[
https://issues.apache.org/jira/browse/TS-3312?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14368159#comment-14368159
]
Dzmitry Markovich edited comment on TS-3312 at 3/19/15 5:08 AM:
----------------------------------------------------------------
[~amc] If I set this timeout after server_session->attach_hostname it will be
overwritten to transaction timeout once we call server_session->release() -
setting this timeout after we call server_session->release() is also not a good
idea because we will simply overwrite the value we just set (transaction
inactivity timeout to keepalive timeout).
Also in HttpSM we call release() method that releases the session for K-A reuse
5 times:
1 time in HttpSM::tunnel_handler_server
1 time in HttpSM::tunnel_handler_ua
2 times in HttpSM::do_http_server_open
1 time in HttpSM::release_server_session
So to clarify one more time, currently HttpServerSession::release() sets
timeout to transaction inactivity time out (not keep alive time out).
And there are just to options:
1) Pass a parameter to release() - like I proposed;
2) Call set_inactivity_timeout(with keep alive timeout) every time after we
call release()
I personally don't like #2 since in this case set_inactivity_timeout will be
called twice: once in release() and second time right after release().
And we cant set this timeout inside release(), because we don't have access to
overrideable keep alive timeout value with in HttpServerSession.
was (Author: dmich):
[~amc] If I set this timeout after server_session->attach_hostname it will be
overwritten to transaction timeout once we call server_session->release() -
setting this timeout after we call server_session->release() is also not a good
idea because we will simply overwrite the value we just set (transaction
inactivity timeout to keepalive timeout).
Also in HttpSM we call release() method that releases the session for K-A reuse
5 times:
1 time in HttpSM::tunnel_handler_server
1 time in HttpSM::tunnel_handler_ua
2 times in HttpSM::do_http_server_open
1 time in HttpSM::release_server_session
So to clarify one more time, currently HttpServerSession::release() sets
timeout to transaction inactivity time out (not keep alive time out).
And there are just to options:
1) Pass a parameter to release() - like I proposed;
2) Call set_inactivity_timeout(with keep alive timeout) every time after we
call release()
I personally don't like #2 since in this case set_inactivity_timeout will be
called twice: once in release() and second time right after release()
> KA timeout to origin does not seem to honor configurations
> ----------------------------------------------------------
>
> Key: TS-3312
> URL: https://issues.apache.org/jira/browse/TS-3312
> Project: Traffic Server
> Issue Type: Bug
> Components: Core, HTTP
> Reporter: Leif Hedstrom
> Assignee: Brian Geffon
> Fix For: 5.3.0
>
> Attachments: keep_alive3.diff
>
>
> Doing some basic testing, with the following settings:
> {code}
> CONFIG proxy.config.http.keep_alive_no_activity_timeout_out INT 120
> CONFIG proxy.config.http.transaction_no_activity_timeout_out INT 30
> {code}
> I see ATS timing out the origin sessions after 30sec, with a
> {code}
> CONFIG proxy.config.http.transaction_no_activity_timeout_out INT 30
> {code}
> What's also interesting, after I made a config change per Geffon's suggestion:
> {code}
> CONFIG proxy.config.http.origin_min_keep_alive_connections INT 10
> {code}
> I see the following in the diagnostic trace:
> {code}
> [Jan 21 14:19:19.416] Server {0x7fb1b4f06880} DEBUG: (http_ss) [0] [release
> session] session placed into shared pool
> [Jan 21 14:19:49.558] Server {0x7fb1b4f06880} DEBUG: (http_ss) [0]
> [session_bucket] session received io notice [VC_EVENT_INACTIVITY_TIMEOUT],
> reseting timeout to maintain minimum number of connections
> [Jan 21 14:20:19.633] Server {0x7fb1b4f06880} DEBUG: (http_ss) [0]
> [session_bucket] session received io notice [VC_EVENT_INACTIVITY_TIMEOUT],
> reseting timeout to maintain minimum number of connections
> [Jan 21 14:20:19.670] Server {0x7fb1b4f06880} DEBUG: (http_ss) [0]
> [session_pool] session 0x1cc5aa0 received io notice [VC_EVENT_EOS]
> {code}
> So, not only is it resetting the timeout twice, it also gets a VC_EVENT_EOS.
> I first though it was the origin that closed the connection, but from what I
> could tell, the timeout on the origin was set to 60s.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
