01251d6d91
merge bug19730 into default
2008-11-10 11:25:39 +00:00
bba6ec51b3
Merged bug18381 into bug19684
2008-11-09 23:32:13 +00:00
26cd77cc37
First cut at handling channel.flow in the Erlang client
2008-11-09 22:19:18 +00:00
ca3b1520af
Merged default into bug19331, QA feedback
2008-11-08 14:55:03 +00:00
f907f49027
merge bug19250 into default
2008-11-07 17:45:07 +00:00
65e270e7b2
merge bug19250 into default
2008-11-07 17:44:40 +00:00
6a64d36c9d
Cleaned the Makefile. Goodbye generic_stage!
2008-11-07 17:42:54 +00:00
92a954f052
Revert to 0-9 definition.
2008-11-07 17:19:35 +00:00
146d4a1eca
Revert to 0-9 definition.
2008-11-07 17:18:58 +00:00
f821572634
Do not include source files in the binary packages
2008-11-07 16:26:59 +00:00
c83165eca1
Verified
2008-11-07 11:25:23 +00:00
252620e5a0
info_all
...
this is resilient to disappearing queues
2008-11-07 10:36:41 +00:00
74eaf36721
beginnings of queue info API
2008-11-06 23:07:39 +00:00
d367479bba
merge bug18776 into default
2008-11-06 17:51:07 +00:00
a7cd1578cd
Merge version 2 of 19468 back into default
2008-11-05 12:02:54 +00:00
6e887e5ac1
fix signature
...
mfa() is predefined as {atom(), atom(), byte()}
2008-11-04 20:53:32 +00:00
fb15adb9e4
Merged 19623 into default
2008-11-04 16:51:09 +00:00
0861a3a965
Merged 19587 into default
2008-11-04 15:08:11 +00:00
6a6665ccd5
Merged default into 18776
2008-11-03 15:37:23 +00:00
18f495420f
Merged 19468 into default
2008-11-03 14:58:58 +00:00
282b258378
Merge default into 19468
2008-11-03 12:11:34 +00:00
19b14e5e60
Alter signatures to the latest 0-91 draft XML
2008-10-31 11:37:12 +00:00
771097544a
Alter signatures to the latest 0-91 draft XML, and support nowait
2008-10-31 11:36:49 +00:00
692f208182
merge default into bug19250
2008-10-31 11:07:57 +00:00
8d9ec7e278
merge default into bug19250
2008-10-31 11:07:23 +00:00
a39933ca41
Merge two default branches into single one
2008-10-30 17:04:20 +00:00
eae657af80
Merge two default branches into one
2008-10-30 17:00:08 +00:00
88e870720b
merge bug19250 into default
2008-10-30 14:10:55 +00:00
44295f44d7
merge bug19250 into default
2008-10-30 14:10:55 +00:00
6a5a9f667b
merged default into bug19250
2008-10-30 13:53:54 +00:00
6c1106d5ca
Added close handler to network driver
2008-10-27 16:15:19 +00:00
2076ae22ba
Started work on this bug, might need to add callback handlers
...
in the driver modules to propagted this properly.
2008-10-27 12:00:55 +00:00
16c1b59c40
don't complain on shutdown
...
when sending notify_down et al to queues
2008-10-25 23:44:52 +01:00
bffa8f6d0d
log *actual* listening IP & port
...
This caters for o/s-assigned listening ports, i.e. when the given port
is 0.
2008-10-25 22:09:41 +01:00
50a0752fb1
Merge bug18732 into default
2008-10-25 13:26:50 +01:00
31cb538ba3
Added new line to logrotate scripts, debian packaging no longer throws warning
2008-10-25 13:05:55 +01:00
f3e9579037
merge bug19572 into default
2008-10-25 11:33:45 +01:00
153837a2ca
Removed packages that are already set as dependencies of the others
2008-10-25 11:09:24 +01:00
7700cc6e4b
hibernate writers
...
With this in place I am finally unable to make rabbit grind to a halt
due to garbage being held by idle processes.
2008-10-25 06:13:06 +01:00
4b2330d4c5
Minimal version of build dependencies on debian
2008-10-24 18:32:55 +01:00
d7343c5d7f
hibernate buffering_proxies
...
My tests got stuck after about an hour, and the cause was
buffering_proxies holding on to memory
2008-10-24 17:46:50 +01:00
051da30ab8
Merge bug19552 into default
2008-10-24 11:06:10 +01:00
feb1e47a9c
Merge bug19552 into default
2008-10-24 10:49:15 +01:00
24558d7f68
Merge bug19382 into default
2008-10-23 16:11:32 +01:00
b19e4bb800
merge bug19468_channels into bug19468
2008-10-22 07:51:58 +01:00
25a63c3e34
hibernate some processes to conserve memory
...
In my experiments I encountered situations where rabbit would not
recover from a high memory alert even though all messages had been
drained from it. By inspecting the running processes I determined that
queue and channel processes sometimes hung on to garbage. Erlang's gc
is per-process and triggered by process reduction counts, which means
an idle process will never perform a gc. This explains the behaviour -
the publisher channel goes idle when channel flow control is activated
and the queue process goes idle once all messages have been drained
from it.
Hibernating idle processes forces a gc, as well as generally reducing
memory consumption. Currently only channel and queue processes are
hibernating, since these are the only two that seemed to be causing
problems in my tests. We may want to extend hibernation to other
processes in the future.
2008-10-21 15:59:41 +01:00
bcd4affd32
raise memory alarm threshold
...
The default 80% is just too low for many systems - I have less than
that on tanto most of the time.
It remains to be seen whether the new figure works ok for most users.
2008-10-21 15:46:05 +01:00
96af7bf702
Removed check_tools script since Build-Depends
...
will handle it itself.
2008-10-19 22:52:00 +01:00
5b995a2b87
Forgot about perl
2008-10-19 02:19:38 +01:00
743944c0e7
Added missing build dependencies for debian
2008-10-19 02:13:47 +01:00
Tony Garnock-Jones
Emile Joubert
Ben Hood
Matthias Radestock
Hubert Plociniczak
Simon MacMullen