Re: Memory leak at head?

Done: https://github.com/apache/trafficserver/issues/5042 On Wed, Feb 20, 2019 at 7:20 PM 宋辰伟 <616955...@qq.com> wrote: > Could you guys open issues. It would be more helpful to find the detail > than email. > > Thanks > Scw00 > > > 在 2019年2月21日，上午11:13，zzz 写道： > > > > a37f0d91ec0bbc5cd1f0700a69

Re: Memory leak at head?

Lol twist my arm amc ;D Leif's leak is different that what I'm seeing, mine is trivially reproduced but here is the ASAN result: https://paste.apache.org/6ttx On Wed, Feb 20, 2019 at 3:55 PM Alan Carroll wrote: > I don't think he's "coming" back, as far as I am concerned he's already > returne

Memory leak at head?

I just pulled down a fresh copy of master, did nothing special: autoreconf -i ./configure --prefix=/tmp/ats make -j12 && make install /tmp/ats/bin/traffic_server ab -n 1000 -c 1000 http://127.0.0.1:8080/ I made no changes to any config files other to to enable the memory debug output every 3

Re: TS-4523: Add the ability to pause/resume data consumption in the CPP API

long running or async action. I think we may need to schedule some time to get on IRC or a skype to just flush out some details. Brian On Wednesday, June 22, 2016, Brian Geffon wrote: > Is this analysis some horrifically long running thing that cannot happen > in the same thread? That mig

Re: TS-4523: Add the ability to pause/resume data consumption in the CPP API

Is this analysis some horrifically long running thing that cannot happen in the same thread? That might be the only reason I can think of for needed a pause... Brian On Wednesday, June 22, 2016, Brian Geffon wrote: > I think you might be misunderstanding certain continuation guarant

Re: TS-4523: Add the ability to pause/resume data consumption in the CPP API

data. For that we first need to produce all the file data > (Read chunks of it and consume it * N times), that's why we need to pause > the transformation we don't want to start buffering the data that keeps on > coming while we consume the file, when we are done consuming the

Re: TS-4523: Add the ability to pause/resume data consumption in the CPP API

The thing I'm missing about this, why can't you just keep buffering to the same file while you're doing analysis? I don't see why you need to pause anything? Brian On Friday, June 17, 2016, David Ben Zakai wrote: > Hi all, > > I'd like to suggest an API change in the CPP API Transformation inte

Re: client CA list in TS-3125

Hi James, Unfortunately, I don't fully remember. I believe it was based on an observation from the nginx guys around security related to session keys. Brian On Mon, May 2, 2016 at 6:50 AM, James Peach wrote: > Hi Brian, > > In TS-3125 you generate a server ID by hashing all the subject names in

Re: Unit testing for plugin

Take a look at the open pull requests, there are c++ API mocks that are hopefully going to land soon too. Brian On Friday, April 22, 2016, James Peach wrote: > > > On Apr 22, 2016, at 7:01 AM, Karthik Sivaraman < > karthiksivara...@outlook.com > wrote: > > > > Hi > > > > > > We are planning to

Re: [trafficserver] branch master updated: TS-3922: Websocket Timeouts, transaction overrideable & clang format

git > > > > The following commit(s) were added to refs/heads/master by this push: > > new 4fdf8ac TS-3922: Websocket Timeouts, transaction > overrideable & clang format > > 4fdf8ac is described below > > > > commit 4fdf8ac21e4dcac97297f6afb03c50a5c997

Re: [trafficserver] branch master updated: TS-4312 Add config to strictly parse URL according to RFC 3986. This closes #574

Will fix in a few hours. On Monday, April 18, 2016, James Peach wrote: > > > On Apr 18, 2016, at 1:05 AM, bri...@apache.org wrote: > > > > This is an automated email from the ASF dual-hosted git repository. > > > > briang pushed a commit to branch master > > in repository https://git-dual.apach

Re: merging github pull requests

I'm not thrilled with it either. I agree that the email addresses are definitely not desireable, also I don't like how it also produces a merge commit. On Friday, April 15, 2016, James Peach wrote: > Hi all, > > I just merged PR #563 using the "merge and squash" button on github. I'm > pretty un

Re: About ATS plugin to read post data

request. Brian On Thu, Apr 7, 2016 at 4:50 PM, Brian Geffon wrote: > I also have an open pull request : > https://github.com/apache/trafficserver/pull/351 that does exactly this, > as you've observed with transformations it's unable to block the downstream > connection an

Re: About ATS plugin to read post data

I also have an open pull request : https://github.com/apache/trafficserver/pull/351 that does exactly this, as you've observed with transformations it's unable to block the downstream connection and so this pull request exposes a new API to buffer the request body before allowing the downstream con

Re: https

It does whatever is in remap. So if you choose to make a request to /foo and you have a remap to https://www.foo.com/ that will be an outgoing HTTPS connection. Brian On Sunday, March 27, 2016, Dk Jack wrote: > Does atscppapi::AsyncFetch support output https connections? Haven't tried > it yet.

