flood
flood copied to clipboard
Flood-UI
Disk usage not reporting correctly
Type: Bug Report
- [x] Try to follow the update procedure described in the README and try again before opening this issue.
- [x] Please check the F.A.Q..
- [x] Please check the Troubleshooting wiki section.
Your Environment
Flood is running in a Docker container using the provided Dockerfile.
- Version used: 86008d5ffb5d073278bf29f46b874b85495980ed
Docker engine version
Version: 18.06.1-ce API version: 1.38 (minimum version 1.12) Go version: go1.10.3 Git commit: e68fc7a Built: Tue Aug 21 17:23:21 2018 OS/Arch: linux/amd64 Experimental: false
- Operating System and version (inside container):
Alpine Linux 3.9.4
Summary
Disk widget indicates NaN%
Expected Behavior
Disk usage widget reporting space usage
Current Behavior
NaN% is show instead of targeted storage usage
Possible Solution
Review df options to take account of various Linux distributions
Steps to Reproduce
- Build docker image at root of repository
- Run it with a volume to host downloaded data
- Download torrent
Appendix
Output of the "df" command executed inside the container
/usr/src/app # df --block-size=1024 --portability df: unrecognized option: block-size=1024 BusyBox v1.29.3 (2019-01-24 07:45:07 UTC) multi-call binary.Usage: df [-PkmhTai] [-B SIZE] [FILESYSTEM]...
Print filesystem usage statistics
-P POSIX output format -k 1024-byte blocks (default) -m 1M-byte blocks -h Human readable (e.g. 1K 243M 2G) -T Print filesystem type -a Show all filesystems -i Inodes -B SIZE Blocksize
Output of "df -B 1024" executed inside the container
/usr/src/app # df -B 1024 Filesystem 1K-blocks Used Available Use% Mounted on overlay 1902052484 1527146244 278264552 85% / tmpfs 65536 0 65536 0% /dev tmpfs 8165532 0 8165532 0% /sys/fs/cgroup /dev/sda3 1902052484 1527146244 278264552 85% /data /dev/sda3 1902052484 1527146244 278264552 85% /data/rtorrent.sock /dev/sda3 1902052484 1527146244 278264552 85% /etc/resolv.conf /dev/sda3 1902052484 1527146244 278264552 85% /etc/hostname /dev/sda3 1902052484 1527146244 278264552 85% /etc/hosts shm 65536 0 65536 0% /dev/shm /dev/sda3 1902052484 1527146244 278264552 85% /data/server/db /dev/sda3 1902052484 1527146244 278264552 85% /home/data/storage tmpfs 8165532 0 8165532 0% /proc/acpi tmpfs 65536 0 65536 0% /proc/kcore tmpfs 65536 0 65536 0% /proc/keys tmpfs 65536 0 65536 0% /proc/timer_list tmpfs 65536 0 65536 0% /proc/sched_debug tmpfs 8165532 0 8165532 0% /proc/scsi tmpfs 8165532 0 8165532 0% /sys/firmware
I have flood running in a docker container, I have added the volume for my NAS, but I am still getting the above error. can someone explain exactly where this is looking?
@MattMattV I'm not a dev, only issue manager. Only @jfurrow is a dev on this project but he doesn't have much time.
To see if it's a bug only happening with docker or not, please install Flood on a machine and see if the bug is happening. If only happening with docker it's maybe related in the way you mount volumes and how docker see them.
@MattMattV it looks like it can be resolved by changing server/services/diskUsageService.js
on line 22 change 'df --block-size=1024 --portability | tail -n+2' to 'df -k | tail -n+2'
I have not tested this on my debian machine yet, just in docker, but it is working.
https://github.com/Flood-UI/flood/blob/86008d5ffb5d073278bf29f46b874b85495980ed/server/services/diskUsageService.js#L22
into df -k | tail -n+2
By the way using --block-size is useless when not using something like -h because bytes are always bytes so when you display them without grouping (when not using human readable format) there is no grouping by 1000 or 1024. So -B is only useful when combined with -h or when is specify the size unit like K,M (power fo 1024) or KB,MB (power of 1000). But anyway using power of 1024 is already the default for df so in this case using -block-size=1024 is two times useless.
If your remove the -P you remove the posix format output.
If you add -k = --block-size=1K = -B 1K which is still useless because you don't use -h. Using -B K will just add a K and using -B KB will change the display.
So the only thing you really did was removing -P. And using -P = --portability only change teh first line
-Filesystem 1K-blocks Used Available Use% Mounted on
+Filesystem 1024-blocks Used Available Capacity Mounted on
So it only change the display of the first line from 1K-blocks to 1024-blocks (which is the same thing) and Use% to Capacity. BUT since we use tail -n+2 to strip the two first lines that's exatly the same output in final.
So in reality df --block-size=1024 --portability | tail -n+2 strictly has the same output as df -k | tail -n+2 and as df | tail -n+2 alone. In the first place the option @jfurrow used where useless but you @slicepaperwords changed nothing.
As a proof look at this :
To convince yourself you can do man df.
I still don't know where does the problem comes from but changing the options of df here changes nothing.
the issue is in the docker alpine image, neither of those 2 options exist for df.
looks like https://github.com/Flood-UI/flood/pull/825 should resolve it for docker as well.
Yeah if df was failing in alpine because the options don't exist on such a platform so not using any option should make it work. You try the change and build the docker image with that change and see if it works @MattMattV. if so @jfurrow could merge it.
Thanks! ah I assumed all dfs on linux were the same haha ;(
So lets hope then that df always return 1K blocks!
By the way @MattMattV , was it obvious enough to find the diskUsageService field in the config template?
To be honest I did not check at all if there was a new setting !
This explains why there is so much entries in my screenshot :smile: