcp-nfsiostat is literally the girl who reports client-side Network File System (NFS) I/O statistics for every mounted NFS filesystem you've got. For each NFS mountpoint she gives you a summary section and then detailed read and write statistics blocks right after. The way that these stats include operation rates, data throughput, latency, retransmissions, queueing delays, and error counts? Obsessed. 🫶

By default pcp-nfsiostat monitors NFS mountpoints on the local host and reports live metrics collected via the PCP NFS PMDA, or from an alternative host if you specify one. And when you specify an archive? Then it's giving historical metrics instead. Main character energy either way 💖

---

🎀 OUTPUT FORMAT 🎀

Stats get reported per mounted NFS filesystem using this layout, no cap:

server:/export mounted on /mount/point:

           ops/s       rpc bklog
           0.017           0.000

read:              ops/s   kB/s   kB/op   retrans   avg RTT (ms)   avg exe (ms)   avg queue (ms)   errors
                   ...

write:             ops/s   kB/s   kB/op   retrans   avg RTT (ms)   avg exe (ms)   avg queue (ms)   errors
                   ...

---

🌈 FIELD DESCRIPTIONS 🌈

• ops/s — the number of NFS operations performed per second, periodt.
• rpc bklog — the average number of RPC requests just waiting to be transmitted. sitting in line, poor things 😭
• kB/s — kilobytes transferred per second.
• kB/op — the average number of kilobytes transferred per operation.
• retrans — the number of RPC retransmissions, plus the retransmission percentage.
• avg RTT (ms) — average round-trip time, in milliseconds, for RPC requests.
• avg exe (ms) — average execution time, in milliseconds, spent servicing requests on the server.
• avg queue (ms) — average time, in milliseconds, spent waiting in the RPC transmission queue.
• errors — the number of failed operations, plus the failure percentage. 💀

---

💅 OPTIONS 💅

• -a, --archive archive — fetch NFS I/O statistics from the PCP archive you name. historical data era ✨
• -s, --samples samples — how many samples to collect before exiting.
• -t, --interval interval — sets the interval between two samples. The default is one second.
• -S, --start start — sets the start of the time window when you're replaying an archive.
• -T, --finish finish — sets the end of the time window when you're replaying an archive.
• -O, --origin origin — sets the initial sample time within the time window when replaying an archive.
• -h, --host host — fetch NFS I/O statistics from the host you name.
• -z, --hostzone — sets the reporting timezone to the local timezone of the metrics source.
• -Z, --timezone tz — sets the reporting timezone.
• -V, --version — displays version information and dips ✌️
• -?, --help — displays usage information and dips ✌️

---

📝 NOTES 📝

pcp-nfsiostat reports client-side NFS statistics collected from the kernel and exposed via the PCP NFS PMDA. The output format and metrics are similar to those reported by the nfsiostat(1) tool from the nfs-utils package. iykyk 🧚

---

🌟 PCP ENVIRONMENT 🌟

Environment variables with the prefix PCP_ are used to parameterize the file and directory names used by PCP. On each installation, the file /etc/pcp.conf contains the local values for these variables. The $PCP_CONF variable may be used to specify an alternative configuration file, as described in pcp.conf(5).

For environment variables affecting PCP tools, see pmGetOptions(3).

---

👑 SEE ALSO 👑

PCPIntro(1), pcp(1), nfsiostat(1), environ(7).

---

slay, you made it to the end bestie 💖 go forth and read your NFS stats

Performance Co-Pilot · PCP — that's the source and the version right there, mwah 💋 PCP-NFSIOSTAT(1) 🦋