search

htop-dev / htop

htop - an interactive process viewer
https://htop.dev/
GNU General Public License v2.0
6.56k stars 442 forks source link

IO-rate-graph cuts of spikes > 100% when the IO-rate drops under 100% again #927

Open binarynoise opened 2 years ago

binarynoise commented 2 years ago

I have set my cpu, network and disk activity as graphs. The problem is that the disk activity can grow >100% when multiple disks are attached (like 200% when copying large files from one to the other hdd). When the io activity drops to 100% because one disk is slower than the other, the graph looks like this: grafik and the spikes beyond 100% are cut off. grafik That also happens to the network-IO

My suggestion would be to ensure that the max value is always in scale and maybe lock that scale to multiples of 100%

Explorer09 commented 2 years ago

Currently the graph doesn't support dynamic scaling (but I tried implementing one in the pull request #714 ). Your problem here is probably the "total" value of the Disk IO meter is incorrect when there are multiple disks.

binarynoise commented 2 years ago

The total value is not incorrect (100% per disk, just like with the CPUs), it just has weird scaling. But good to know, that you are already working on that 👍