--- name: "iosnoop-perf" section: "8" platform: "linux" package: "perf-tools-unstable" version: "1.0.1~20200130+git49b8cdf-1" release: "Debian 13 (Trixie)" platforms: ["linux"] url: "https://new.carta.tech/iosnoop-perf" --- # iosnoop-perf(8) trace block I/O events as they occur. Uses Linux ftrace. ## NAME iosnoop - trace block I/O events as they occur. Uses Linux ftrace. ## SYNOPSIS **iosnoop** \[-hQst\] \[-d device\] \[-i iotype\] \[-p pid\] \[-n name\] \[duration\] ## DESCRIPTION iosnoop prints block device I/O events as they happen, with useful details such as PID, device, I/O type, block number, I/O size, and latency. This traces disk I/O at the block device interface, using the block: tracepoints. This can help characterize the I/O requested for the storage devices and their resulting performance. I/O completions can also be studied event-by-event for debugging disk and controller I/O scheduling issues. NOTE: Use of a duration buffers I/O, which reduces overheads, but this also introduces a limit to the number of I/O that will be captured. See the duration section in OPTIONS. Since this uses ftrace, only the root user can use this tool. ## REQUIREMENTS FTRACE CONFIG, and the tracepoints block:block\_rq\_insert, block:block\_rq\_issue, and block:block\_rq\_complete, which you may already have enabled and available on recent Linux kernels. And awk. ## OPTIONS - \-d device Only show I/O issued by this device. (eg, "202,1"). This matches the DEV column in the iosnoop output, and is filtered in-kernel. - \-i iotype Only show I/O issued that matches this I/O type. This matches the TYPE column in the iosnoop output, and wildcards ("\*") can be used at the beginning or end (only). Eg, "\*R\*" matches all reads. This is filtered in-kernel. - \-p PID Only show I/O issued by this PID. This filters in-kernel. Note that I/O may be issued indirectly; for example, as the result of a memory allocation, causing dirty buffers (maybe from another PID) to be written to storage. With the -Q option, the identified PID is more accurate, however, LATms now includes queueing time (see the -Q option). - \-n name Only show I/O issued by processes with this name. Partial strings and regular expressions are allowed. This is a post-filter, so all I/O is traced and then filtered in user space. As with PID, this includes indirectly issued I/O, and -Q can be used to improve accuracy (see the -Q option). - \-h Print usage message. - \-Q Use block I/O queue insertion as the start tracepoint (block:block\_rq\_insert), instead of block I/O issue (block:block\_rq\_issue). This makes the following changes: COMM and PID are more likely to identify the origin process, as are -p PID and -n name; STARTs shows queue insert; and LATms shows I/O time including time spent on the block I/O queue. - \-s Include a column for the start time (issue time) of the I/O, in seconds. If the -Q option is used, this is the time the I/O is inserted on the block I/O queue. - \-t Include a column for the completion time of the I/O, in seconds. - duration Set the duration of tracing, in seconds. Trace output will be buffered and printed at the end. This also reduces overheads by buffering in-kernel, instead of printing events as they occur. The ftrace buffer has a fixed size per-CPU (see /sys/kernel/debug/tracing/buffer\_size\_kb). If you think events are missing, try increasing that size (the bufsize\_kb setting in iosnoop). With the default setting (4 Mbytes), I'd expect this to happen around 50k I/O. ## EXAMPLES - Default output, print I/O activity as it occurs: \# **iosnoop** - Buffer for 5 seconds (lower overhead) and write to a file: \# **iosnoop 5 \> outfile** - Trace based on block I/O queue insertion, showing queueing time: \# **iosnoop -Q** - Trace reads only: \# **iosnoop -i '\*R\*'** - Trace I/O issued to device 202,1 only: \# **iosnoop -d 202,1** - Include I/O start and completion timestamps: \# **iosnoop -ts** - Include I/O queueing and completion timestamps: \# **iosnop -Qts** - Trace I/O issued when PID 181 was on-CPU only: \# **iosnoop -p 181** - Trace I/O queued when PID 181 was on-CPU (more accurate), and include queue time: \# **iosnoop -Qp 181** ## FIELDS - COMM Process name (command) for the PID that was on-CPU when the I/O was issued, or inserted if -Q is used. See PID. This column is truncated to 12 characters. - PID Process ID which was on-CPU when the I/O was issued, or inserted if -Q is used. This will usually be the process directly requesting I/O, however, it may also include indirect I/O. For example, a memory allocation by this PID which causes dirty memory from another PID to be flushed to disk. - TYPE Type of I/O. R=read, W=write, M=metadata, S=sync, A=readahead, F=flush or FUA (force unit access), D=discard, E=secure, N=null (not RWFD). - DEV Storage device ID. - BLOCK Disk block for the operation (location, relative to this device). - BYTES Size of the I/O, in bytes. - LATms Latency (time) for the I/O, in milliseconds. ## OVERHEAD By default, iosnoop works without buffering, printing I/O events as they happen (uses trace\_pipe), context switching and consuming CPU to do so. This has a limit of about 10,000 IOPS (depending on your platform), at which point iosnoop will be consuming 1 CPU. The duration mode uses buffering, and can handle much higher IOPS rates, however, the buffer has a limit of about 50,000 I/O, after which events will be dropped. You can tune this with bufsize\_kb, which is per-CPU. Also note that the "-n" option is currently post-filtered, so all events are traced. The overhead may be acceptable in many situations. If it isn't, this tool can be reimplemented in C, or using a different tracer (eg, perf\_events, SystemTap, ktap.) ## SOURCE This is from the perf-tools collection. - https://github.com/brendangregg/perf-tools Also look under the examples directory for a text file containing example usage, output, and commentary for this tool. ## OS Linux ## STABILITY Unstable - in development. ## AUTHOR Brendan Gregg ## SEE ALSO [iolatency(8)](/iolatency/8.md), [iostat(1)](/iostat/1.md), [lsblk(8)](/lsblk/8.md)