Breakpad/Status Meetings/2010-Dec-29: Difference between revisions
< Breakpad | Status Meetings
Jump to navigation
Jump to search
(load testing) |
(notes) |
||
| Line 1: | Line 1: | ||
= Load Testing = | = Load Testing = | ||
* Grinder seems like overkill for the initial smoketest | * Grinder seems like overkill for the initial smoketest | ||
** we could add a few features to submitter | ** we could add a few features to submitter, to simulate peak traffic (SWAG - 10k crashes-per-second?) | ||
* IT can't get us VMs until next week | * IT can't get us VMs until next week | ||
* make sure the network path from test machines -> server is realistic | * make sure the network path from test machines -> server is realistic | ||
* make sure that client VMs can consistently produce enough load | |||
** we should be able to measure this on the server, to make sure we're getting the correct number of crashes-per-second | |||
= PHX status = | = PHX status = | ||
Latest revision as of 19:51, 29 December 2010
Load Testing
- Grinder seems like overkill for the initial smoketest
- we could add a few features to submitter, to simulate peak traffic (SWAG - 10k crashes-per-second?)
- IT can't get us VMs until next week
- make sure the network path from test machines -> server is realistic
- make sure that client VMs can consistently produce enough load
- we should be able to measure this on the server, to make sure we're getting the correct number of crashes-per-second
PHX status
- not ready
- need Zeus configuration for postgresql
Action items
- lars to add stats and looping to submitter
- rhelmer to follow up on phx status