Dashboard

The dashboard is the start page of the interface and shows the state of the system at a glance. It refreshes every 5 seconds while the browser tab is visible. The dashboard only reads and never changes the system.

Figures at the top

• Uptime since the last boot.
• CPU: usage since the previous sample and the load of the last minute.
• Memory: share in use (including the ZFS ARC).
• Pools: how many pools are healthy. A pool counts as unhealthy when it is not ONLINE or reports data or scrub errors.

When a new Bunkerbox version is available, a notice with a link to Updates appears above.

Pools

For every ZFS pool: state (ONLINE, DEGRADED, …), usage as a bar (yellow from 80 %, red from 90 %), used and free space and the result of the last scrub or resilver. A running scrub shows progress and remaining time.

Disks

All disks with model, type (HDD, SSD, NVMe), size, temperature, SMART health and usage (system disk, pool or free). The system disk is always listed first. Temperatures are colour coded:

Type	yellow from	red from
HDD	46 °C	55 °C
SSD	60 °C	70 °C
NVMe	65 °C	75 °C

CPU

Model, number of threads, total usage, load (1, 5 and 15 minutes), temperature and one bar per thread. Usage is the difference between two samples of /proc/stat, so right after the start the card briefly shows Measuring …. The temperature comes from sensors -j (package lm-sensors) or, without it, directly from /sys/class/hwmon. The CPU package sensor is preferred (Intel Package id 0, AMD Tctl); yellow from 80 °C, red from 90 °C.

Memory

The bar splits memory into programs, ZFS ARC, cache and free. The ARC is the ZFS read cache. Linux counts it as used, but it is released when programs need memory. How large the ARC may grow is set by the automatic ARC tuning under Storage. Below the bar: available memory, swap, the ARC hit rate and the ARC target and maximum.

Network

Per interface: link state and speed (from /sys/class/net/<name>/speed), current receive and send throughput (from /proc/net/dev, difference to the previous sample) and the IP addresses. Loopback, docker0, veth* and Docker bridges (br-…) are hidden.

Sessions and open files

• SMB sessions: user, computer, connected shares and protocol (from smbstatus -j).
• Open files: the first 8 files opened over SMB with share, user and access mode.
• NFS clients: connected clients with their NFS version (from /proc/fs/nfsd/clients).

Apps and updates

The apps card shows whether Docker is installed and running, and how many stacks and containers are running. The update card shows the installed and the latest version and the time of the last check.

When something is missing

Every source is queried on its own and may take at most 8 seconds. If one fails (for example because Samba is not installed), only the affected card shows not available. The rest of the dashboard is still shown.