The Accounting Report shows the App Statistics dashboard, but the Service Usage dashboard is missing.
When calling the /system_report/service_usages endpoint, the app-usage-server responds with a 502 error code.
$ curl -k -H "Authorization: $(cf oauth-token)" https://app-usage.system.example.com/system_report/service_usages -i -vvv
* Trying 10.x.x.x...
* TCP_NODELAY set
* Connected to app-usage.system.example.com (10.x.x.x) port 443 (#0)
...
> GET /system_report/service_usages HTTP/2
> Host: app-usage.system.example.com
> User-Agent: curl/7.61.1
...
< HTTP/2 502
HTTP/2 502
...
< content-length: 67
content-length: 67
<
502 Bad Gateway: Registered endpoint failed to handle the request.
While calling the above endpoint, the memory usage of the app instance goes up significantly to almost 1GB:
$ cf app app-usage-server
Showing health and status for app app-usage-server in org system / space system as admin...
name: app-usage-server
requested state: started
routes: app-usage.system.example.com
last uploaded: Mon 15 Dec 12:43:21 UTC 2025
stack: cflinuxfs4
buildpacks:
name version detect output buildpack name
ruby_buildpack 1.10.31 ruby ruby
type: web
sidecars:
instances: 2/2
memory usage: 1024M
state since cpu memory disk logging cpu entitlement details
#0 running 2025-12-15T12:44:52Z 21.8% 141.8M of 1G 237.7M of 1G 28B/s of 16K/s 82.7%
#1 running 2025-12-17T04:19:51Z 70.2% 998.6M of 1G 237.7M of 1G 40B/s of 16K/s 265.9%
TPCF v6.0.17
The service_instance_usages table grew to a point where it has 130K rows. This volume causes the high memory usage by the app-usage-server when it tries to generate the report.
Scale up the memory of the app-usage-server to 2GB.
cf scale app-usage-server -m 2G
Some enhancements will be included in a future release. There are optimizations developed that should significantly reduce the memory consumption. The app-usage-server memory setting would also be user-configurable.