Skip to content

Output

Every export writes one .zip. The three formats hold the same data in the same sections — only the rendering differs — so pick the one that fits who reads it.

Format Best for Inside the zip
Excel --format Xlsx (default) Analysts and capacity planning: native filters, sorting and pivots in Excel or LibreOffice Calc report.xlsx, one sheet per section and per node, VM and container
HTML --format Html Sharing: a static site that opens offline in any browser, with a sidebar, search, light and dark theme index.html, one page per section and per node, VM and container, assets/
JSON --format Json Automation: scripts, CI, snapshot diffs, Power BI metadata.json, one file per section and per node, VM and container

Every zip also contains network-diagram.svg, the network diagram of the cluster.

Open the Summary — the first sheet, or index.html. It lists every section with its row count and the time it took, and links to each one. If the report has an Issues page, part of the cluster could not be read — see Troubleshooting.

Sections come in this order: the cluster at a glance, then the inventory, then time series, then the cluster deep-dives. A section whose setting is off is left out. The Sections pages list every table and column; Settings explains each switch.

Section What it holds Written when
Summary Date, application version, filters applied, every section with its row count and duration, linked always
Issues API calls that failed while reading the cluster — see Troubleshooting at least one call failed
Cluster Status, options, firewall options, backup jobs, replication, storages, metric servers, mappings Cluster.Include
Storages Every storage with size, usage and type always
Nodes Every node with hardware, subscription, DNS, kernel always
VMs Every QEMU VM with guest agent info, OS, BIOS, CPU, memory, disks always
Containers Every LXC container with hostname, swap, nameserver, privileges always
Network Node interfaces, SDN VNets and the NICs of every VM and container always
Storage Content ISO images, templates, disk images Storage.IncludeContent
Backups Backup files on every storage Storage.IncludeBackups
Disks Disks of every VM and container Guest.IncludeDisks
Partitions Filesystems seen inside running VMs by the guest agent Guest.IncludePartitions
Snapshots Snapshots of every VM and container Guest.IncludeSnapshots
Firewall Firewall rules, aliases and IP sets of cluster, nodes, VMs and containers Firewall.Enabled
Replication Replication jobs of every node: last and next sync, errors, duration Node.IncludeReplication
RRD Nodes CPU, memory, swap, disk and network history of every node Node.RrdData.Enabled
RRD Storage Size and usage history of every storage Storage.RrdData.Enabled
RRD Guests CPU, memory, disk and network history of every VM and container Guest.RrdData.Enabled
Syslog System journal of every node, split into date, time, host, service, PID and message Node.Syslog.Enabled
Cluster Access Users, API tokens, two-factor authentication, groups, roles, ACL, realms Cluster.Include
Cluster SDN SDN zones, VNets, controllers, IPAMs, subnets Cluster.Include
Cluster HA HA resources, groups and status Cluster.Include
Cluster Pools Pools with their VMs, containers and storages Cluster.Include
Cluster Log Cluster log with user, node, service and message Cluster.Log.Enabled
Cluster Tasks Recent tasks of the whole cluster with status, duration and node Cluster.IncludeTasks
Node name One per online node: services, network, /etc/hosts, disks, SMART, ZFS, APT, firewall log, certificates, tasks Node.Detail.Enabled
VM id / CT id One per VM and container: status, configuration (NICs and disks included), guest agent OS info (VMs), firewall log, tasks Guest.Detail.Enabled

The Summary has no Backups row: backup files are counted in the Storage Content row.

Node.Names and Guest.Ids in the settings limit the report to part of the cluster — some nodes, a pool, a tag, a range of IDs. Both default to @all. The Summary shows the filters that were applied, so a partial report is never mistaken for a complete one.

The Nodes, VMs, Containers and Storages sections have a Health column: one number from 0 to 100 that sums up how loaded the resource is right now. Higher is healthier. It lets you sort a list of hundreds of guests and see the busiest first.

Resource Formula
Node 100 − (CPU% × 0.4 + RAM% × 0.4 + Disk% × 0.2)
VM or container, running 100 − (CPU% × 0.5 + RAM% × 0.5)
VM or container, stopped none — not measurable
Storage 100 − Disk%

The result is kept between 0 and 100 and rounded to one decimal. HTML shows it as a badge: green from 80, yellow from 60, red below. Excel colours the column on a red–yellow–green scale with yellow at 60. JSON has the number under the healthScore key. The formula is the same one cv4pve-admin uses, so the two tools show the same score.

The score is a snapshot of usage at the moment of the report, not a diagnosis. To find what is actually wrong, use cv4pve-diag.

Every table is sorted, and numbers inside names are compared as numbers: 105 before 1000, vmbr2 before vmbr10, net2 before net10. Two reports of an unchanged cluster differ only in live values — usage, uptime, RRD — so comparing two of them shows real changes. See snapshot diffs.