A single point-in-time snapshot of server-side compute grid metrics. More...
#include <grid_sample.hpp>
Collaboration diagram for grid_sample:
Public Attributes | |
| std::chrono::system_clock::time_point | sampled_at |
| When this sample was taken (hypertable partition key). | |
| utility::uuid::tenant_id | tenant_id = utility::uuid::tenant_id::system() |
| Tenant this sample belongs to. | |
| int | total_hosts {0} |
| Total registered hosts (all versions). | |
| int | online_hosts {0} |
| Hosts whose last heartbeat is within the 5-minute reaper window. | |
| int | idle_hosts {0} |
| Online hosts that have no InProgress result assigned to them. | |
| int | results_inactive {0} |
| Results in state 1 (Inactive). | |
| int | results_unsent {0} |
| Results in state 2 (Unsent). | |
| int | results_in_progress {0} |
| Results in state 4 (InProgress). | |
| int | results_done {0} |
| Results in state 5 (Done). | |
| int | total_workunits {0} |
| Total workunits. | |
| int | total_batches {0} |
| Total batches. | |
| int | active_batches {0} |
| Batches that have at least one InProgress result. | |
| int | outcomes_success {0} |
| int | outcomes_client_error {0} |
| int | outcomes_no_reply {0} |
Detailed Description
A single point-in-time snapshot of server-side compute grid metrics.
Written by the compute_grid_poller in ores.compute.service every ~30 seconds. One row per sample interval per tenant in ores_compute_grid_samples_tbl.
