Hi,
THANK YOU!!!
The patch itself didn't help very mutch (stalled after the debug
statements), but it gave me a good starting point for more debugging.
Took me an hour, but then I found that this line never finished:
$new_toks{$LAST_ATIME_DELTA_MAGIC_TOKEN} = $newdelta;
This is the first access to the new database. And then I found a
leftover and damaged "_toks.new" database lying around in
~/.spamassassin. Seems there is a sanity check missing. I added this
near the beginning of this sub:
# since DB_File will not shrink a database (!!), we need to *create*
# a new one instead.
my $main = $self->{bayes}->{main};
my $path = $main->sed_path ($main->{conf}->{bayes_path});
my $name = $path.'_toks.new';
+ CHECKLOOP: while (1) {
+ for my $ext (@DB_EXTENSIONS) {
+ if (-e $name.$ext) {
+ warn "$name already exists (died expiry run?), please
clean up!";
+ $name .= $$;
+ next CHECKLOOP;
+ }
+ }
+ last CHECKLOOP;
+ }
and had to change one line near the end:
# now rename in the new one. Try several extensions
for my $ext (@DB_EXTENSIONS) {
- my $newf = $path.'_toks.new'.$ext;
+ my $newf = $name.$ext;
cu/2
Michael
On Sun, 11 Jan 2004 10:56:34 -0500, Bryan Hoover wrote:
>Michael Jacob wrote:
>>
>> Hi,
>>
>> something seems to be wrong with my bayes db, or is this considered
>> 'normal'? Log:
>
>I backed up my Bayes db, and ran with -D force-expire to see what it'd
>do. I got pretty much what you show, except after the "something fishy"
>line, it paused for a moment or two, and then finished up with:
>
>debug: bayes: couldn't find a good delta atime, need more token
>difference, skipping expire.
>debug: Syncing complete.
>debug: bayes: 90800 untie-ing
>debug: bayes: 90800 untie-ing db_toks
>debug: bayes: 90800 untie-ing db_seen
>debug: bayes: files locked, now unlocking lock
>debug: unlock: 90800 unlink /usr/home/bhoover/.spamassassin/bayes.lock
>
>So that's the way it's supposed to go. Did your run ever return?
>
>I saw a related thread in SA dev that might be helpful. It regards a
>"problem" along the same lines, though not precisely -- your run never
>returned -- that may (along with application of the associated patch)
>give some insight. The linked patch adds more debug output:
>
>news://news.gmane.org/[EMAIL PROTECTED]
>
>Copy the patch to a file, say, bayesstore-patch-file.patch, in your
>../Mail-SpamAssassin-2.60/lib/Mail/SpamAssassin directory, and run:
>
>patch < bayesstore-patch-file.patch
>
>You'll get more debug output when you run force-expire again.
>
>Most notable, with regards to the "problem", is the play between how big
>a set of tokens you want to expire -- given current number of tokens in
>db, and bayes_expiry_max_db_size -- and whether sa-learn can find a
>suitable atime for expiration selection. The additional debug output
>helps to clarify this, but see the link (and sub-links there) for
>explaination. This may not be your problem, but the added output may be
>helpful.
>
>> synced Bayes databases from journal in 0 seconds: 457 unique entries
>> (562 total entries)
>> debug: bayes: expiry check keep size, 75% of max: 112500
>> debug: bayes: token count: 196598, final goal reduction size: 84098
>
>So I read the above output to mean sa-learn wants to expire 84098
>tokens, bringing the size down to 112500 from 196598. I assume then
>that's what it'll look for -- an atime that'll allow for that. My
>understanding is if sa-learn can not find an atime allowing for less
>than or equal to the target number of token expirations, expiration will
>not be carried out.
>
>> debug: bayes: First pass? Current: 1073772599, Last: 1073562102,
>> atime: 22118400, count: 2972, newdelta: 781658, ratio: 28.2967698519515
>> debug: bayes: something fishy, calculating atime (first pass)
>
>The expiration routine did not like what it saw in the above, so it went
>on to calculate/find a new atime with which to select tokens to expire,
>and apparently, never returned... It's this part to which the patch
>adds more debug output.
>
>> and after that nothing, for
>>
>> PID USER PRI NI SIZE RSS SHARE STAT LIB %CPU %MEM TIME
>> COMMAND
>> 11464 root 14 0 15092 14M 1600 R 0 94.6 5.7 30:25
>> sa-learn
>>
>> and running.
