-----BEGIN PGP SIGNED MESSAGE-----
Dan Moschuk wrote: > >I noticed on a very high traffic'd webserver, I have just over 4000 sockets >stuck in the TIME_WAIT state. Ideally, I want to "bend" the RFC a bit and >close the descriptor before it hits that state, or, ignore the 2MSL wait >when it enters that state. Are they casuing you a problem? If so, what? I'm interested because I wrote a paper about this problem (http://www.isi.edu/~faber/pubs/html/infocom99/ for the html, http://www.isi.edu/~faber/pubs/time_wait.ps for postscript) and various people have told me that although the PCBs pile up, the performance effects are minimal in a reasonable OS (specifically in FreeBSD). And more directly for you, if they're not harming your performance, I wouldn't do *anything*. They do serve a purpose. > >I take it there is no sysctl switch to trigger this, so, am I going >kernel diving? I don't think there is one. Although I don't think it's a good idea to turn off the TW state, I think it's pretty easy to do so. I'm going to be away from my email for a few days, so don't be surprised if you respond and don't hear from me immediately. If folks have evidence that accumulating TW states cause a performance problem (although I doubt they do), I would love to know about it. - -- - ---------------------------------------------------------------------- Ted Faber fa...@isi.edu USC/ISI Computer Scientist http://www.isi.edu/~faber (310) 822-1511 x190 PGP Key: http://www.isi.edu/~faber/pubkey.asc -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAwUBN16qc4b4eisfQ5rpAQEregP/T/AjHQMX0mKdtSKKsmIxE7RI32UxMIV0 1tKS5mEshWwO4Y7PtGrbFHuGYBh9D1PnDcKbHZ9PI/j54zM/yfTYvhkiIsamM4G2 B4/kUPRMzrWQCPX8hKSe3zYz7ZoHjI/mHWj+i3AW18rMtA1M6AoQrNOMyyQFFu8I I4XF9dyFYGE= =6L+A -----END PGP SIGNATURE----- To Unsubscribe: send mail to majord...@freebsd.org with "unsubscribe freebsd-hackers" in the body of the message