Re: Intercept Plugin

13, 2016 at 6:22 PM, Brian Geffon wrote: > > > Take a look at the example, you can set the body, status code, headers, > > basically anything you want. The only reason you might want to use an > > intercept plugin is it you wanted to stream a response. > > > > Br

Re: Intercept Plugin

seem to include body. I wanted to > include body as well... > > Bhasker. > > > On Sun, Mar 13, 2016 at 6:01 PM, Brian Geffon > wrote: > > > I would suggest looking at the custom error response plugin in the > examples > > folder, it will give you same result

Re: Intercept Plugin

I would suggest looking at the custom error response plugin in the examples folder, it will give you same result and it's much easier, Brian On Monday, March 14, 2016, Dk Jack wrote: > Hi, > I am using atscppapi intercept Plugin. Does it still require a remap rule > if I am using SERVER_INTERCE

Re: ts lua - http2

Yes it is. The H2 implementation uses the internal Fetch State Machine to make requests for a new stream. This means that you can hook into read request headers, send response headers, etc, or even create a transformation. Brian On Wednesday, February 10, 2016, Luca Rea wrote: > Hi, > > Does is

Re: Build with ASAN

Make sure to disable freelists too otherwise you'll miss out a number of potential issues. Also, I've always disabled lua_jit w/ ASAN. Brian On Wed, Jan 13, 2016 at 2:21 AM, Bryan Call wrote: > This is how I configure my bulids: > > With ASAN: > [bcall@homer ~]$ type tsconfig_asan > tsconfig_as

Re: Github vs Jira for comments

Just out of curiosity is the only way to close a github pull request still to just make a phantom commit? Brian On Wed, Dec 2, 2015 at 4:41 PM, Luca Rea wrote: > Hi, > about... "To answer Bcall’s email, yes, Github comments ends up on the > Jira, but not the other way around. So people reading

Re: Github vs Jira for comments

Just a constructive comment but I perosonslly am sort of sick of getting 3 emails for every GitHub comment :/ anything we can do about that regardless? Brian On Tuesday, December 1, 2015, Sudheer Vinukonda wrote: > It seems like there are pros and cons in both approaches - > While having the di

Re: Progressive download

It just sounds like he was using tsxs to build the plugin and linked with only 1 of 6 or so of the compilation units required. Compiling with -O0 shouldn't make a difference. Brian On Wed, Oct 7, 2015 at 2:54 PM, Nick Kew wrote: > On Wed, 2015-10-07 at 13:24 +0800, Brian Geff

Re: Progressive download

Sounds like it wasn't properly linked, it's missing the symbol BgFetchConfig::readConfig(const char *config_file) Which is configs.cc. I checked Makefile.am for background_fetch and it appears that it should be compiling and linking properly. Did you build this plugin in a strange way, for exampl

Re: Req. Info Gzip

FWIW, there is also a copy of the gzip plugin available w/ the C++ api. Brian On Wed, Sep 30, 2015 at 9:30 PM, Leif Hedstrom wrote: > > On Sep 30, 2015, at 2:05 AM, Luca Rea wrote: > > Hi, > Does is possible to compress “on-the-fly” and store in cache gzipped > version (and only that) of pushe

Re: active timeouts broken

I'm suspecting: https://git-wip-us.apache.org/repos/asf?p=trafficserver.git;a=commitdiff;h=974e8e3abc9a3f60f516b24302edde5618a3cfc5 from TS-3313: https://issues.apache.org/jira/browse/TS-3313 Brian On Mon, Sep 21, 2015 at 4:22 PM, Brian Geffon wrote: > All, > While adding ne

active timeouts broken

All, While adding new websocket specific timeouts it seems that active timeouts are completely broken on master. I quickly tested and verified that they are working properly as of 5.2.x (which is the branch we're on over here), so somewhere since then a huge regression was introduced...does anyone

