|
Mantis - Resin
|
|||||
| Viewing Issue Advanced Details | |||||
|
|
|||||
| ID: | Category: | Severity: | Reproducibility: | Date Submitted: | Last Update: |
| 1780 | minor | always | 06-06-07 10:36 | 10-05-07 19:22 | |
|
|
|||||
| Reporter: | sam | Platform: | |||
| Assigned To: | ferg | OS: | |||
| Priority: | normal | OS Version: | |||
| Status: | closed | Product Version: | 3.1.1 | ||
| Product Build: | Resolution: | unable to reproduce | |||
| Projection: | none | ||||
| ETA: | none | Fixed in Version: | 3.1.3 | ||
|
|
|||||
| Summary: | 0001780: Performance slowdown with mod_caucho when 2nd backend is introduced | ||||
| Description: |
(rep by M Morawetz) I'm evaluating the resin-cluster capabilities. My problem is that with the current setup using just one node performs much better than having two nodes running. The setup is: Two apache server as frontend, I use the mod_caucho module as load-balancer, two resin server. One Apache and one resin instance is running on one physical server. I tested the performance with Jmeter. Throughput with one resin node running is about the double of the throughput when I use both resin nodes (same load-Jmeter setup / 80 Users). I was expecting an increased throughput having 2 nodes running. Could that be a configuration issue? What are common reasons for that behaviour? Any ideas? |
||||
| Steps To Reproduce: | |||||
| Additional Information: | |||||
| Relationships | |||||
| Attached Files: | |||||
| Notes | |||||
|
|
|||||
|
|
||||