Question regarding hostdb

serve_stale_for, that is the only reason i can think you are still going to the OOR host. I'd guess the design decision there was that if traffic to that host is THAT low, then adding the latency of a DNS lookup doesn't matter. On Tue, Mar 12, 2019 at 2:49 PM Nishant Gaurav wrote:

Question regarding hostdb

Hi, I was trying to debug an issue where a low QPS service's host keeps getting traffic even after it goes out of rotation. While debugging I observed that hostdb record for any given origin doesn't get updated unless trafficserver gets a call corresponding to that origin. This happen even when

Question about connection termination and hostdb

Hi, I had a question regarding timeout of a connection in traffic server. Let's say that the DNS name mentioned in `remap.config` to reach a downstream origin server has 5 hosts behind it (namely H1, H2, H3, H4, H5). Let's assume that the following sequence of events happen - at timestamp T, a

Re: connection attempts timeout difference b/w GET and POST requests in trafficserver

Disregard this mail, Susan had already answered this one. Thank you Nishant On Tue, Aug 21, 2018 at 5:18 PM Nishant Gaurav wrote: > Hi, > > I see that there is a large difference b/w the default value of timeout > for GET and POST (and PUT) requests (in seconds)

connection attempts timeout difference b/w GET and POST requests in trafficserver

Hi, I see that there is a large difference b/w the default value of timeout for GET and POST (and PUT) requests (in seconds). CONFIG proxy.config.http.connect_attempts_timeout INT 30 CONFIG proxy.config.http.post_connect_attempts_timeout INT 1800 I couldn’t think of a reason why it should be tha

Re: connection attempts timeout difference b/w GET and POST requests in trafficserver

Just nudging this to the top of your inbox. :) It will be of great help if anyone can help with this. On 8/20/18, 9:50 AM, "Nishant Gaurav" wrote: I was trying to debug a crash in trafficserver that happens when a server that only handles POST request goes down. I see that

connection attempts timeout difference b/w GET and POST requests in trafficserver

Hi, I was trying to debug a crash in trafficserver that happens when a server that only handles POST request goes down. I see that there is a large difference b/w the default value of timeout for GET and POST (and PUT) requests (in seconds). CONFIG proxy.config.http.connect_attempts_timeout INT

Re: logging

Just check if you have a LogObject defined in log_xml.config (wiki : https://docs.trafficserver.apache.org/en/6.2.x/admin-guide/monitoring/logging/log-formats.en.html#defining-log-objects) Nishant On 8/8/18, 12:02 PM, "Dk Jack" wrote: 6.2.1. Thanks for you help.

Re: Reloading remap.config without restarting trafficserver

inside ReverseProxy.cc will be called. Since this code is being run as a part of async thread, I don't mind it being blocking. Just wanted to get your thought on this one. Regards, Nishant On Fri, May 19, 2017 at 1:50 PM, Leif Hedstrom wrote: > > > On May 17, 2017, at 4:54 PM,

Reloading remap.config without restarting trafficserver

Hi everyone, I work as a part of Traffic team at LinkedIn. Our team has a use case where we want to reload ‘remap.config’ without restarting trafficserver. I understand that I can use ‘traffic_ctl config reload’ command for the same, but we don’t want dependence on an external process to run traf