Re: TLS Session Ticket: Key Rotation

Bump, any last feedback regarding my previous email? On Wed, Aug 12, 2015 at 1:54 PM, Brian Geffon wrote: > I'd like to close the loop on this discussion. In general I believe there > is a consensus that perhaps ssl_multicert is not the place to deal with > ticket rotation and

Re: TLS Session Ticket: Key Rotation

I'd like to close the loop on this discussion. In general I believe there is a consensus that perhaps ssl_multicert is not the place to deal with ticket rotation and that if you're willing to have global session tickets (meaning not tied to a specific domain) then the implementation that would acco

Re: Proposal: Consistent hashing DNS selection

t 12:10 PM, Phil Sorber wrote: > On Sun, Aug 9, 2015 at 9:10 PM Brian Geffon wrote: > > > All, > > We currently support strict and timed round robin for round robin DNS > > records and I'd like to propose adding consistent hashing based DNS > > selection. We&#x

Proposal: Consistent hashing DNS selection

All, We currently support strict and timed round robin for round robin DNS records and I'd like to propose adding consistent hashing based DNS selection. We're trying to replace our current ATS -> Haproxy setup with only ATS and that's been the motivation behind all of our recent HostDB improvement

Re: Binary body in plugin of traffic server

If you attach a transformation plugin the std::string will be populated with whatever the body is, even if it's binary data. You'll need to use .size()/.length() to determine the buffer size and then just make sure you do not use string functions if it's binary data (check the content-type). Brian

Re: [GitHub] trafficserver pull request: Maintain and use a mapping of hostname...

I agree that this is the preferred pattern given this situation, the message is very clear: this object is no longer in use and then when it's assigned again at the start it's clear it's in use again. On Mon, Jul 6, 2015 at 9:51 PM, SolidWallOfCode wrote: > Github user SolidWallOfCode commented

Re: Concurrency Kit Future

I like the idea of CK++, I'd love to see this, my only question is: who is officially going to be the maintainer of CK++, will it be the CK community? Also, would CK++ require C++11/14 or would it try to be backwards compatible with older versions of C++? Brian On Tue, Jun 16, 2015 at 11:57 AM, P

Re: Removing CHANGES file from git

+1 I've been doing separate commits to changes to make cherry picks easier, let's nuke it. Brian On Friday, June 5, 2015, Thomas Jackson wrote: > +1 on killing manual CHANGES file :D > > On Fri, May 29, 2015 at 7:58 AM, Alan Carroll < > solidwallofc...@yahoo-inc.com.invalid > > > wrote: > >> +1

Re: Adding Gzip/Gunzip feature in ATS core

ll zip the body if > necessary. > > > > -Dzmitry > > > > ____ > > From: gang li [portl4t...@gmail.com ] > > Sent: Tuesday, March 24, 2015 6:38 AM > > To: dev@trafficserver.apache.org > > Cc: Roland Zink; Brian Gef

SSL hook ordering

Hi, When looking at apidefs ( https://git-wip-us.apache.org/repos/asf?p=trafficserver.git;a=blob;f=lib/ts/apidefs.h.in;h=329adf27f9d0fc3fc677c21d5be38e39456dab73;hb=HEAD#l291) the SSL hooks appear to be at the end of the hook list. While this isn't a huge problem, in some situations it's helpful to

Re: Performance tuning advice for plugin

I've also observed unexplained latency when it comes to transformations, I think it's time that we dig into this more. The reason we're observing an increase in latency without a corresponding increase in CPU load is because TS simply isn't doing anything, it appears that it's just rescheduling tra

Re: Adding headers to C++ api AsyncHttpFetch Streaming request.

Take a look at line 91 in the following example: https://git1-us-west.apache.org/repos/asf?p=trafficserver.git;a=blob;f=lib/atscppapi/examples/async_http_fetch/AsyncHttpFetch.cc Brian On Thu, Feb 26, 2015 at 4:43 PM, Sandeep Davu wrote: > Hi, > > How do you add headers to a request you want to

Re: Sending chunked data on intercept plugins

You'll have to handle chunking yourself; it should be relatively simple to do. On Wed, Feb 11, 2015 at 2:22 PM, Sandeep Davu wrote: > Who will be responsible for chunking? Plugin has access to response and not > chunked data. If Intercept writes a raw byte stream, how does the data get > chunke

