Skip to content
Go to dashboard

13.6Filtering and pages of results

List calls return results in pages (a part of the list). Send limit to choose the page size.

  1. Send the call with a limit.
  2. If the answer has nextCursor, more results are waiting. A cursor is a marker that says where the next page starts.
  3. Send the call again with the same fields, plus "cursor" set to that nextCursor value.
  4. Stop when nextCursor is missing or empty. That was the last page.
CallDefaultMaxFiltersOrder
ListMonitors50200NoneNewest first
ListAlertChannels50200None
ListExecutions (check results)501,000monitorId (required); statusFilter: CHECK_STATUS_UP, CHECK_STATUS_DOWN or CHECK_STATUS_DEGRADED (empty = all)Newest first
ListIncidents50200status: "firing" or "resolved" (empty = all)Newest first
ListNotifications50200monitorId (optional)
ListEvaluations2001,000monitorId (required); windowSeconds (default 24 hours). No pages.Newest first

ListApiKeys, ListSecrets and ListMaintenanceWindows have no pages. They return everything. Maintenance windows come soonest first.

There is no search by name, type or tag, no sort option and no total count. To find a monitor, page through ListMonitors.

Page through failed checks
# First page: failed checks, 100 at a time
curl -s "$API/result/obsrvly.v1.ResultService/ListExecutions" \
  -H "Authorization: Bearer $OBSRVLY_KEY" \
  -H "Content-Type: application/json" \
  -d '{"monitorId": "<monitor-id>", "limit": 100,
       "statusFilter": "CHECK_STATUS_DOWN"}'

# Next page: add the nextCursor from the answer
curl -s "$API/result/obsrvly.v1.ResultService/ListExecutions" \
  -H "Authorization: Bearer $OBSRVLY_KEY" \
  -H "Content-Type: application/json" \
  -d '{"monitorId": "<monitor-id>", "limit": 100,
       "statusFilter": "CHECK_STATUS_DOWN",
       "cursor": "<nextCursor>"}'

Uptime and response-time summaries#

  • ListMonitorSummaries takes monitorIds (at least 1, at most 1,000) and a window.
  • window can be SUMMARY_WINDOW_1H, SUMMARY_WINDOW_24H, SUMMARY_WINDOW_7D or SUMMARY_WINDOW_30D. If you leave it out, you get 24 hours.
  • GetMonitorSeries returns points over time. You can set bucketSeconds. You get at most 1,500 points. A bucket that is too small is made larger.
  • uptimePct is -1 when nothing was measured. Show it as "no data", not 0%.
  • overdue: true means there is no recent result. Do not show lastStatus as the current status.

Last updated