3 innduct \- quickly and reliably stream Usenet articles to remote site
11 implements NNTP peer-to-peer news transmission including the streaming
12 extensions, for sending news articles to a remote site.
14 You need to run one instance of innduct for each peer site. innduct
15 manages its interaction with innd, including flushing the feed as
16 appropriate, etc., so that articles are transmitted quickly, and
17 manages the retransmission of its own backlog. innduct includes the
18 locking necessary to avoid multiple simutaneous invocations.
20 By default, innduct reads the default feedfile corresponding to
24 .IR pathoutgoing / site )
25 and feeds it via NNTP, streaming if possible, to the host
30 is not specified, it defaults to
33 innduct daemonises after argument parsing, and all logging (including
34 error messages) are sent to syslog (facility
37 The best way to run innduct is probably to periodically invoke innduct
38 for each feed (e.g. from cron), passing innduct it the
40 option to arrange that it silently exits if an innduct is already
41 running for that site.
42 .SH INNDUCT VS INNFEED/NNTPSEND/INNXMIT
45 does roughly the same thing as innduct. However, the way it receives
46 information from innd can result in articles being lost (not offered
47 to peers) if innfeed crashes for any reason. This is an inherent
48 defect in the innd channel feed protocol. innduct uses a file feed,
49 constantly "tailing" the feed file, and where implemented uses
51 to reduce the latency which would come from having to constantly poll
52 the feed file. innfeed is capable of feeding multiple peers from a
53 single innfeed instance, whereas each innduct process handles exactly
54 one peer. innduct is much smaller and simpler, at <4kloc to innfeed's
55 25kloc. innfeed needs a separate wrapper script or similar
56 infrastructure (of which there is an example in its manpage), whereas
57 innduct can be run directly and doesn't need help from shell scripts.
60 processes feed files in batch mode. That is, you have to periodically
61 invoke nntpsend, and when you do, the feed is flushed and articles
62 which arrived before the flush are sent to the peer. This introduces
63 a batching delay, and also means that the NNTP connection to the peer
64 needs to be remade at each batch. nntpsend (which uses innxmit)
65 cannot make use of multiple connections to a single peer site.
66 However, nntpsend can be left to find automatically which sites need
71 is the actual NNTP feeder program used by nntpsend.
74 .BR \-f | \-\-feedfile= \fIfeedfile\fR
77 If the specified value ends in a
79 it is taken as a directory to use as if it were
81 and the actual feed file used is
82 .IR specified_feedfile / site .
84 .BR \-q | \-\-quiet-multiple
85 Makes innduct silently exit (with status 0) if another innduct holds
86 the lock for the site. Without \fB-q\fR, this causes a fatal error to
87 be logged and a nonzero exit.
90 Do not daemonise. innduct runs in the foreground and all messages
91 (including all debug messages) are written to stderr. A control
92 command line is also available on stdin/stdout.
95 Do not try to use the streaming extensions to NNTP (for use eg if the
96 peer can't cope when we send MODE STREAM).
99 Do not try to use the file change monitoring support to watch for
100 writes by innd to the feed file; poll it instead. (If file monitoring
101 is not compiled in, this option just downgrades the log message which
102 warns about this situation.)
104 .BR \-C | \-\-inndconf= \fIFILE\fR
107 instead of the default
110 .BR \-\-cli= \fIDIR\fR / |\fIPATH\fR
111 Listen for control command line connections on
113 (if the value ends with a
117 (if it doesn't). See CONTROLLING INNDUCT, below.
118 This option may be essential, if the
121 is too long, as there is a fairly short limit on the paths to AF_UNIX
122 sockets. The default is
123 .IR feedfile \fB_cli\fR.
128 at the remote site rather than to the NNTP port (119).
131 Just print a brief usage message and list of the options to stdout.
133 You should not normally need to adjust these. Time intervals may
134 specified in seconds, or as a number followed by one of the following
137 .BR "sec min hour day" ,
140 .BI \-\-max-connections= max
141 Restricts the maximum number of simultaneous NNTP connections used by
146 There is no global limit on the number of connections.
148 .BI \-\-max-queue-per-conn= per-conn-max
149 Restricts the maximum number of outstanding articles queued on any
150 particular connection to
152 (Non-streaming connections can only handle one article at a time.)
156 .BI \-\-max-queue-per-file= max
157 Restricts the maximum number articles read into core from any one
160 The default is twice the value of per-conn-max.
162 .BI \-\-feedfile-flush-size= bytes
163 Specifies that innduct should flush the feed and start a new feedfile
164 when the existing feedfile size exceeds
166 the effect is that the innduct will try to avoid the various
167 batchfiles growing much beyond this size while the link to the peer is
168 working. The default is
171 .BI \-\-period-interval= PERIOD-INTERVAL
172 Specifies wakup interval and period granularity.
173 innduct wakes up every
175 to do various housekeeping checks. Also, many of the timeout and
176 rescan intervals (those specified in this manual as
178 are rounded up to the next multiple of
179 .IR PERIOD-INTERVAL .
