Ticket #147 (new defect)
--seed is fully random on some platforms
| Reported by: | aturner | Owned by: | aturner |
|---|---|---|---|
| Priority: | low | Milestone: | Future Release |
| Component: | libtcpedit | Version: | 3.0.beta13 |
| Keywords: | Cc: | ||
| Operating System: | Add to FAQ?: | yes | |
| Hardware: | All | ||
| Output of tcpreplay -V: | |||
Description
currently --seed is just used for srandom() to seed the random generator. On most systems, random() uses a single well defined algorithm, but a few systems (like Win32) use another algorithm. The result is the --seed test fails and the results aren't determnistic.
Honestly, it's probably not that big of a deal, but it is something I would like to see.
Change History
comment:5 in reply to: ↑ description Changed 2 years ago by anonymous
Replying to aturner:
currently --seed is just used for srandom() to seed the random generator. On most systems, random() uses a single well defined algorithm, but a few systems (like Win32) use another algorithm. The result is the --seed test fails and the results aren't determnistic.
Honestly, it's probably not that big of a deal, but it is something I would like to see.
