Skip to content

Commit 8e05c4d

Browse files
Merge branch 'master' into rabbitmq-server-1303
2 parents 1f52406 + 2b2ea11 commit 8e05c4d

File tree

3 files changed

+96
-2203
lines changed

3 files changed

+96
-2203
lines changed

docs/rabbitmq.conf.example

Lines changed: 11 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -252,8 +252,18 @@
252252

253253
## Fraction of the high watermark limit at which queues start to
254254
## page message out to disc in order to free up memory.
255+
## For example, when vm_memory_high_watermark is set to 0.4 and this value is set to 0.5,
256+
## paging can begin as early as when 20% of total available RAM is used by the node.
255257
##
256-
## Values greater than 0.9 can be dangerous and should be used carefully.
258+
## Values greater than 1.0 can be dangerous and should be used carefully.
259+
##
260+
## One alternative to this is to use durable queues and publish messages
261+
## as persistent (delivery mode = 2). With this combination queues will
262+
## move messages to disk much more rapidly.
263+
##
264+
## Another alternative is to configure queues to page all messages (both
265+
## persistent and transient) to disk as quickly
266+
## as possible, see http://www.rabbitmq.com/lazy-queues.html.
257267
##
258268
# vm_memory_high_watermark_paging_ratio = 0.5
259269

docs/rabbitmq.config.example

Lines changed: 11 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -236,8 +236,18 @@
236236

237237
%% Fraction of the high watermark limit at which queues start to
238238
%% page message out to disc in order to free up memory.
239+
%% For example, when vm_memory_high_watermark is set to 0.4 and this value is set to 0.5,
240+
%% paging can begin as early as when 20% of total available RAM is used by the node.
239241
%%
240-
%% Values greater than 0.9 can be dangerous and should be used carefully.
242+
%% Values greater than 1.0 can be dangerous and should be used carefully.
243+
%%
244+
%% One alternative to this is to use durable queues and publish messages
245+
%% as persistent (delivery mode = 2). With this combination queues will
246+
%% move messages to disk much more rapidly.
247+
%%
248+
%% Another alternative is to configure queues to page all messages (both
249+
%% persistent and transient) to disk as quickly
250+
%% as possible, see http://www.rabbitmq.com/lazy-queues.html.
241251
%%
242252
%% {vm_memory_high_watermark_paging_ratio, 0.5},
243253

0 commit comments

Comments
 (0)