181 .BI \-\-connection-timeout= TIME
182 How long to allow for a connection setup attempt before giving up.
186 .BI \-\-stuck-flush-timeout= TIME
187 How long to wait for innd to respond to a flush request before giving
191 .BI \-\-feedfile-poll= TIME
192 How often to poll the feedfile for new articles written by innd
195 or equivalent) is not available. (When file monitoring is available,
196 there is no need for periodic checks and we wake immediately up
197 whenever the feedfile changes.)
201 .BI \-\-no-check-proportion= PERCENT
202 If the moving average of the proportion of articles being accepted
203 (rather than declined) by the peer exceeds this value, innduct uses
204 "no check mode" - ie it just sends the peer the articles with TAKETHIS
205 rather than checking first with CHECK whether the article is wanted.
206 This only affects streaming connections. The default is
210 .BI \-\-no-check-response-time= ARTICLES
211 The moving average mentioned above is an alpha-smoothed value with a
217 .BI \-\-reconnect-interval= RECONNECT-PERIOD
218 Limits initiation of new connections to one each
219 .IR RECONNECT-PERIOD .
220 This applies to reconnections if the peer has been down, and also to
221 ramping up the number of connections we are using after startup or in
222 response to an article flood. The default is
225 .BI \-\-flush-retry-interval= PERIOD
226 If our attempt to flush the feed failed (usually this will be because
227 innd is not running), try again after
232 .BI \-\-earliest-deferred-retry= PERIOD
233 When the peer responds to our offer of an article with a 431 or 436
234 NNTP response code, indicating that the article has already been
235 offered to it by another of its peers, and that we should try again,
238 before offering the article again. The default is
241 .BI \-\-backlog-rescan-interval= BACKLOG-SCAN-PERIOD
242 We scan the directory containing
244 for backlog files at least every
245 .IR BACKLOG-SCAN-PERIOD ,
246 in case the administrator has manually dropped in a file there for
250 .BI \-\-max-flush-interval= PERIOD
251 We flush the feed at least every
253 even if the current instance of the feedfile has not reached the size
258 .BI \-\-max-flush-interval= PERIOD
259 We flush the feed and start a new feedfile at least every
261 even if the current instance of the feedfile has not reached the size
266 .BI \-\-flush-finish-timeout= FLUSH-FINISH-PERIOD
268 .IR FLUSH-FINISH-PERIOD
269 ago, and are still trying to finish processing articles that were
270 written to the old feed file, we forcibly and violently make sure that
271 we do by abandoning and deferring all the work (which includes
272 unceremoniously dropping any connections on which we've sent some of
273 those articles but not yet had replies, as they're probably stuck
278 .BI \-\-idle-timeout= PERIOD
279 Connections which have had no activity for
281 will be closed. This includes connections where we have sent commands
282 or articles but have not yet had the responses, so this same value
283 doubles as the timeout after which we conclude that the peer is
284 unresponsive or the connection has become broken.
288 .BI \-\-max-bad-input-data-ratio= PERCENT
289 We tolerate up to this proportion of badly-formatted lines in the
290 feedfile and other input files. Every badly-formatted line is logged,
291 but if there are too many we conclude that the corruption to our
292 on-disk data is too severe, and crash; to successfully restart,
293 administrator intervention will be required. This avoids flooding the
294 logs with warnings and also arranges to abort earlyish if an attempt
295 is made to process a file in the wrong format.
300 .BI \-\-max-bad-input-data-init= LINES
301 Additionally, we tolerate this number of additional badly-formatted
302 lines, so that if the badly-formatted lines are a few but at the start
303 of the file, we don't crash immediately.
306 (which would suffice to ignore one whole corrupt 4096-byte disk block
307 filled with random data, or one corrupt 1024-byte disk block filled
308 with an inappropriate text file with a mean line length of at least
310 .SH CONTROLLING INNDUCT
311 If you tell innd to drop the feed, innduct will (when it notices,
312 which will normally be the next time it decides flushes) finish up the
313 articles it has in hand now, and then exit. It is harmless to cause
314 innd to flush the feed (but innduct won't notice and flushing won't
315 start a new feedfile; you have to leave that to innduct).
317 If you want to stop innduct you can send it SIGTERM or SIGINT, or the
319 control command, in which case it will report statistics so far and
320 quickly exit. If innduct receives SIGKILL nothing will be broken or
321 corrupted; you just won't see some of the article stats.
323 innduct listens on an AF_UNIX socket, and provides a command-line
324 interface which can be used to trigger various events and for
325 debugging. The socket is called
327 and when connected reads and writes lines (with unix line endings).
328 It can most easily be accessed with a program like
332 .IR feedfile \fB_cli\fR)
338 The following control commands are supported:
341 Print a list of all the commands understood. This list includes
342 undocumented commands which mess with innduct's internal state and
343 should only be used by a developer in conjuction with the innduct
347 Start a new feed file and trigger a flush of the feed. (Or, cause
349 .I FLUSH-FINISH-PERIOD
350 to expire early, forcibly completing a previously started flush.)
