Hi Gareth, - Smartmon/smarctl does not report any hw issues on the HDDs. - Fragmentation score is 1 (not fragmented at all) - 18% used only - RAID status is green (force-resynced) - rebooted several times - the IO utilization is almost zero(!) - chart attached- tried to change the io scheduler of the disks from mq-deadline to noop: does not bring change.
- the read latency increased on ALL 4 discs by the bullseye upgrade! - no errors/warnings in kern.log/syslog/messages Br, Mihaly On Tue 8 Nov 2022, at 09:48, Vukovics Mihály <vm@informatik.hu> wrote:
Hello Community, since I have upgraded my debian 10 to 11, the read IO wait time of all disks have been increased dramatically.
[...]
Chart attached, you can clearly see the date of the upgrade. Any ideas?
Hello, I'm not an expert in the area, but a few thoughts: --- (Seemingly unlikely issues) Hardware problems coincident with upgrade - smartmon tests? Fragmentation? What % capacity is used? What filesystem? --- RAID status? Have you rebooted?After a quick web search, this article suggests iowait may not be the best indicator of problems (though not sure if this extends to wait times for particular operations)...
https://ioflood.com/blog/2021/01/25/iowait-in-linux-is-iowait-too-high/ ...does iostat's %util give cause for concern? (per the article)HDDs or SSDs? The article suggests ~100ms may be reasonble for HDDs - your chart only seems to show one outstandingly high wait time - do you actually notice a difference in performance?
Not sure if any of that helps but I will follow the thread with interest. Nice chart btw. What produced the data and the chart itself? Best wishes, Gareth -- Köszönettel: Vukovics Mihály
Attachment:
md1-iopersec.png
Description: PNG image
