[
https://jira.jboss.org/browse/JGRP-1238?page=com.atlassian.jira.plugin.sy...
]
Bela Ban edited comment on JGRP-1238 at 9/20/10 10:48 AM:
----------------------------------------------------------
We could also implement something like RATE_LIMITER for certain events ! E.g. we cannot
send more than 5 STABLE events / 10 seconds. If we exceed this, the event in question
would simply get discarded...
This functionality could be added into RATE_LIMITER ! By default, we check against
Event.MSG, but we could also add other events into the hashmap and check against them in
down()...
was (Author: bela(a)jboss.com):
We could also implement something like RATE_LIMITER for certain events ! E.g. we
cannot send more than 5 STABLE events / 10 seconds. If we exceed this, the event in
question would simply get discarded...
STABLE: determine best value for max_bytes
------------------------------------------
Key: JGRP-1238
URL:
https://jira.jboss.org/browse/JGRP-1238
Project: JGroups
Issue Type: Feature Request
Reporter: Bela Ban
Assignee: Bela Ban
Fix For: 2.11
max_bytes should be determined as a function of the available memory, a max value that
cannot be exceeded and the number of nodes in the cluster.
If we have more cluster nodes, and everyone sends messages, then *not* changing max_value
would lead to too many STABLE messages being sent !
An example could be: max_bytes="1m" cap="0.25" // 25% of the
available heap
If we have 5 nodes in the cluster, max_bytes would be set to 5m. With a heap of 100MiB,
if we have 30 nodes, max_bytes would be set to max(30Mib, 0.25 * 100 Mib) == 25 MiB.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
https://jira.jboss.org/secure/Administrators.jspa
-
For more information on JIRA, see:
http://www.atlassian.com/software/jira