SpamAssassin 2.60 Upgrade Question - bayes_jour nal file needed?

[Desai, Jason]

Perhaps the files will appear after spamassassin has seen 200 spam and 200
ham?

Jason

> -----Original Message-----
> From: Richard Bollinger [mailto:rabollinger at COMCAST.NET]
> Sent: Tuesday, September 23, 2003 9:53 PM
> To: MAILSCANNER at JISCMAIL.AC.UK
> Subject: [MAILSCANNER] SpamAssassin 2.60 Upgrade Question -
> bayes_journal file needed?
>
>
> Today I upgraded two nearly identical Linux servers running
> MS with sendmail and SpamAssassin
> 2.55 to version 2.60.  After the upgrades, I noticed that
> only one of them was using a
> bayes_journal file.  As I watched, I could see a
> bayes_toks.new file appear, grow to full size
> and the replace bayes_toks only on the server which didn't
> have a bayes_journal file.
>
> It didn't look right, so taking the suggestion from another
> user's email, I stopped MS, wiped
> out the "state" directories on each server, then restarted
> MS.  It seems to be working, but now
> I have just two files, bayes_toks and bayes_seen, on each
> server... no bayes_journal files on
> either one.
>
> Have a missed a step?  Do I need the bayes_journal files?
>
> It looks like the bayes database is being rebuilt just fine
> without it:
>
> # sa-learn --dbpath /u/MailScanner/var/spamassassin --dump magic
> 0.000          0          2          0  non-token data: bayes
> db version
> 0.000          0         17          0  non-token data: nspam
> 0.000          0         62          0  non-token data: nham
> 0.000          0       8291          0  non-token data: ntokens
> 0.000          0 1064349609          0  non-token data: oldest atime
> 0.000          0 1064367209          0  non-token data: newest atime
> 0.000          0          0          0  non-token data: last
> journal sync atime
> 0.000          0          0          0  non-token data: last
> expiry atime
> 0.000          0          0          0  non-token data: last
> expire atime delta
> 0.000          0          0          0  non-token data: last
> expire reduction count
>
> Thanks, Rich Bollinger
>