forked from johann8384/memcachedb
-
Notifications
You must be signed in to change notification settings - Fork 0
/
ChangeLog
246 lines (185 loc) · 10.7 KB
/
ChangeLog
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
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
2009-04-15 Steve Chu <[email protected]>
* Remove stats malloc and stats maps due to a security issue. Thanks <[email protected]> for reporting this.
2009-03-10 Steve Chu <[email protected]>
* new option '-G' for setting log file dir(sometime we want to put log files in a different place, such as SSD disk partition)
2009-03-05 Steve Chu <[email protected]>
* Fix some warnings on 64-bit compiler, thanks to Nicolas<[email protected]>
2009-02-26 Steve Chu <[email protected]>
* Bugfix: 64bit box can not specify memory pool size larger than 2G with '-m' option
* new option '-E' to automatically remove log files that are no longer needed, but PLEASE NOTICE:
- "Automatic log file removal is likely to make catastrophic recovery impossible."
- "Replication applications will rarely want to configure automatic log file removal as it increases the likelihood a master will be unable to satisfy a client's request for a recent log record." So this option will be disabled when replication is on regardless of whether you specify it.
* Bugfix: slight security issue when client specify a negative value length
2009-01-13 Steve Chu <[email protected]>
* some default settings policy changed(because most of us is using powerful box):
max simultaneous connections ('-c'): 1024 -> 4096
item buffer size('-b'): 512 bytes -> 2048 bytes
in-memmory cache size of BerkeleyDB('-m'): 64MB -> 256MB
log buffer size('-L'): 32KB -> 4MB
2009-01-06 Steve Chu <[email protected]>
* new 'rget' command patch for 'memcache.py'
2008-12-25 Version 1.2.1-beta released
2008-12-10 Steve Chu <[email protected]>
* new option '-X' to make the BerkeleyDB allocate region memory from the heap instead of from memory backed by the filesystem or system shared memory. If this option enabled, you can not use db_* standalone utilities(use private commands instead), but get performance improved.
2008-11-27 Steve Chu <[email protected]>
* new flexible query command 'rget' implemented, can do range query based
on keys. See 'doc/rget.txt' for details.
2008-11-26 Steve Chu <[email protected]>
* more graceful exit
2008-11-24 Version 1.2.0 released
2008-10-16 Version 1.2.0-beta released
2008-10-14 Steve Chu <[email protected]>
* a new unittest suit 'mdbtest.py' is added
* a new benchmark suit 'mcben.py' is added, now you can get the benchmark in your environment
and find out how fast MemcacheDB can run.
2008-10-08 Steve Chu <[email protected]>
* lots of statistics code clean up; stats command adjusted for better rendering; now more replication statistics info is provided.
* more compact and verbose message for logging
* lots of replication code clean up for more reliable electing, a new option "-n" is added that allows user to customize the votes of electing.
* "memcache.py" is nearly patched for recent changes on private commands.
2008-09-04 Steve Chu <[email protected]>
* New option '-e' to set percent of the pages in the cache that should be clean, used for memp_trickle call(only available when '-T' is presented).
2008-09-03 Steve Chu <[email protected]>
* Replace the database type setting option '-T' with '-B'
* Now '-T' has new meaning that it fires a separate thread to do memp_trickle every xx seconds.
This feature makes the memory pool more effective by reducing too much frequent pages swapping.
2008-09-01 Steve Chu <[email protected]>
* Now 'append' and 'prepend' commands are supported.
* Merge bugfix code from Memecached community.
* lots of refactors for better implementation and a bit code tuning.
2008-08-28 Version 1.1.0 released
2008-08-28 Version 1.0.4 released(maitaining release)
2008-07-25 Steve Chu <[email protected]>
* Warning: 'pkget' and 'pvget' private command now has been discarded.
MemcacheDB will be a more memcache protocol compatible daemon(append and
prepend command will be implemented later), and will not add relational
data model any more, the structural data storage will be issued by another
project (now internal only).
2008-07-25 Steve Chu <[email protected]>
* Bugfix: "multiple get" did not return the right value
2008-07-23 Steve Chu <[email protected]>
* add more other replication configure parameters
2008-07-14 Steve Chu <[email protected]>
* rename "default.db" to "data.db" for less ambiguity
2008-07-04 Steve Chu <[email protected]>
* work well with Mac OS X(tested on Lepard 10.5.4)
* new 'db_compact' private command to compact a btree database
2008-07-03 Version 1.1.0-beta released
2008-06-26 Steve Chu <[email protected]>
* unix socket connectivity now has be well tested.
2008-06-13 Steve Chu <[email protected]>
* some code refactoring to improve a bit performance
2008-06-11 Steve Chu <[email protected]>
* now '-b' option is a tuning way, not a limit. item size that smaller than
'-b' value use fast freelist alloc, otherwise, use system 'malloc' instead.
Many thanks to Davies Liu <[email protected]> for patches and ideas.
2008-06-03 Steve Chu <[email protected]>
* new support to BerkeleyDB 4.7
- new version is BerkeleyDB 4.7 only, due to the changed bdb api
- 'rep_set_request' now use time, not the number of messages
* Bugfix: 'pkget' and "multiple get" that return lots of items may
cause memory overflow so the daemon crashes. Thanks to Davies Liu <[email protected]>
* flooded verbose messages of bdb and replication now move to '-vv' option
2008-04-13 Steve Chu <[email protected]>
* Changes:
- allow limitation to 'pkget' and 'pvget' command,
use 'pkget/pvget <prefix> [limit]' to get limited matched items.
