+1 On Thu, Nov 19, 2020 at 6:15 PM Susan Hinrichs <shinr...@verizonmedia.com.invalid> wrote:
> There is also a proxy.config.http.parent_proxy.connect_attempts_timeout. I > have less familiarity with the parent selection path, but I question the > value of separating that case out as well. > > On Thu, Nov 19, 2020 at 6:05 PM Susan Hinrichs <shinr...@verizonmedia.com> > wrote: > > > I just got surprised by the existence of > > proxy.config.http.post_connect_attempts_timeout again while tracking down > > an issue for the ops team. There is also a > > proxy.config.http.connect_attempts_timeout which covers the connect > > timeouts for all the methods expect POST. If you forget to set the post > > one, you get the default connect timeout of 1800 seconds for POST > methods. > > > > > > > https://urldefense.proofpoint.com/v2/url?u=https-3A__docs.trafficserver.apache.org_en_latest_admin-2Dguide_files_records.config.en.html-3F-23proxy-2Dconfig-2Dhttp-2Dconnect-2Dattempts-2Dtimeout&d=DwIBaQ&c=sWW_bEwW_mLyN3Kx2v57Q8e-CRbmiT9yOhqES_g_wVY&r=5nE_8e-Jc1t5vF6GVeub9BCN4FzSc_6kU7_mjSiUrDs&m=NRficC7PAQI1hQG7LzqwzUKCi0s7HZEjAddMwLiWcd8&s=DZDwc8-oZ_qRVFlY-Zm8vOI0YFQ9JkOAcxaHQyaVhRw&e= > > > > I presume it was useful to have two distinct timeouts when the connect > > timeout was really the timeout for time-to-first-byte. Now that the > > *connect_attempts_timeout just covers the time to complete the TCP or TLS > > handshake, there is not much point in having a separate control for the > > POST method. > > > > > > >
