On Thursday 01 February 2007 12:34, James Harper wrote: > > > Yes, providing you don't mind prematurely killing off non-keepalive > > > programs > > > that are inactive during the reduced keepalive period you have set. > > > > This should be relatively easy to test... assuming we can't find a > > document somewhere that clarifies it one way or another. > > Okay... if you are willing to accept a random page from the internet as > gospel... > > >From http://ipsysctl-tutorial.frozentux.net/chunkyhtml/tcpvariables.html > > (emphasis mine) > > " > 3.3.10. tcp_keepalive_time > > The tcp_keepalive_time variable tells the TCP/IP stack how often to send > TCP keepalive packets to keep an connection alive if it is currently > unused. ___This value is only used when keepalive is enabled.___
Yes, it is used for keeping the line alive when keepalive is enable. Otherwise, it is used to determine when to timeout an inactive line. I don't have my copy of Stevens in front of me, so I cannot 100% guarantee what I am saying. However, if keepalive does not serve to terminate an inactive line then there must be another kernel variable that does set the line timeout, and that is typically not something the TCP/IP code does --- it is very lean and does not keep multiple copies of variables that do the same thing. > > The tcp_keepalive_time variable takes an integer value which is counted > in seconds. The default value is 7200 seconds, or 2 hours. This should > be a good value for most hosts and will not take too much network > resources from you. Do not set this value to low since it will then use > up your network resources with unnecessary traffic. > " > > And Documentation/networking/ip-sysctl.txt in the Linux source tree: > > " > tcp_keepalive_time - INTEGER > How often TCP sends out keepalive messages when keepalive is > enabled. > Default: 2hours. > " > > Again... not explicitely saying that applications that haven't enabled > SO_KEEPALIVE will have their connections terminated after that many > seconds of idle, but I think that if that were the case it would be > mentioned. > > I'll check the kernel source and see if it has anything useful to say > about it. > > The other nice thing about the keepalives is, as you mention above, if > the remote client does go dead, the director will figure it out sooner > and be able to move along without holding things up. Without keepalives, > an application that is waiting for traffic will literally wait forever. > Bacula does it's own application keepalives so it shouldn't suffer from > this. > > A downside though, is that if the remote client were to go off the air > for 6 minutes, the connection would be terminated, even if it would > otherwise have been able to continue. > > James > > > > ------------------------------------------------------------------------- > Using Tomcat but need to do more? Need to support web services, security? > Get stuff done quickly with pre-integrated technology to make your job > easier. Download IBM WebSphere Application Server v.1.0.1 based on Apache > Geronimo > http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 > _______________________________________________ > Bacula-users mailing list > Bacula-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/bacula-users ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier. Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