>
>Is bayes.lock in .spamassassin? I assume you've killed the process by
>now, but you'll need to get rid of that lock file if it got left
>behind. It's used when autolearning too, btw.
>
>Bryan
>
>> SpamAssassin is 2.61, Perl is:
>>
>> Summary of my perl5 (revision 5.0 version 6 subversion 1)
>> configuration:
>> Platform:
>> osname=linux, osvers=XXX, archname=i686-linux
>> uname='linux sps-c1s9p14.j-e-b.net XXX XXX i686 unknown '
>> config_args=''
>> hint=recommended, useposix=true, d_sigaction=define
>> usethreads=undef use5005threads=undef useithreads=undef
>> usemultiplicity=undef
>> useperlio=undef d_sfio=undef uselargefiles=define usesocks=undef
>> use64bitint=undef use64bitall=undef uselongdouble=undef
>> Compiler:
>> cc='cc', ccflags ='-fno-strict-aliasing -D_LARGEFILE_SOURCE
>> -D_FILE_OFFSET_BITS=64',
>> optimize='-O2',
>> cppflags='-fno-strict-aliasing'
>> ccversion='', gccversion='2.95.3 19991030 (prerelease)',
>> gccosandvers=''
>> intsize=4, longsize=4, ptrsize=4, doublesize=8, byteorder=1234
>> d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=12
>> ivtype='long', ivsize=4, nvtype='double', nvsize=8, Off_t='off_t',
>> lseeksize=8
>> alignbytes=4, usemymalloc=n, prototype=define
>> Linker and Libraries:
>> ld='cc', ldflags =' -L/usr/local/lib'
>> libpth=/usr/local/lib /lib /usr/lib
>> libs=-lnsl -lndbm -ldb -ldl -lm -lc -lposix -lcrypt -lutil
>> perllibs=-lnsl -ldl -lm -lc -lposix -lcrypt -lutil
>> libc=/lib/libc-2.1.1.so, so=so, useshrplib=false, libperl=libperl.a
>> Dynamic Linking:
>> dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun=undef,
>> ccdlflags='-rdynamic'
>> cccdlflags='-fpic', lddlflags='-shared -L/usr/local/lib'
>>
>> Characteristics of this binary (from libperl):
>> Compile-time options: USE_LARGE_FILES
>> Built under linux
>> Compiled at Jan 8 2002 10:55:38
>> @INC:
>> /opt/perl/lib/5.6.1/i686-linux
>> /opt/perl/lib/5.6.1
>> /opt/perl/lib/site_perl/5.6.1/i686-linux
>> /opt/perl/lib/site_perl/5.6.1
>> /opt/perl/lib/site_perl
>> .
>>
>> -------------------------------------------------------
>> This SF.net email is sponsored by: Perforce Software.
>> Perforce is the Fast Software Configuration Management System offering
>> advanced branching capabilities and atomic changes on 50+ platforms.
>> Free Eval! http://www.perforce.com/perforce/loadprog.html
>
>--
>My thinking is a passion. I am very good at rooting out truffles for
>others; I myself take no pleasure in them. I root out the problems with
>my snout, but all I can do with them is toss them back over my head. -
>(Soren Kierkegaard - Either/Or)
>
>http://www.wecs.com/content.htm
>
>This signature file is generated by Pick-a-Tag !
>Written by Jeroen van Vaarsel
>http://www.google.com/search?hl=en&ie=ISO-8859-1&q=pick-a-tag
>
>
>
>-------------------------------------------------------
>This SF.net email is sponsored by: Perforce Software.
>Perforce is the Fast Software Configuration Management System offering
>advanced branching capabilities and atomic changes on 50+ platforms.
>Free Eval! http://www.perforce.com/perforce/loadprog.html
>_______________________________________________
>Spamassassin-talk mailing list
>[EMAIL PROTECTED]
>https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
-------------------------------------------------------
This SF.net email is sponsored by: Perforce Software.
Perforce is the Fast Software Configuration Management System offering
advanced branching capabilities and atomic changes on 50+ platforms.
Free Eval! http://www.perforce.com/perforce/loadprog.html
_______________________________________________
Spamassassin-talk mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
