| 1 | Revision history for AnyEvent::MP |
1 | Revision history for AnyEvent::MP |
| 2 | |
2 | |
| 3 | TODO: global protocol fix? |
|
|
| 4 | TODO: use seednodes only to broadcast? |
|
|
| 5 | TODO: testsuite |
3 | TODO: testsuite |
| 6 | TODO: intro: maybe simple job pool example? |
4 | TODO: intro: maybe simple job pool example? |
| 7 | TODO: intro: mention watchdog... |
5 | TODO: intro: mention watchdog... |
| 8 | TODO: group monitoring |
6 | TODO: use gvpe method of making contact, if possible. |
| 9 | |
7 | |
| 10 | TODO: cal, cal_func |
8 | 1.21 Mon Oct 5 21:58:36 CEST 2009 |
| 11 | TODO: fix the "trying to contatc node under wrogn address" problem. |
9 | - always delay kil messages. |
|
|
10 | - remote monitor requests were never cleaned up (found by |
|
|
11 | Sadrak and analysed by elmex). |
|
|
12 | - move $VERSION to AnyEvent::MP and AnyEvent::MP only. |
|
|
13 | |
|
|
14 | 1.2 Thu Oct 1 22:59:35 CEST 2009 |
|
|
15 | - fix a rather big bug where message processing could be |
|
|
16 | inhibited completely. |
|
|
17 | - grp_mon could not return a guard due to a typo. |
|
|
18 | - implement primitive aemp shell. |
|
|
19 | - do not use bound address as listen length (reported by |
|
|
20 | Felix Antonius Wilhelm Ostman). |
|
|
21 | - no longer export an undefined "reg" function from AE::MP. |
|
|
22 | - the mon_guard return value no longer keeps an additional |
|
|
23 | reference to the passed refs. |
|
|
24 | |
|
|
25 | 1.1 Fri Sep 11 04:34:03 CEST 2009 |
|
|
26 | - bumped the transport protocol to version 1 - sorry, but there |
|
|
27 | were too many bugfixes/changes. |
|
|
28 | - new function AnyEvent::MP::cal for simple rpc. |
| 12 | - renamed AnyEvent::MP::Global functions to grp_reg|get|mon. |
29 | - renamed AnyEvent::MP::Global functions to grp_reg|get|mon. |
|
|
30 | - implemented monitoring for Global groups. |
| 13 | - removed all userspace time-outs from the transport, instead |
31 | - removed all userspace time-outs from the transport, instead |
| 14 | rely on tcp retransmit timeouts and tcp keepalive. |
32 | rely on tcp retransmit timeouts and tcp keepalive. |
| 15 | - spawn now delays spawns on the local node artificially, so |
33 | - spawn now delays spawns on the local node artificially, so |
| 16 | it can return before invoking the init function. |
34 | it can return before invoking the init function. |
| 17 | - break endless recursion between ping-pong ports on the local |
35 | - break endless recursion between ping-pong ports on the local |
| … | |
… | |
| 25 | - rely on perl srand'ing better than we could. |
43 | - rely on perl srand'ing better than we could. |
| 26 | - try to backport to 5.8.2, unfortunately, recursive dependencies |
44 | - try to backport to 5.8.2, unfortunately, recursive dependencies |
| 27 | rely on Module::Build, which fails at 5.8.7, so YMMV. |
45 | rely on Module::Build, which fails at 5.8.7, so YMMV. |
| 28 | - use a vastly different reseeding strategy. |
46 | - use a vastly different reseeding strategy. |
| 29 | - use seed nodes to notify nodes about new nodes. |
47 | - use seed nodes to notify nodes about new nodes. |
|
|
48 | - in the case where a ndoe learns new addresses after it tried |
|
|
49 | connecting, prefer the new addrseses over the old ones. |
| 30 | |
50 | |
| 31 | 1.0 Wed Sep 2 20:40:43 CEST 2009 |
51 | 1.0 Wed Sep 2 20:40:43 CEST 2009 |
| 32 | - randomise more connection times, but reduce delay to 0.2s (seeds) |
52 | - randomise more connection times, but reduce delay to 0.2s (seeds) |
| 33 | and 0.05s (normal connects). |
53 | and 0.05s (normal connects). |
| 34 | - upgrade to and require AnyEvent::Watchdog 1.0. |
54 | - upgrade to and require AnyEvent::Watchdog 1.0. |
