Skip to content

Using JGroups on a crappy network for reliable comms #862

Answered by belaban
snyderra asked this question in Q&A
Discussion options

You must be logged in to vote

Hi Rob

it depends... what transport are you using? UDP would probably be recommended over TCP, as the SYN-ACK-SYNACK connection establishment costs you dearly. Even worse when using TLS.

+1 on increasing the GMS.join_timeout and FD_SOCK2.connect_timeout timeouts. You might even consider removing FD_SOCK2 altogether and only use FD_ALL3

If you use RPCs, make sure you set the call timeouts accordingly. Perhaps also use async RPCs in this case.

Is the latency fluctuating, or is it always 5s?
Also note that you could use DELAY to simulate this environment in your test env, if test env != prod env.

Replies: 1 comment 2 replies

Comment options

You must be logged in to vote
2 replies
@snyderra
Comment options

@belaban
Comment options

Answer selected by snyderra
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Category
Q&A
Labels
None yet
2 participants