|
On a Debian Sarge with CyrusIMAP MURDER 2.2.8 and MTA Postfix, the
following strange phenomenon appears : LMTP dialog between frontend and backend slow down, with an increasing mailq. What happends : few frontend lmtpproxy processes fall in a CLOSE_WAIT state on the backend(s) lmtp port. Until the timeout nothing happends, the process doesn't work anymore. Whenever I kill one of those, a new fresh process starts, working fine. It appears under charge when several dozens of mails are in the mailq. Is it symptomatic of a bad parameter or a bad network configuration? If not, what direction can I go for a better diagnostic? Is it possible to shorten this timeout in order to let a new process take place faster? Thanks for help. Laurent. "No phenomenon is a phenomenon until it is an observed phenomenon" Niels Bohr |
---- Cyrus Home Page: http://asg.web.cmu.edu/cyrus Cyrus Wiki/FAQ: http://cyruswiki.andrew.cmu.edu List Archives/Info: http://asg.web.cmu.edu/cyrus/mailing-list.html