353 Log statistics and exit. (Same effect as SIGTERM or SIGINT.)
356 Writes information about innduct's state to a plain text file
357 .IR feedfile \fB_dump\fR.
358 This overwrites any previous dump.
360 is a summary including general state and a list of connections;
362 also includes information about each article innduct is dealing with.
365 Requests that innduct rescan for new backlog files at the next
367 poll. Normally innduct assumes that any backlog files dropped in by
368 the administrator are not urgent and may not get around to noticing
370 .IR BACKLOG-SCAN-PERIOD .
373 Resets the connection startup delay counter so that innduct may
374 consider making a new connection to the peer right away, regardless
376 .IR RECONNECT-PERIOD .
377 A connection attempt will still only be made if innduct feels that it
378 needs one, and innduct may wait up to
380 before actually starting the attempt.
381 .IR BACKLOG-SCAN-PERIOD .
385 An instance of innduct is already running for this
392 The feed has been dropped by innd, and we (or previous innducts) have
393 successfully offered all the old articles to the peer site. Our work
397 innduct was invoked with bad options or command line arguments. The
398 error message will be printed to stderr, and also (if any options or
399 arguments were passed at all) to syslog with severity
403 Things are going wrong, hopefully shortage of memory, system file
404 table entries; disk IO problems; disk full; etc. The specifics of the
405 error will be logged to syslog with severity
407 (if syslog is working!)
410 Things are going badly wrong in an unexpected way: system calls which
411 are not expected to fail are doing so, or the protocol for
412 communicating with innd is being violated, or some such. Details will
413 be logged with severity
415 (if syslog is working!)
418 These exit statuses are used by children forked by innduct to
419 communicate to the parent. You should not see them. If you do, it is
422 innduct dances a somewhat complicated dance with innd to make sure
423 that everything goes smoothly and that there are no races. (See the
424 two ascii-art diagrams in innduct.c for details of the protocol.) Do
425 not mess with the feedfile and other associated files, other than as
428 .IP \fIpathoutgoing\fR/\fIsite\fR
429 .IX Item "default feedfile"
434 Main feed file as specified in
436 This and other batchfiles used by innduct contains lines each of which
438 \& \fItoken\fR \fImessageid\fR
439 where \fItoken\fR is the inn storage API token. Such lines can be
440 written by \fBTf,Wnm\fR in a \fInewsfeeds\fR(5) entry. During
441 processing, innduct overwrites lines in the batch files which
442 correspond to articles it has processed: the line is replaced with
443 one containing only spaces. Only innd should create this file, and
444 only innduct should remove it.
445 .IP \fIfeedfile\fR_lock
447 Lockfile, preventing multiple innduct invocations for the same
448 feed. A process holds this lock after it has opened the lockfile,
449 made an fcntl F_SETLK call, and then checked with stat and fstat that
450 the file it now has open and has locked still has the name
451 \fIfeedfile\fR_lock. (Only) the lockholder may delete the lockfile.
452 For your convenience, after the lockfile is locked,
459 are all written to the lockfile. NB that stale lockfiles may contain
460 stale data so this information should not be relied on other than for
462 .IP \fIfeedfile\fR_flushing
463 .IX Item "flushing file"
464 Batch file: the main feedfile is renamed to this filename by innduct
465 before it asks inn to flush the feed. Only innduct should create or
467 .IP \fIfeedfile\fR_defer
468 .IX Item "flushing file"
469 Batch file containing details of articles whose transmission has
470 recently been deferred at the request of the recipient site. Created,
471 written, read and removed by innduct.
472 .IP \fIfeedfile\fR_backlog.\fItime_t\fR.\fIinum\fR
473 .IX Item "backlog file"
474 Batch file containing details of articles whose transmission has less
475 recently been deferred at the request of the recipient site. Created
476 by innduct, and will also be read and removed by innduct. However you
477 (the administrator) may also safely remove backlog files.
478 .IP \fIfeedfile\fR_backlog\fIsomething\fR
479 .IX Item "manual backlog file"
480 Batch file manually provided by the administrator. The file should be
481 complete and ready to process at the time it is renamed or hardlinked
482 to this name. innduct will then automatically find and read and
483 process it and eventually remove it. The administrator may also
484 safely remove backlog files. \fIsomething\fR may not contain \fB#\fR
485 \fB~\fR or \fB/\fR. Be sure to have finished writing the file before
486 you rename it to match the pattern \fIfeedfile\fR\fB_backlog\fR*, as
487 otherwise innduct may find and process the file and read it to EOF
488 before you have finished creating it.
489 .IP \fIfeedfile\fR_cli
490 .IX Item "control command line socket"
491 Default AF_UNIX listening socket for the control command line. See
492 CONTROLLING INNDUCT, above.
493 .IP \fIfeedfile\fR_dump
494 .IX Item "debug dump file"
495 On request via a control connection innduct dumps a summary of its
496 state to this text file. This is mostly useful for debugging.
497 .IP /etc/news/inn.conf
501 if none is specified, for finding how to communicate with innd,
507 Written by Ian Jackson <ijackson@chiark.greenend.org.uk>
~ian / inn-innduct.git / blob