1OVDB(5) InterNetNews Documentation OVDB(5)
2
6 ovdb - Overview storage method for INN
7
9 The ovdb overview is a storage method that uses the Berkeley DB library
10 to store overview data. It requires version 4.4 or later of the
11 Berkeley DB library (4.7+ is recommended because older versions suffer
12 from various issues).
13 The ovdb overview method makes use of the full
15 transaction/logging/locking functionality of the Berkeley DB
16 environment. Berkeley DB may be downloaded from
17 <http://www.oracle.com/technetwork/database/database-technologies/berkeleydb/overview/index.html>
18 and is needed to build the ovdb backend.
19
21 There are several versions of the ovdb storage method:
22 • Version 1, the initial version shipped with INN 2.3.0 up to
24 INN 2.3.5.
25 • Version 2, with improved performance, since INN 2.4.0.
27 • Version 3, corresponding to version 2 with compression enabled,
29 starting with INN 2.5.0.
30 If you have a database created with a previous version of ovdb, your
32 database will need to be upgraded using ovdb_init. See the
33 ovdb_init(8) man page for upgrade instructions, as well as the
34 COMPRESSION section below.
35 Note that when the Berkeley DB library is updated to a newer version,
37 the ovdb database also needs being upgraded.
38
40 If the Berkeley DB library is found at configure time, INN will be
41 built with Berkeley DB support unless the --without-bdb flag is
42 explicitly passed to configure. By default, configure will search for
43 Berkeley DB in standard locations; there will be a message in the
44 configure output indicating the pathname that will be used.
45 You can override this pathname by adding a path to the option, for
47 instance --with-bdb=/usr/BerkeleyDB.4.4. This directory is expected to
48 have subdirectories include and lib (lib32 and lib64 are also checked),
49 containing respectively db.h, and the library itself. In case non-
50 standard paths to the Berkeley DB libraries are used, one or both of
51 the options --with-bdb-include and --with-bdb-lib can be given to
52 configure with a path.
53 The ovdb database may take up more disk space for a given spool than
55 the other overview methods. Plan on needing at least 1.1 KB for every
56 article in your spool (not counting crossposts). So, if you have 5
57 million articles, you'll need at least 5.5 GB of disk space for ovdb.
58 With compression enabled, this estimate changes to 0.9 KB per article,
59 so you'll need at least 4.5 GB of disk space for 5 million articles.
60 See the COMPRESSION section below. Plus, you'll need additional space
61 for transaction logs: at least 100 MB. By default, the transaction
62 logs go in the same directory as the database. To improve performance,
63 they can be placed on a different disk -- see the DB_CONFIG section.
64
66 To enable the ovdb overview method, set the ovmethod parameter in
67 inn.conf to "ovdb". The ovdb database is stored in the directory
68 specified by the pathoverview parameter in inn.conf. This is the
69 "DB_HOME" directory. To start out, this directory should be empty
70 (other than an optional DB_CONFIG file; see DB_CONFIG for details), and
71 innd (or makehistory) will create the files as necessary in that
72 directory. Also, make sure the directory is owned by the news user.
73 Other parameters for configuring ovdb are in the ovdb.conf
75 configuration file. The following parameters can be set in that file:
76 compress
78 If INN was compiled with zlib, and this compress parameter is true,
79 ovdb will compress overview records that are longer than 600 bytes.
80 See the COMPRESSION section below.
81 cachesize
83 Size of the memory pool cache, in kilobytes. The cache will have a
84 backing store file in the DB directory which will be at least as
85 big. In general, the bigger the cache, the better. Use "ovdb_stat
86 -m" to see cache hit percentages. To make a change of this
87 parameter take effect, shut down and restart INN (be sure to kill
88 all of the nnrpd processes when shutting down). Default is 8000
89 (KB), which is adequate for small to medium-sized servers. Large
90 servers will probably need at least 20000 (KB).
91 ncache
93 Number of regions across which to split the cache. The region size
94 is equal to cachesize divided by ncache. Default is 1 for ncache,
95 that is to say the cache will be allocated contiguously in memory.
96 numdbfiles
98 Overview data is split between this many files. Currently, innd
99 will keep all of the files open, so don't set this too high or innd
100 may run out of file descriptors. nnrpd only opens one at a time,
101 regardless. May be set to one, or just a few, but only do that if
102 your OS supports large (> 2 GB) files. Changing this parameter has
103 no effect on an already-established database. Default is 32.
104 txn_nosync
106 If txn_nosync is set to false, Berkeley DB flushes the log after
107 every transaction. This minimizes the number of transactions that
108 may be lost in the event of a crash, but results in significantly
109 degraded performance. Default is true.
