Preliminary version. The functionality may change, but the basic features will be preserved. Compatibility with future versions is ensured, but may require additional migration actions.
Scaling is based on CodeInstance parameter scaling.targetUsercount — according to its value, module changes HPA, PDB, CPU and
memory requests/limits
Warning! Any “burstable” instance types in cloud infrastructure are not recommended due to inconsistent performance.
Architecture reference
10 Users
Single no-HA installation for demo purposes only. We do not measure load here and do not guarantee reliability / redundancy here.
Resource capacity table
All calculations are made based on scaling.targetUserCount parameter
Given table represents resources needed for single replicas of each component. For HA mode consider at least twice more capacity
| Users/Component | 100 | 300 | 500 | 1000 |
|---|---|---|---|---|
| Webservice default | 2CPU / 4Gb 2 worker, 8 thread |
3CPU / 6Gb 3 worker, 8 thread |
3CPU / 6Gb 3 worker, 8 thread |
4CPU / 8Gb 4 worker, 8 thread |
| Webservice internal | 2CPU / 4Gb 2 worker, 8 thread |
2CPU / 4Gb 2 worker, 8 thread |
2CPU / 4Gb 2 worker, 8 thread |
2CPU / 4Gb 2 worker, 8 thread |
| Sidekiq | 0.25CPU / 1.5Gb 1.5 CPU / 3Gb |
0.25CPU / 1.5Gb 1.5CPU / 3Gb |
0.25CPU / 1.5Gb 1.5CPU / 3Gb |
0.25CPU / 1.5Gb 1.5CPU / 3Gb |
| Shell | 0.01CPU / 24Mb 0.5 CPU / 600Mb |
0.01CPU / 24Mb 0.5 CPU / 600Mb |
0.01CPU / 24Mb 0.5 CPU / 600Mb |
0.01CPU / 24Mb 0.5 CPU / 600Mb |
| Toolbox | 0.05CPU / 350Mb 1 CPU / 2Gb |
0.05CPU / 350Mb 1CPU / 2Gb |
0.05CPU / 350Mb 1CPU / 2Gb |
0.05CPU / 350Mb 1CPU / 2Gb |
| Praefect | 0.1CPU / 128Mb 0.3 CPU / 600Mb |
0.1CPU / 128Mb 0.3CPU / 600Mb |
0.1CPU / 128Mb 0.3CPU / 600Mb |
0.1CPU / 128Mb 0.3CPU / 600Mb |
| Gitaly | 1.5CPU / 2Gb 1.5CPU / 2Gb |
1.5CPU / 2Gb 1.5CPU / 2Gb |
1.5CPU / 2Gb 1.5CPU / 2Gb |
2CPU / 4Gb 2CPU / 4Gb |
| MRA | 0.5CPU / 250Mb 1CPU / 500Mb |
0.5CPU / 250Mb 1CPU / 500Mb |
0.5CPU / 250Mb 1CPU / 500Mb |
0.5CPU / 250Mb 1CPU / 500Mb |
| Code-operator | 0.02CPU / 128Mb 1CPU / 256Mb |
0.02CPU / 128Mb 1CPU / 256Mb |
0.02CPU / 128Mb 1CPU / 256Mb |
0.02CPU / 128Mb 1CPU / 256Mb |
| Registry* | 1 CPU/ 1Gb 1.5CPU / 2Gb |
1CPU / 1Gb 1.5CPU / 2Gb |
1 CPU/ 1Gb 1.5CPU / 2Gb |
1CPU / 1Gb 1.5CPU / 2Gb |
| Pages* | 0.9CPU / 1Gb 1.5CPU / 2Gb |
0.9CPU / 1Gb 1.5CPU / 2Gb |
0.9CPU / 1Gb 1.5CPU / 2Gb |
0.9CPU / 1Gb 1.5CPU / 2Gb |
| Mailroom* | 0.05CPU / 150Mb 0.25CPU / 0.5Gb |
0.05CPU / 150Mb 0.25 CPU / 0.5Gb |
0.05CPU / 150Mb 0.25CPU / 0.5Gb |
0.05CPU / 150Mb 0.25CPU / 0.5Gb |
| HAProxy* | 0.25CPU / 128Mb 0.5CPU / 256Mb |
0.25CPU / 128Mb 0.5CPU / 256Mb |
0.25CPU / 128Mb 0.5CPU / 256Mb |
0.25CPU / 128Mb 0.5CPU / 256Mb |
| Total(Min components) | 7CPU / 13Gb 11CPU / 17Gb |
8CPU / 15.5Gb 12CPU / 19.5Gb |
8CPU / 15.5Gb 12CPU / 19.5Gb |
9.5CPU / 17.5Gb 13CPU / 21.5Gb |
| Total(All components) | 9CPU / 15Gb 14CPU / 22Gb |
10CPU / 17.5Gb 15CPU / 24Gb |
10CPU / 17.5Gb 15CPU / 24Gb |
11.5CPU / 19.5Gb 16CPU / 26Gb |
Optional components are marked with
*The first line in each cell stands for kubernetesresources, the second - forlimits
Tunning Git storage
If you see a non-empty ‘OOM events’ table in Grafana or a firing GitalyCgroupMemoryOOM alert in Prometheus, you likely need to adjust the memory and CPU resources for Gitaly. Increase the instanceSpec.gitData.resources in your configuration (e.g., set memory to 16Gi and CPU to 4). After updating, apply the changes and monitor Gitaly for improvements.