-
Type:
Task
-
Resolution: Done
-
Priority:
Major - P3
-
None
-
Affects Version/s: None
-
Component/s: Tools
-
None
-
Storage Engines, Storage Engines - Persistence
-
0.001
-
SE Persistence backlog
-
None
Issue Summary
We want to create observability around database size through Grafana, surfacing the clusters that consume the most storage in the disaggregated (DSC) world.
Context
- Part of the WT-16385 effort to define and expose checkpoint/database size for disagg.
- The idea is to print the top N clusters with the largest database size and/or the ones exceeding a specific value.
- In Grafana, the metric to use is mongodb_wiredTiger_disagg_database_size.
Proposed Solution
Build a Grafana panel/dashboard using one of the following queries:
# Sort all clusters
sort_desc(
sum(
mongodb_wiredTiger_disagg_database_size{rs_state="1", xgen_environment=~"dev", xgen_provider=~"aws"}
) by (xgen_environment, xgen_provider, cl_name, rs_name)
)
# Select the top N
topk(
20,
sort_desc(
sum(
mongodb_wiredTiger_disagg_database_size{rs_state="1", xgen_environment=~"dev", xgen_provider=~"aws"}
) by (xgen_environment, xgen_provider, cl_name, rs_name)
)
)
# Hard cutoff
sum(
mongodb_wiredTiger_disagg_database_size{rs_state="1", xgen_environment=~"dev", xgen_provider=~"aws"}
) by (xgen_environment, xgen_provider, cl_name, rs_name)
> 1000000000000000000