110 useshm
112 If useshm is set to true, Berkeley DB will use shared memory
113 instead of mmap for its environment regions (cache, lock, etc).
114 With some platforms, this may improve performance. Default is
115 false.
116 shmkey
118 Sets the shared memory key used by Berkeley DB when useshm is true.
119 Berkeley DB will create several (usually 5) shared memory segments,
120 using sequentially numbered keys starting with "shmkey". Choose a
121 key that does not conflict with any existing shared memory segments
122 on your system. Default is 6400.
123 pagesize
125 Sets the page size for the DB files (in bytes). Must be a power of
126 2. Best choices are 4096 or 8192. The default is 8192. Changing
127 this parameter has no effect on an already-established database.
128 minkey
130 Sets the minimum number of keys per page. See the Berkeley DB
131 documentation for more information. Default is based on page size
132 and whether compression is enabled:
133 default_minkey = MAX(2, pagesize / 2600) if compress is false
135 default_minkey = MAX(2, pagesize / 1500) if compress is true
136 The lowest allowed minkey is 2. Setting minkey higher than the
138 default is not recommended, as it will cause the databases to have
139 a lot of overflow pages. Changing this parameter has no effect on
140 an already-established database.
141 maxlocks
143 Sets the Berkeley DB lk_max parameter, which is the maximum number
144 of locks that can exist in the database at the same time. Default
145 is 4000.
146 nocompact
148 The nocompact parameter affects the behaviour of expireover. The
149 expireover function in ovdb can do its job in one of two ways: by
150 simply deleting expired records from the database; or by re-writing
151 the overview records into a different location leaving out the
152 expired records. The first method is faster, but it leaves 'holes'
153 that result in space that can not immediately be reused. The
154 second method 'compacts' the records by rewriting them.
155 If this parameter is set to 0, expireover will compact all
157 newsgroups; if set to 1, expireover will not compact any
158 newsgroups; and if set to a value greater than one, expireover will
159 only compact groups that have less than that number of articles.
160 Experience has shown that compacting has minimal effect (other than
162 making expireover take longer) so the default is 1. This parameter
163 will probably be removed in the future.
164 readserver
166 When the readserver parameter is set to false, each nnrpd process
167 directly accesses the Berkeley DB environment. The process of
168 attaching to the database (and detaching when finished) is fairly
169 expensive, and can result in high loads in situations when there
170 are lots of reader connections of relatively short duration.
171 When the readserver parameter is set to true, the nnrpd processes
173 will access overview via a helper server (ovdb_server -- which is
174 started by ovdb_init). All ovdb reads will then be funnelled
175 through a single process with a cleaner interface to the underlying
176 Berkeley DB database. This will result in cleaner shutdowns for
177 the database, improving stability and avoiding deadlocks, timing
178 issues and corrupted databases. That's why you should try to set
179 this parameter to true if you are experiencing any instability in
180 the ovdb overview method.
181 Default value is true.
183 numrsprocs
185 This parameter is only used when readserver is true. It sets the
186 number of ovdb_server processes. As each ovdb_server can process
187 only one transaction at a time, running more servers can improve
188 reader response times. Default is 5.
189 maxrsconn
191 This parameter is only used when readserver is true. It sets a
192 maximum number of readers that a given ovdb_server process will
193 serve at one time. This means the maximum number of readers for
194 all of the ovdb_server processes is (numrsprocs * maxrsconn). This
195 does not limit the actual number of readers, since nnrpd will fall
196 back to opening the database directly if it can't connect to an
197 ovdb_server. Default is 0, which means an unlimited number of
198 connections is allowed.
199
201 The ovdb storage method has the ability to compress overview data
202 before it is stored into the database. In addition to consuming less
203 disk space, compression keeps the average size of the database keys
204 smaller. This in turn increases the average number of keys per page,
205 which can significantly improve performance and also helps keep the
206 database more compact. This feature requires that INN be built with
207 zlib. Only records larger than 600 bytes get compressed, because that
208 is the point at which compression starts to become significant.
209 If compression is not enabled (either from the compress option in
211 ovdb.conf or INN was not built with zlib support), the database will be
212 backward compatible with older versions of ovdb. However, if
213 compression is enabled, the database is marked with a newer version
214 that will prevent older versions of ovdb from opening the database.
215 You can upgrade an existing database to use compression simply by
217 setting compress to true in ovdb.conf. Note that existing records in
218 the database will remain uncompressed; only new records added after
219 enabling compression will be compressed.
