Github user acaciocenteno closed the pull request at:
https://github.com/apache/trafficserver/pull/106
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the f
Github user SolidWallOfCode commented on the pull request:
https://github.com/apache/trafficserver/pull/106#issuecomment-56677407
Yes, putting in the Machine instance sounds good. But it must be
non-persistent across process restarts, because the SM id isn't. The goal is
to have a UU
Github user zwoop commented on the pull request:
https://github.com/apache/trafficserver/pull/106#issuecomment-5661
On Sep 23, 2014, at 10:22 AM, Alan M. Carroll
wrote:
> Something that came up with Acacio is rather than doing the UUID for each
transaction, we do it
Github user SolidWallOfCode commented on the pull request:
https://github.com/apache/trafficserver/pull/106#issuecomment-56547180
Something that came up with Acacio is rather than doing the UUID for each
transaction, we do it for the traffic_server process and make it possible to
log
Github user acaciocenteno commented on the pull request:
https://github.com/apache/trafficserver/pull/106#issuecomment-54376276
Thank you for your comments, as I'm just starting to hack with ATS code,
there're certainly many aspects I'll have to learn yet.
But, all in all, do
Github user SolidWallOfCode commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/106#discussion_r17050782
--- Diff: proxy/http/HttpSM.cc ---
@@ -371,6 +392,107 @@ HttpSM::destroy()
httpSMAllocator.free(this);
}
+#ifdef TS_HAS_UU
Github user marcusgrando commented on the pull request:
https://github.com/apache/trafficserver/pull/106#issuecomment-54241232
About the X- headers, maybe you guys can agree about one debug header like
X-Debug or another one and it can be used on the xdebug plugin too. Maybe a new
API
Github user zwoop commented on the pull request:
https://github.com/apache/trafficserver/pull/106#issuecomment-54227923
Oh, and almost certainly, don't use std:string in the HttpSM.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub
Github user zwoop commented on the pull request:
https://github.com/apache/trafficserver/pull/106#issuecomment-54227662
Also, I'm not keen on naming headers like x-ats-uuid (we should not
introduce more X- headers into the wild).
---
If your project is set up for it, you can reply t
Github user zwoop commented on the pull request:
https://github.com/apache/trafficserver/pull/106#issuecomment-54225453
But, if we exposed the http sm_id as both a log tag and plugin API, you
could do whatever you want? You could generate your own "UUID" based on the
sm_id? I'm assumi
Github user acaciocenteno commented on the pull request:
https://github.com/apache/trafficserver/pull/106#issuecomment-54223450
The problem with HttpSM::sm_id is that it's just a counter, and as so it's
almost impossible to grep for it... the UUID, though, is guaranteed to be
unique.
Github user acaciocenteno commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/106#discussion_r17018609
--- Diff: proxy/http/HttpSM.cc ---
@@ -371,6 +392,107 @@ HttpSM::destroy()
httpSMAllocator.free(this);
}
+#ifdef TS_HAS_UUID
Github user SolidWallOfCode commented on the pull request:
https://github.com/apache/trafficserver/pull/106#issuecomment-54215944
`HttpSM::sm_id` does not appear to be exposed to the plugin API, and I see
no reason that would be a problem. The example use cases are all from a single
r
Github user zwoop commented on the pull request:
https://github.com/apache/trafficserver/pull/106#issuecomment-54215437
Yeah, I was just about to comment the same. the sm_id is a sequence number
though, and is not guaranteed unique across ATS instances.
---
If your project is set up
Github user SolidWallOfCode commented on the pull request:
https://github.com/apache/trafficserver/pull/106#issuecomment-54215168
How is this different than `HttpSM::sm_id`?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well.
Github user bgaff commented on the pull request:
https://github.com/apache/trafficserver/pull/106#issuecomment-54208758
It appears there is no code in here to detect a previously set UUID, which
means that each tier will generate its own UUID, it would be nice if the
downstream could
Github user bgaff commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/106#discussion_r17011490
--- Diff: proxy/http/HttpSM.cc ---
@@ -371,6 +392,107 @@ HttpSM::destroy()
httpSMAllocator.free(this);
}
+#ifdef TS_HAS_UUID
+vo
Github user bgaff commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/106#discussion_r17011520
--- Diff: proxy/http/HttpSM.cc ---
@@ -371,6 +392,107 @@ HttpSM::destroy()
httpSMAllocator.free(this);
}
+#ifdef TS_HAS_UUID
+vo
Github user bgaff commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/106#discussion_r17011136
--- Diff: proxy/http/HttpSM.cc ---
@@ -371,6 +392,107 @@ HttpSM::destroy()
httpSMAllocator.free(this);
}
+#ifdef TS_HAS_UUID
+vo
GitHub user acaciocenteno opened a pull request:
https://github.com/apache/trafficserver/pull/106
Added Unique Identifier to HttpSM, to aid on debugging
While debugging the plugins we've written, we sometimes had trouble
correlating the log messages to one specific request. That
20 matches
Mail list logo