- 'memcache.py' is also patched for this feature.
2008-04-03 Steve Chu <[email protected]>
* New features:
- add '-T' option to set database type, 'btree' or 'hash'
- add '-E' option to enable second database, so we can use 'pvget'
- new command 'pkget' that you can get items with a key prefix
- new command 'pvget' that you can get items with a value prefix
* Changes:
- add support to memcache.py for 'pkget' and 'pvget' command, see
'tools/memcache.py'
2008-03-21 Version 1.0.3 released
2008-03-21 Steve Chu <[email protected]>
* Bugfix: --with-bdb can not find libdb.so and db.h
2008-03-17 Steve Chu <[email protected]>
* add a '-A' option to set page size, and also can get this value from
'stats bdb'
2008-03-07 Steve Chu <[email protected]>
* new private command 'rep_set_request' available, to set wehther it allows bulk
transfer and also can get this value from 'stats rep'. See 'doc/replication.txt'
* mdbtop.py update
2008-03-06 Steve Chu <[email protected]>
* new private command 'rep_set_bulk' available, to set wehther it allows bulk
transfer and also can get this value from 'stats rep'. See 'doc/replication.txt'
2008-03-05 Steve Chu <[email protected]>
* add 'next_lsn' to 'stats rep' to show master-slave sync status
* Make the default ack policy 'DB_REPMGR_ACKS_ONE_PEER'
* upgrade 'mdbtop.py' for better rendering
2008-03-04 Version 1.0.2-beta released
2008-03-03 Steve Chu <[email protected]>
* replication configure changed and improved:
- Make the default ack policy 'DB_REPMGR_ACKS_ALL_PEERS'
- Make the default ack timeout 20 millisecond
- two new private commands 'rep_set_ack_policy' and
'rep_set_ack_timeout' are available to set ack_policy and ack_timeout,
and 'stats rep' also includes this two value. See
'docs/replication.txt'
2008-02-29 Steve Chu <[email protected]>
* give a warning if item buffer size larger than 256KB
* memcache.py is patched for private commands. See 'tools/memcache.py'
2008-02-26 Steve Chu <[email protected]>
* Bugfix: incr/decr is not atomic in multithread, porting mistake from memcached
2008-02-21 Version 1.0.1-beta released
2008-02-21 Steve Chu <[email protected]>
* Bugfix: some private commands may cause seg fault
2008-02-20 Steve Chu <[email protected]>
* Bugfix: incr/decr command did not update length string
2008-02-18 Steve Chu <[email protected]>
* merge all read-only private command of replication into 'stats rep' that memcached client
can read it. Original commands are still available. 'stats bdb' is also provided.
* 'mdbtop.py', a new monitor tool is added for easy maintaining. See 'tools/mdbtop.py'
2008-02-14 Big Version 1.0.0-beta released
2008-02-01 Steve Chu <[email protected]>
* Almost entire code is rewritten based on Memcached 1.2.x.
* More memcache protocol compatibility, now multiple get and flags are supported,
also udp and unix socket is ready(not yet tested).
* big performance improved due to memcached 1.2.x code base. Thread is
used to resolve the blocked I/O.
* use standard build tool, "./configure;make;make install" and done.
* two new pravite command for replication to set/get priority. See
"doc/replication.txt".
2008-01-21 Version 0.1.1 released
2008-01-21 Steve Chu <[email protected]>
* Bug fix: out string 'NOT STORED' breaks clients, it should be
'NOT_STORED'!
2007-11-05 Version 0.1.0 released
2007-10-30 Cao Kai <[email protected]>, Steve Chu <[email protected]>
* big replication code merged, let's ship it!
using option "-R" to enable replication, two private command is added for
replication: db_ismaster, db_whoismaster
* code refactor for performance and readability
2007-10-23 Version 0.0.4 released
2007-10-22 Steve Chu <[email protected]>
* add a new struct db_settings for all db related configures.
* big code cleanup for performance.
* "-L" option now uses a unit of kbytes.
2007-10-19 Cao Kai <[email protected]>
* add option "-C" and create a separate thread to do periodic checkpoint
* add option "-D" and create a separate thread to do periodic deadlock detecting
2007-10-15 Novey Donar <[email protected]>
* remove the item key in data.data, about 50% db file size reduced, but not compatible
with previous versions, warning to use.
2007-10-08 Steve Chu <[email protected]>
* add option "-L" to allow to set transaction log buffer
* add option "-N" to allow to set DB_TXN_NOSYNC flag, if someone wanted
lots of performance improved, but warning to use, because it loses transaction's durability
2007-09-29 Version 0.0.3 released
2007-09-29 Steve Chu <[email protected]>
* add command "flush_all" that will truncate a database, warning to use.
* lots of code formated
* Bugfix: Now using "-u" option can get proper db file mode
2007-09-21 Version 0.0.2 released
2007-09-20 Steve Chu <[email protected]>
* move some macro to memcachedb.h
* rename chkpoint to db_checkpoint, for consistency.
* add new command "db_archive" to remove log files that are no longer needed
* README updated
2007-09-19 Steve Chu <[email protected]>
* enlarge incr value to unsigned int to conform memcache protocol, now supports max value of 4294967295.
2007-09-18 Version 0.0.1 released