220 If you disable compression on a database that previously had it
222 enabled, new records will be stored uncompressed, but the database will
223 still be incompatible with older versions of ovdb (and will also be
224 incompatible with this version of ovdb if INN was not built with zlib
225 support). So to downgrade to a completely uncompressed database, you
226 will have to rebuild the database using makehistory.
227
229 A file called DB_CONFIG may be placed in the database directory
230 (pathoverview in inn.conf) to customize where the various database
231 files and transaction logs are written. By default, all of the files
232 are written in the "DB_HOME" directory. One way to improve performance
233 is to put the transaction logs on a different disk. To do this, put:
234 DB_LOG_DIR /path/to/logs
236 in the DB_CONFIG file. If the pathname you give starts with a "/", it
238 is treated as an absolute path; otherwise, it is relative to the
239 "DB_HOME" directory. Make sure that any directories you specify exist
240 and have proper ownership/mode before starting INN, because they won't
241 be created automatically. Also, don't change the DB_CONFIG file while
242 anything that uses ovdb is running.
243 Another thing that you can do with this file is to split the overview
245 database across multiple disks. In the DB_CONFIG file, you can list
246 directories that Berkeley DB will search when it goes to open a
247 database.
248 For example, let's say that you have pathoverview set to /mnt/overview
250 and you have four additional file systems created on /mnt/ovX. You
251 would create a file /mnt/overview/DB_CONFIG containing the following
252 lines:
253 set_data_dir /mnt/overview
255 set_data_dir /mnt/ov1
256 set_data_dir /mnt/ov2
257 set_data_dir /mnt/ov3
258 set_data_dir /mnt/ov4
259 Distribute your ovNNNNN files into the four filesystems (say, 8 each).
261 When called upon to open a database file, the db library will look for
262 it in each of the specified directories (in order). If said file is
263 not found, one will be created in the first of those directories.
264 Whenever you change DB_CONFIG or move database files around, make sure
266 all news processes that use the database are shut down first (including
267 nnrpd processes).
268 The DB_CONFIG functionality is part of Berkeley DB itself, rather than
270 something provided by ovdb. See the Berkeley DB documentation for
271 complete details for the version of Berkeley DB that you're running.
272
274 When starting the news system, rc.news will invoke the ovdb_init
275 program. See the ovdb_init(8) man page for information about the tasks
276 it performs. ovdb_init must be run before using the database.
277 And when stopping INN, rc.news kills the ovdb_monitor processes after
279 the other INN processes have been shut down.
280
282 Problems relating to ovdb are logged to news.err with "OVDB" in the
283 error message.
284 INN programs that use overview will fail to start up if the
286 ovdb_monitor processes aren't running. Be sure to run ovdb_init before
287 running anything that accesses overview.
288 Also, INN programs that use overview will fail to start up if the user
290 running them is not the news user.
291 If a program accessing the database crashes, or otherwise exits
293 uncleanly, it might leave a stale lock in the database. This lock
294 could cause other processes to deadlock on that stale lock. To fix
295 this, shut down all news processes (using "kill -9" if necessary) and
296 then restart. ovdb_init should perform a recovery operation which will
297 remove the locks and repair damage caused by killing the deadlocked
298 processes.
299
301 pathetc/inn.conf
302 The ovmethod and pathoverview parameters are relevant to ovdb.
303 pathetc/ovdb.conf
305 Optional configuration file for tuning. See CONFIGURATION above.
306 pathoverview
308 Directory where the database goes. Berkeley DB calls it the
309 "DB_HOME" directory.
310 pathoverview/DB_CONFIG
312 Optional file to configure the layout of the database files.
313 pathrun/ovdb.sem
315 A file that gets locked by every process that is accessing the
316 database. This is used by ovdb_init to determine whether the
317 database is active or quiescent.
318 pathrun/ovdb_monitor.pid
320 Contains the process ID of ovdb_monitor.
321
323 Implement a way to limit how many databases can be open at once (to
324 reduce file descriptor usage); maybe using something similar to the
325 cache code in legacy ov3.c file.
326
328 Written by Heath Kehoe <hakehoe@avalon.net> for InterNetNews.
329 $Id: ovdb.pod 10525 2021-01-20 11:51:15Z iulius $
331
333 inn.conf(5), innd(8), makehistory(8), nnrpd(8), ovdb_init(8),
334 ovdb_monitor(8), ovdb_stat(8).
335 Berkeley DB documentation: in the docs directory of the Berkeley DB
337 source distribution, or on the Oracle Berkeley DB web page
338 (<http://www.oracle.com/technetwork/database/database-technologies/berkeleydb/overview/index.html>).
339INN 2.6.4 2021-01-21 OVDB(5)