On 06/19/2013 02:04 PM, . . wrote: > Hi, I've tested dbmail (commit e10143b6804824bffc20752251578de734a01df8) > with dovecot's imaptest (imaptest-20130617 + dovecot-20130618) and here > is the summary:
some additional tests were fixed since then > > 35 test groups: 12 failed, 11 skipped due to missing capabilities > base protocol: 69/337 individual commands failed > extensions: 6/26 individual commands failed > > 1 search-sets > 19 search-addresses > 5 search-flags > 2 copy > 1 fetch-body > 1 fetch-envelope > 12 search-body > 3 sort-date > 3 sort-subject > 25 search-date > 2 subscribe > > Also stresstesting (imaptest checkpoint=1) with several users gave lots > of errors like these ones: I think Timo once said that imaptest output when testing concurrent access to a mailbox is not very reliable. I use stress testing for load-testing, not for compliance testing. Without checkpoint it looks better anyway. > Error: testuser[217]: uid=9067 200301061418.31353.b...@tpi.com: > RFC822.SIZE size changed 2411 -> '2412': * 2 FETCH (RFC822.SIZE 2412 > FLAGS () UID 9067) dbmail has it's only quite accurate idea of message size which is sometimes off-by-one compared to the original message fed to it. Pretty safe to ignore. > So my question is - are these errors non-critical for production usage > of dbmail? Since dbmail is in production use on some very large installation, I would say so. > It is also said here http://imapwiki.org/ImapTest/ServerStatus that > dbmail's UID/sequence mapping becomes wrong. Correct, but that was long ago. UID mapping has been fixed afaik. -- ________________________________________________________________ Paul J Stevens pjstevns @ gmail, twitter, skype, linkedin * Premium Hosting Services and Web Application Consultancy * www.nfg.nl/i...@nfg.nl/+31.85.877.99.97 ________________________________________________________________ _______________________________________________ Dbmail-dev mailing list Dbmail-dev@dbmail.org http://mailman.fastxs.nl/cgi-bin/mailman/listinfo/dbmail-dev
