Weird BalenaCloud dashboard in relation to storage

[tim-littlefair]

It looks like the dashboard is telling me that this device has over 23G used out of a capacity of under 14G - and it seems confused about which end of the storage bar to mark red.

View is https://dashboard.balena-cloud.com/devices/00a4376f76ba9a3b57387a9913401b70/summary

I’ve enabled support access on that device in case anyone at balena.io is interested in taking a look.

Device seems to be working fine, I suspect the SD card is 32GB, will take it offline and update this post with the partition structure when I finish the tasks I booted it up for.

Both OS and supervisor version are updated today - no reboot or power cycle since either update, so I’ll check whether the issue goes away when I do get round to bringing it down and up again.

[tim-littlefair]

It’s a 16GB class 10 Patriot microSD card, the output of lsblk (on an Ubuntu workstation, not the Pi) looks like this:

mmcblk0 179:0 0 14.8G 0 disk
├─mmcblk0p1 179:1 0 40M 0 part
├─mmcblk0p2 179:2 0 320M 0 part
├─mmcblk0p3 179:3 0 320M 0 part
├─mmcblk0p4 179:4 0 1K 0 part
├─mmcblk0p5 179:5 0 20M 0 part
└─mmcblk0p6 179:6 0 14.1G 0 part

Viewing the device an BalenaCloud dashboard after a power cycle still has the same weird pattern - the storage statistic reported is “21.3 GB/13.8 GB”, the bar indicator for storage is grey on the left and red on the right.

Output of ‘df’ run on the Host OS on the device:

Connecting to 00a4376f76ba9a3b57387a9913401b70…
Spawning shell…

Welcome to balenaOS

=============================================================
root@00a4376:~# df
Filesystem                     1K-blocks    Used Available Use% Mounted on
devtmpfs                          920784       0    920784   0% /dev
tmpfs                             971312    9336    961976   1% /run
/dev/mmcblk0p3                    307087  283778      2829 100% /mnt/sysroot/active
/dev/disk/by-state/resin-state     17849      90     16326   1% /mnt/state
overlay                           307087  283778      2829 100% /
/dev/mmcblk0p6                  14467208 2032700  11676520  15% /mnt/data
tmpfs                             971312       0    971312   0% /dev/shm
tmpfs                               4096       0      4096   0% /sys/fs/cgroup
tmpfs                             971312       8    971304   1% /tmp
/dev/mmcblk0p1                     40314    8470     31844  22% /mnt/boot
tmpfs                             971312      72    971240   1% /var/volatile
/dev/mmcblk0p2                    307087  281880      4727  99% /mnt/sysroot/inactive

[Firat]

Hello,

We have the same issue:

Used supervisor version: 17.6.1

BR

Firat

[alanb128]

Thanks for bringing this to our attention. I haven’t been able to reproduce this so far. Can you provide the UI version from your balenaCloud dashboard, in the lower left under “Changelog”?

[cywang117]

Hi @tim-littlefair ,

We traced this to a bump in systeminformation, the Supervisor’s system info reporting library. The affected version is limited to 17.6.1. We’ll be releasing 17.6.2 soon so keep an eye out for that, and let us know if the issue persists.

Regards,

Christina

[tim-littlefair]

Dashboard version is reported as 32.1.3.

As of power on this morning, the storage is reported as 30.0 GB/13.8 GB and the red area which was at the right end of the bar has shrunken to nothing.

[tim-littlefair]

Balena supervisor 17.6.2 was on offer so I installed it. Installation was almost instantaneous, and the data now looks correct:

I’ll mark this as solved.

Thanks @alanb128 and @cywang117 for rapid action on this.