68

My disk often is utilized, but top (and htop, a custom replacement) show nothing suspicious.

Is there a way to sort processes by I/O (more specific: disk) utilization?

EDIT
Found out using iotop that those strange processes are flush-8:16 and jbd2/sdb3-7. Seems to have to do with usual filesystem operations.

java.is.for.desktop
  • 1,718
  • 3
  • 18
  • 20
  • 1
    If I'm correct flush and jbd concerns the sync of the journal (FS metadata) to the disk. Which means you must have some processes either writing to the disk or reading a lot of data and you have the atime option on your mount. I don't recommend this because some software relies on it (mutt and I have heard one backup tool) but you can set your mount to relatime or even "better" noatime. The latter will completely stop updating the access time (which incures a disk write) each time a file is read. – Huygens Jul 27 '12 at 21:48
  • http://unix.stackexchange.com/questions/55212/how-can-i-monitor-disk-io – Ciro Santilli OurBigBook.com May 18 '16 at 12:37
  • https://serverfault.com/questions/25032/linux-disk-io-load-breakdown-by-filesystem-path-and-or-process/25034#25034 – qwr Jul 05 '19 at 21:44

3 Answers3

68

Have you tried iotop ?

You may need to install it before. Also, it depends on a kernel feature that may or may not enabled in your specific distribution.

b0fh
  • 2,245
  • 16
  • 18
27

You might want to give atop a try. It seems to do a good job of letting you know what is going on.

afluth
  • 371
  • 3
  • 3
5

iostat is still the king of detailed I/O information, the kind of info that can confirm if your SSD is living up to expectations, for example.

$ iostat -xht 5 nvme0n1
Linux 4.15.0-43-generic (myhost)    06/02/2021  _x86_64_    (8 CPU)

06/02/2021 08:59:20 AM
avg-cpu:  %user   %nice %system %iowait  %steal   %idle
           2.6%    0.1%    0.8%    0.1%    0.0%   96.4%

Device            r/s     w/s     rkB/s     wkB/s   rrqm/s   wrqm/s  %rrqm  %wrqm r_await w_await aqu-sz rareq-sz wareq-sz  svctm  %util
nvme0n1
                95.03   20.80      1.8M    300.1k     0.00    13.15   0.0%  38.7%    0.24    0.57   0.03    19.2k    14.4k   0.02   0.3%

Every five seconds, this will print detailed IO stats for the NVMe drive. The most important ones are usually not actually read/write bandwidth -- rKB/s and wKB/s -- but rather the reads and writes per second (r/s and w/s, aka IOPS) and, perhaps most important, the average time a process waited for each read and write -- r_await and w_await (in milliseconds).

All of these values feed into each other using the surprisingly useful Little's law formula, translated a bit to latency = queue_size / IOPS, or await = aqu_sz / (r/s + w/s).

CivMeierFan
  • 179
  • 1
  • 4