ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/AnyEvent-MP/Changes
(Generate patch)

Comparing AnyEvent-MP/Changes (file contents):
Revision 1.60 by root, Tue Sep 8 13:46:25 2009 UTC vs.
Revision 1.158 by root, Thu Aug 31 15:39:51 2017 UTC

1Revision history for AnyEvent::MP 1Revision history for AnyEvent::MP
2 2
3TODO: global protocol fix?
4TODO: use seednodes only to broadcast?
5TODO: testsuite 3TODO: testsuite
6TODO: intro: maybe simple job pool example? 4TODO: intro: maybe simple job pool example?
7TODO: intro: mention watchdog... 5TODO: intro: mention watchdog...
8TODO: group monitoring 6TODO: maybe disbale warnings by default?
7TODO: listener-scopes (10.0.0.1:4040@vpn) and connect-scopes ("vpn,public")
8TODO: document env-variable usage
9TODO: make node objects responsible for keepalive?
9 10
10TODO: cal, cal_func 11faq: can't se anything
11TODO: mon_nodes should init-call for all up nodes 12faq: all is asynch
12TODO: better autoseeding, timout*(nconnect+1) 13faq: how to interface to non-perl nodes?
13TODO: fix the "trying to contatc node under wrogn address" problem. 14
142009-09-08 03:58:42 <9> trying connect to seed node 10.0.0.18:4040. 15TODO: check gproto, nproto, on connect
152009-09-08 03:58:42 <9> 10.0.0.18:4040 frank already connected, not connecting again. 16TODO: limiting reconnecting speed when unreachable? somehow use same interval timers as for seeding and keepalive?
17TODO: multiple profiles? also some default profiles?
18TODO: export keepalive?
19TODO: $guard = con $cb->($up)
20TODO: aemp readline support
21TODO: gleeco re: AE::MP::DataConn -
22
232
24 - INCOMPATIBLE CHANGE: the node protocol used by this version is not
25 compatible to versions <2 - you have to upgrade all nodes.
26 - INCOMPATIBLE CHANGE: many APIs have changed, see the section
27 "PORTING FROM VERSION 1" in the AnyEvent::MP manpage for details.
28 - AnyEvent::MP::Global is now optional - only seed nodes autoload it
29 and create a fully-meshed network with each other, normal nodes only
30 have to connect to seed nodes, other connections are on-demand.
31 - remove md6 auth algorithms (and use algorithms based on SHA3 instead).
32 - implement CBOR framing and make it the preferred format.
33 - new "norc" and "force" options for MP::configure to ignore
34 or override the rc file.
35 - a shitload of minor and major fixes all reported and debugged
36 by Felix Ostmann.
37 - move seed code to Kernel.
38 - non-json receive framing was probably broken.
39 - fixed small typo in MP::Transport.
40 - fix (unused) provider-kv (version sometimes missing).
41 - reduce the default export list of MP::Kernel.
42 - reduced various random delays to zero, as they
43 are hopefully no longer needed with the new global
44 protocol.
45 - implement a secure mode that can suppress (some) code execution
46 requests.
47 - do not use GMP for generating alphanumeric strings even when
48 available, use a more tricky algorithm instead.
49 - anonymous node names and remote port names have been shortened
50 considerably.
51 - $AnyEvent::MP::Kernel::SRCNODE contains the node ID of the origin
52 node during message receives.
53 - aemp shell now supports "package" selection and offers an
54 $ECHO port you can send messages to.
55 - rely on perl rand() instead of using /dev/urandom when available, as
56 hopefully it is random enough.
57 - aemp shell no longer leaks one port per command.
58 - nodenames can contain %-escapes such as %n or %u.
59 - aemp now uses aemp/%n/%u as nodename.
60 - configure supports "eval", a small perl snippet to initialsie a node,
61 most useful in "aemp run eval ...".
62 - known_nodes is gone, it has little value, use all_nodes as
63 replacement.
64 - lots of small change and improvements.
65 - node_is_known has been removed.
66 - TRACE now shows the portid on SND.
67 - aggressively reduce number of external modules, to reduce
68 memory usage for nodes.
69 - rework most of the documentation.
70 - clean up MP::Kernel exports.
71 - include AnyEvent::MP in AnyEvent::MP::Kernel's @CARP_NOT to
72 hopefully improve error messages.
73
741.30 Thu Jun 30 11:30:39 CEST 2011
75 - connection errors at the right time would kill node
76 connections in a bad way (patch by Malcolm Studd).
77
781.29 Fri May 7 20:13:39 CEST 2010
79 - codename "Sadrak".
80 - fix error in callback at AnyEvent/MP/Global.pm line 339
81 (found by Sadrak).
82 - listener-less nodes had trouble sending keepalive
83 messages on write timeouts (found by Sadrak).
84 - the monitor guard could cause a memleak due
85 to autovivification if a mon was cleared after
86 the port was gone (analysed by Sadrak).
87 - do not overwrite the config file if we couldn't read it
88 for some reason (lesson demonstrated by Sadrak).
89
901.28 Thu Apr 1 21:23:54 CEST 2010
91 - accepted connections didn't correctly set up the SRCNODE
92 in some cases, leading to intra-node messages (such as monitoring)
93 to fail silently (great testcase by Sadrak).
94
951.27 Sat Mar 20 21:11:16 CET 2010
96 - avoid exceptions in the transport code if the connection
97 fails during TLS negotiation.
98 - delay new address broadcasts in seednodes, to reduce the
99 likelyhood of nodes connecting at the same time, leading to
100 message loss.
101 - improve general message passing efficiency by 8%, and special
102 case json framing for a 50% speed increase.
103 - add hints how to interface to aemp nodes from other programming
