Questions tagged [iostat]
The iostat command is used for monitoring the CPU utilization and input/output load of devices attached to a system by observing the time the devices are active in relation to their average transfer rates.
51 questions
1
vote
1
answer
80
views
Linux kernel phantom reads
Why if i write to a raw hard disk (without FS) the kernel also makes reads.
$ sudo dd if=/dev/zero of=/dev/sda bs=32k count=1 oflag=direct status=none
$ iostat -xc 1 /dev/sda | grep -E &...
- 973
0
votes
1
answer
250
views
Is there iostat-similar tool that tracks swap area activity and page cache miss?
The iostat tool is able to tell us cpu usage, disk r/w throughput second-by-second.
Is there a similar tool to track swap area activity and page cache miss?
For example, the tool should tell us the ...
0
votes
0
answers
999
views
iostat returns disk utilization greater than 100% while profiling a Beaglebone Black board
I need to profile the performance of software running on a BeagleBone Black (BBB). The BBB has an ARM Cortex-A8 up to 1GHz frequency, 512MB RAM, and 4GB eMMC onboard flash storage. You can find more ...
- 1
0
votes
0
answers
116
views
Our MySQL is read heavy, but iostat reports that almost no reads are taking place. How come?
According to MySQL's STATUS command, we have about 500 reads and ~20 writes to our DB per second.
But iostat is reporting that ~70 writes (w/s) and ~0.5 reads (r/s) are taking place on the ...
- 103
0
votes
0
answers
248
views
How can I monitor whether disk activity is sychronous or asynchronous?
My Google-fu simply cannot find an answer to this.
If I have a process with heavy I/O activity, how can I check whether it's using asynchronous or synchronous writes?
(I want this information to ...
- 213
1
vote
0
answers
297
views
degraded iops and throughput on a linux machine in a cluster
we have a linux-based cluster on AWS with 8 workers.
OS version (taken from /proc/version) is:
Linux version 5.4.0-1029-aws (buildd@lcy01-amd64-021) (gcc version 7.5.0 (Ubuntu 7.5.0-3ubuntu1~18.04)) #...
- 11
0
votes
2
answers
2k
views
In iostat, why are kB_wrtn/s and kB_wrtn the same?
/dev/sdc is a SATA hard drive. Do the kB_read and kB_wrtn fields sometimes, in some situations, show total counts? Here it seems to be just the same as the per second value.
Linux kernel 5.4.0-26-...
- 195
2
votes
0
answers
777
views
Process shows as 100% I/O bound while producing minimal disk activity, disk util is at 100%
We are having quite strange problem. There is a program (cryptocurrency node to be precise) which has local database of all the transactions ever made. Database is huge - around 15 TB. The problem is ...
- 121
0
votes
2
answers
726
views
How do I get the linux kernel to track io stats to a block device I create in a loadable module?
I've been looking and looking and everybody explains the /proc/diskstats file, but nobody seems to explain where that data comes from.
I found this comment:
Just remember that /proc/diskstats is ...
- 143
2
votes
1
answer
2k
views
Does IOSTATS show output since boot or since last execution?
I see conflicting information online about use of IOSTAT. In particular I would like to be able to show an average since boot. Based on information I have read if I have never issued the command ...
- 157
1
vote
1
answer
3k
views
Understanding iostat block measurements
I am trying to understand how data is written to the disk. I'm writing data with dd using various block sizes, but it looks like the disk is always getting hit with the same size blocks, according to ...
- 575
0
votes
1
answer
491
views
How to find which disk is being written to/read from in an LSI HW RAID logical volume?
On this system there is a lot of "await" which is causing slow response. i need to find out which disk behind the LSI logical volume is slowing it down...
IBM blade, 2 HDD with LSI RAID (LV simple ...
- 268
6
votes
1
answer
9k
views
Why is the size of my IO requests being limited, to about 512K?
I read /dev/sda using a 1MiB block size. Linux seems to limit the IO requests to 512KiB an average size of 512KiB. What is happening here? Is there a configuration option for this behaviour?
$ ...
- 53.5k
2
votes
2
answers
4k
views
NVMe disk shows 80% io utilization, partitions show 0% io utilization
I have a CentOS 7 server (kernel 3.10.0-957.12.1.el7.x86_64) with 2 NVMe disks with the following setup:
# lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
nvme0n1 259:0 0 477G 0 ...
- 105
3
votes
1
answer
4k
views
IO wait time is higher than disk utilization. Isn't this impossible? [duplicate]
I am trying to improve my understanding, following this (so far) unanswered question: Possible limiting factor during upgrade of Fedora VM - not disk or cpu or network?
I ran the following test load, ...
- 53.5k