|
Hello List,
Hello Roberto,
Roberto Nibali wrote:
[...]
> Well, the current proc-fs entries are a read-only representation of what
> could be set regarding state timeouts. The 'ipvsadm --set' command will set
>
> a) The time we spend in TCP ESTABLISHED before transition
> b) The time we spend in TCP FIN_WAIT before transition
> c) The time we spend for UDP in general
>
> for IPVS related connection entries.
Let me ask the question of the foreposter again, please:
Where can I get the currently set values of "ipvsadm --set foo bar baz"?
I did not find a way to read them out, I grep through the /proc/sys/foo
and /proc/net foo and was not able to see the numbers I set before. This
was on kernel 2.4.30 at least.
This brings me further to the question if the changes of lvs in recent
2.6 development are being backported to 2.4?
Besides that to my knowledge Linux uses a timeout of about 2h + several
minutes ( 9 * 75 secs?) until a tcp connection timeouts. So, if I set a
lvs tcp timeout about 2h 12 min, lvs would never drop a tcp connection
unless a client is really "unreachable": After 2h Linux sends tcp
keepalive probes serveral times, so there are some byte send through the
connection. lvs will (re)set the internal timer for this connection to
the keepalive time I set with "--set". Or does it recognize that the
bytes send are only probes without a vaild answer and thus drop the
connection?
Best Regards,
Andreas
|
