On 05-04-14 17:18, Reindl Harald wrote: > however, lmtpd should not crash because a typo in a sieve-script and if > that is the trigger how do i get the existing 8 errors cleaned
The only way to make sure that doesn't happen due to a problem in libsieve is by re-validating each script before running it. But there is a reason validation is done at insertion time if you use dbmail-sievecmd or dbmail-timsieved. Runtime validation at execution time would affect performance. I don't know how much, so it might not be dramatic. But adding that just so people can shoot themselves in the foot? I don't know... It would be nice to have some backtraces for the problematic lmtpd process. @Curtis: if you see processes running at 100% cpu without any database load, please try running http://git.dbmail.eu/paul/dbmail/plain/test-scripts/dumpbt.sh against it with 'dbmail-lmtpd' as only argument. -- ________________________________________________________________ Paul J Stevens pjstevns @ gmail, twitter, github, linkedin www.nfg.nl/[email protected]/+31.85.877.99.97 _______________________________________________ DBmail mailing list [email protected] http://mailman.fastxs.nl/cgi-bin/mailman/listinfo/dbmail
