forked from rethinkdb/rethinkdb
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathNOTES
161 lines (100 loc) · 3.47 KB
/
NOTES
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
# Release 1.4.3 (Some Like It Hot) #
## Changes ##
Bug fix update:
* Improve the networking code in the Python driver
* Fix a crash triggered by a type error when using concatMap (#568)
* Fix a crash when running `rethinkdb proxy --help` (#565)
* Fix a bug in the Python driver that caused it to occasionally return `None` (#564)
# Release 1.4.2 (Some Like It Hot) #
## Changes ##
Bug fix update:
* Replace `~` with `About` in the web UI (#485)
* Add framing documentation to the protobuf spec (#500)
* Fix crashes triggered by .orderBy().skip() and .reduce(r.js()) (#522, #545)
* Replace MB with GB in an error message (#526)
* Remove some semicolons from the protobuf spec (#530)
* Fix the `rethinkdb import` command (#535)
* Improve handling of very large queries in the data explorer (#536)
* Fix variable shadowing in the javascript driver (#546)
---
# Release 1.4.1 (Some Like It Hot) #
## Changes ##
Bug fix update:
* Python driver fix for TCP streams (#495)
* Web UI fix that reduces the number of AJAX requests (#481)
* JS driver: added useOutdated to r.table() (#502)
* RDB protocol performance fix in release mode.
* Performance fix when using filter with object shortcut syntax.
* Do not abort when the `runuser` or `rungroup` options are present (#512)
---
# Release 1.4 (Some Like It Hot) #
## Changes ##
* Improved ReQL wire protocol and client drivers
* New build system
* Data explorer query history
---
# Release 1.3.2 (Metropolis) #
## Changes ##
* Fixed security bug in http server.
---
# Release 1.3.1 (Metropolis) #
## Changes ##
* Fixed OS X crash on ReQL exceptions.
---
# Release 1.3.0 (Metropolis) #
## Changes ##
* Native OS X support.
* 32-bit support.
* Support for legacy systems (e.g. Ubuntu 10.04)
---
# Release 1.2.8 (Rashomon) #
## Changes ##
* Updating data explorer suggestions to account for recent `r.row`
changes.
---
# Release 1.2.7 (Rashomon) #
## Changes ##
* Lots and lots of bug fixes
---
# Release 1.2.6 (Rashomon) #
## Changes ##
* Fixed the version string
* Fixed 'crashing after crashed'
* Fixed json docs
---
# Release 1.2.5 (Rashomon) #
## Changes ##
* Checking for a null ifaattrs
---
# Release 1.2.4 (Rashomon) #
## Changes ##
* Local interface lookup is now more robust.
---
# Release 1.2.3 (Rashomon) #
## Changes ##
* Fixes a bug in the query engine that causes large range queries to
return incorrect results.
---
# Release 1.2.0 (Rashomon) #
## Changes ##
This is the first release of the product. It includes:
* JSON data model and immediate consistency support
* Distributed joins, subqueries, aggregation, atomic updates
* Hadoop-style map/reduce
* Friendly web and command-line administration tools
* Takes care of machine failures and network interrupts
* Multi-datacenter replication and failover
* Sharding and replication to multiple nodes
* Queries are automatically parallelized and distributed
* Lock-free operation via MVCC concurrency
## Limitations ##
There are a number of technical limitations that aren't baked into the
architecture, but were not resolved in this release due to time
pressure. They will be resolved in subsequent releases.
* Write requests have minimal batching in memory and are flushed to
disk on every write. This significantly limits write performance,
expecially on rotational disks.
* Range commands currently don't use an index.
* The clustering system has a bottleneck in cluster metadata
propagation. Cluster management slows down significantly when more
than sixteen shards are used.