Skip to content

Error overview

The error overview (menu item Error Overview) is the central list of all checks with problems — across tenants and hosts. Where the host detail page shows a single host, the error overview shows everything that needs attention right now. On a phone the filters open as a drawer from the bottom.

Categories and status

Two filter rows determine what you see:

Categories — a chip group, selected means "shown" (at least one always stays selected):

Category Meaning Default
Open untreated problems — no ACK, no downtime, no info mode on
ACK acknowledged, still marked as an error — being fixed, still counts as a problem off
Accepted acknowledged and classified as "not an error" — accepted exception; ends by itself if the status gets worse off
Downtime inside a maintenance window off
Dependency down the parent dependency (switch, router …) is currently down; grouped under the culprit on
Info checks in info mode — statistics only, never alert off

Clicking "Info" in the header summary shows all info checks, whatever their real status.

Status row — additionally filters by the real status: CRITICAL, WARNING, NO DATA, UNKNOWN and OK (OK is deselected by default). If entries are hidden by the current selection, a hint "N more entries hidden:" lists one pill per reason (e.g. "N× Status OK") and "Show all".

Newly created checks without a result yet appear as PENDING ("waiting for data") and count as NO DATA when filtering.

A plain-language line always summarises what the current combination shows. Further filters: tenant, profile, check type and host tags (OR-combined).

The search at the top finds parts of device name, display name, check and tenant; with * and ? it matches patterns against the whole name (sw-* = everything starting with "sw-"). There is no separate "host pattern" field any more; saved tabs, links and dashboard widgets with an old host pattern keep filtering, and the field stays visible there until the value is removed.

The Hide nothing switch next to the search field makes only the search term count: all other filters are paused (not deleted) and come back when you switch it off; OK, info, downtime and acknowledged entries are loaded too. The switch is stored in the address (alle=1), not in saved tabs and not in TV mode. If a search only finds hidden info checks, the hint names them ("1× Info") and offers the switch.

Sorting

Sortable by status, duration (how long the problem has existed), host, service, tenant or last check time — ascending or descending.

Saved tabs

A filter combination can be saved as a tab of its own — e.g. "Critical only at customer X" or "Network checks of all sites". A tab can be marked as default (opens automatically on the next visit), its link can be shared, and tabs from earlier filter models are migrated automatically.

Views

  • Compact: one line per entry (check, message, error code and logs side by side), as a continuous list. The button switches the personal setting Settings → Preferences → Lists → Density (Compact/Normal; default Compact) — it applies on all devices and also to the hosts list and a host's checks.
  • Group by host: all entries of a host appear as one expandable row with status counters (e.g. "1 Critical · 2 Warning"); hosts with a single entry stay on their own. The choice is saved per user.
  • TV: opens the current filter configuration in TV mode; the link carries the filters in the address.
  • Map: opens the logical map with the current filters.

During a mass outage the list shows at most 1000 entries and says so; filters (tenant, tags, check type, search) are then also applied on the server, so a filter also finds problems beyond the first 1000.

Bulk actions

When you select several rows, an action bar appears: ACK All (n), Downtime (n) and — for checks the Vesana server runs itself (Active checker) — Check now. "Select all" selects only the currently visible rows. ACK requires the permission alerts.acknowledge, downtimes downtimes.create; without it the actions are disabled. Bulk changes to interval, thresholds or info mode are available on the host detail page.

Error code chips and notes

Where a check returns a structured error code (e.g. SNMP_TIMEOUT or HTTP_STATUS_UNEXPECTED), it appears as a small chip next to the entry — with a hint on what to do in the tooltip. For event log / journald checks a link leads to the host's Logs tab, pre-filled with the level, time window and unit of the check (see Logs). If the host, tenant or check has a note, the name is highlighted and the note appears in the tooltip.