104 languages using the aemp protocol in AnyEvent::MP::Transport
105 (it can really be quite simple).
106 - handle unsupported remote framings more gracefully.
107
1081.26 Wed Dec 30 16:48:37 CET 2009
109 - INCOMPATIBLE CHANGE: the high-level node protocol changed,
110 all nodes need to upgrade.
111 - remote kil's did not work (reported by Eugene Grigoriev).
112 - documented monitor_timeout, connect_interval, auth_offer,
113 auth_accept, framing_format, autocork and nodelay options
114 and make them configurable via bin/aemp.
115 - some doc updates.
116 - remote kil's were not synchronous to message sends which affected
117 nobody, as remote kils were broken.
118 - remote mon/unmonitor requests were not synchronous to message
119 sends either sometimes, although it likely didnt matter.
120
1211.24 Wed Dec 9 14:59:52 CET 2009
122 - codename "the foul egg".
123 - increased "unmonitored local port died" message level to 2, so
124 it is displayed by default now.
125 - generate random seed at configure time, not module loading time,
126 so fork&configure works as expected.
127 - decreased "seed"-related message priority in Global from 0 (d'oh)
128 to 9, as they are not signifcant.
129 - new experimental module AnyEvent::MP::DataConn to establish
130 additional data connections between nodes.
131 - bin/aemp now loads AnyEvent::Watchdog::Util before trying
132 to call restart.
133 - bin/aemp setcert didn't properly handle the filename argument.
134 - removed empty and unused timeout= parameter from protocol greeting.
135
1361.23 Fri Nov 6 18:46:26 CET 2009
137 - listener-less nodes were misinformed by their masters about
138 the location of new nodes, and therefore didn't connect
139 properly in all cases.
140 - messages send during node-up processing could sometimes get lost,
141 which would hamper global's ability to mesh the network.
142 - fixed AnyEvent::MP::Global::grp_reg to return a guard also
143 in scalar context.
144 - fixed AnyEvent::MP::Kernel::mon_nodes to return a guard also
145 in scalar context.
146 - try to improve error reporting when automatically loading
147 a function - being unable to load a module will now
148 stop the process and report the error.
149
1501.22 Sat Oct 17 03:41:47 CEST 2009
151 - fix two rcv-bugs: after adding a tagged rcv, the default
152 callback would not be used or changed anymore (reported
153 by Felix Antonius Wilhelm Ostmann, with apologies for
154 taking so long to releae the fix).
155 - added Coro::MP introduction to AnyEvent::MP::Intro.
156 - add peval function.
157
1581.21 Mon Oct 5 21:58:36 CEST 2009
159 - always delay kil messages.
160 - remote monitor requests were never cleaned up (found by
161 Sadrak and analysed by elmex).
162 - move $VERSION to AnyEvent::MP and AnyEvent::MP only.
163
1641.2 Thu Oct 1 22:59:35 CEST 2009
165 - fix a rather big bug where message processing could be
166 inhibited completely.
167 - grp_mon could not return a guard due to a typo.
168 - implement primitive aemp shell.
169 - do not use bound address as listen length (reported by
170 Felix Antonius Wilhelm Ostman).
171 - no longer export an undefined "reg" function from AE::MP.
172 - the mon_guard return value no longer keeps an additional
173 reference to the passed refs.
174
1751.1 Fri Sep 11 04:34:03 CEST 2009
176 - bumped the transport protocol to version 1 - sorry, but there
177 were too many bugfixes/changes.
178 - new function AnyEvent::MP::cal for simple rpc.
16 - renamed AnyEvent::MP::Global functions to grp_reg|get|mon. 179 - renamed AnyEvent::MP::Global functions to grp_reg|get|mon.
180 - implemented monitoring for Global groups.
17 - removed all userspace time-outs from the transport, instead 181 - removed all userspace time-outs from the transport, instead
18 rely on tcp retransmit timeouts and tcp keepalive. 182 rely on tcp retransmit timeouts and tcp keepalive.
19 - spawn now delays spawns on the local node artificially, so 183 - spawn now delays spawns on the local node artificially, so
20 it can return before invoking the init function. 184 it can return before invoking the init function.
21 - break endless recursion between ping-pong ports on the local 185 - break endless recursion between ping-pong ports on the local
… …
28 - new service: AnyEvent::MP::LogCatcher. 192 - new service: AnyEvent::MP::LogCatcher.
29 - rely on perl srand'ing better than we could. 193 - rely on perl srand'ing better than we could.
30 - try to backport to 5.8.2, unfortunately, recursive dependencies 194 - try to backport to 5.8.2, unfortunately, recursive dependencies
31 rely on Module::Build, which fails at 5.8.7, so YMMV. 195 rely on Module::Build, which fails at 5.8.7, so YMMV.
32 - use a vastly different reseeding strategy. 196 - use a vastly different reseeding strategy.
197 - use seed nodes to notify nodes about new nodes.
198 - in the case where a ndoe learns new addresses after it tried
199 connecting, prefer the new addrseses over the old ones.
33 200
341.0 Wed Sep 2 20:40:43 CEST 2009 2011.0 Wed Sep 2 20:40:43 CEST 2009
35 - randomise more connection times, but reduce delay to 0.2s (seeds) 202 - randomise more connection times, but reduce delay to 0.2s (seeds)
36 and 0.05s (normal connects). 203 and 0.05s (normal connects).
37 - upgrade to and require AnyEvent::Watchdog 1.0. 204 - upgrade to and require AnyEvent::Watchdog 1.0.

Diff Legend

– Removed lines
+ Added lines
< Changed lines
> Changed lines