Details
-
Type:
Bug
-
Status:
Closed
-
Resolution: Won't Fix
-
Affects Version/s: 1.5 z
-
Fix Version/s: None
-
Component/s: bucket-engine
-
Labels:None
-
Environment:Operating System: All
Platform: All
Description
please also document on the wiki (see Randi)
I started down the path of adding seeding to the mc-hammer to allow it to do something closer to reproducible, but as it's a multi-threaded network client, I'm not sure how realistic it is to expect a given test to ever produce the same results twice.
That is, multi-threaded network clients that go as fast they can are inherently non-deterministic.
For reproducibility, it might be best to try to use some type of packet capture-based approach. I know there are some tools that do this kind of thing, but I don't know how well they work at load.