Re: [VOTE] Release Apache Traffic Server 4.2.3 (RC0)

Builds and passes regressions on CentOS 6.5 (2.6.32-431.29.2.el6.x86_64) so +1 from me. Brian On Mon, Jan 26, 2015 at 12:36 PM, Phil Sorber wrote: > Hello All, > > I've prepared a release for v4.2.3 (RC0) which is the latest stable > release in the 4.2.x series. This is the third release in our

Re: atscppapi AsyncHttpFetch

Unfortunately you cannot access the state_ member because it's private in AsyncHttpFetch, I think the best option here might be for us to add a getter method for the dispatch controller. Alternatively, you can just skip inheriting from AsyncHttpFetch and implement the AsyncProvider interface and co

Re: Testing Framework Meetup

All, just a reminder if you're coming to discuss testing please be there at 11.30 and ping me if you have any questions. It's the same building the summit was held in, 580 N Mary, Sunnyvale Brian On Thursday, January 15, 2015, Brian Geffon wrote: > From the voting it sounds lik

Re: Testing Framework Meetup

n. Feel free to ping me with any questions. > > Brian > > > On Thu, Jan 15, 2015 at 10:38 AM, Brian Geffon > wrote: > >> All, please vote on date/times at the following doodle: >> http://doodle.com/t3vss2zdz6wyhez9 , I'll send out an email with the &

Re: Testing Framework Meetup

-- Meera > > > > On Jan 14, 2015, at 2:29 PM, Brian Geffon wrote: > > > > Hi All, > > Thomas and Josh have made some pretty good progress on a testing > framework, > > I think it's probably time for another mini meetup to see where we're at &

Testing Framework Meetup

Hi All, Thomas and Josh have made some pretty good progress on a testing framework, I think it's probably time for another mini meetup to see where we're at and get everyone caught up. Would anyone be interested in meeting somewhere in sunnyvale or mountain view early next week? Brian

Re: [VOTE] Release v5.2.0

I forgot to mention that I +1 this release in my previous email. Brian On Wednesday, December 17, 2014, Leif Hedstrom wrote: > > > On Dec 16, 2014, at 3:52 PM, Leif Hedstrom > wrote: > > > > Hi all, > > > > I've prepared a release candidate for v5.2.0, which is primarily a bug > fix release ov

Re: [VOTE] Release v5.2.0

Fixed in master with commit 8993142d44253ded76b132535b6886f971483b9f. I've verified the build on CentOS 6 w/ gcc 4.9.2 all regressions also pass. Brian On Tue, Dec 16, 2014 at 9:07 PM, Brian Geffon wrote: > > I'll take a look. > > Brian > > > On Tuesday, December

Re: [VOTE] Release v5.2.0

#x27;__sync_val_compare_and_swap’ > >ink_queue.cc:338: error: 'struct _InkFreeList' has no member > named 'head’ > > > > Confirmed, it’s related to reclaimable freelist and this commit: 0cdc3cb4 > > > Brian Geffon: Can you take a look please? > > — Leif > >

Re: trafficserver git commit: Fixing AddressSanitizer issue where sizeof(size_t) > sizeof(int).

t; Diff: http://git-wip-us.apache.org/repos/asf/trafficserver/diff/90b63548 > > > > Branch: refs/heads/master > > Commit: 90b635481aa9a819852fbfeafdf334bf06513181 > > Parents: 74d29be > > Author: Brian Geffon > > Authored: Thu Nov 20 11:53:24 2014 -0800 > &g

Re: ATS Integration Testing

; > On 11/4/14, 10:54 AM, "Narayan Balasubramanian (நாராயண் )" > > wrote: > > > > >Brian, > > > > > >Agree with Josh. You have captured the requirements well. > > > > > >AFAICT, most of the contributors are local - LinkedIn, Ap

Re: [1/2] git commit: TS-2986: Adding stats to TLS errors

at 4:15 PM, James Peach wrote: > > > On Aug 6, 2014, at 10:20 AM, James Peach wrote: > > > > On Aug 5, 2014, at 12:16 PM, Brian Geffon wrote: > > > >> Technically you'd probably aggregate all of these into a single variable > >> for production

[REVIEW REQUEST] ClassAllocator Improvements

All, Cynthia Gu has put together a great patch that improves the ClassAllocators. Firstly she added a configurable low watermark for ClassAllocators that prevent it from returning all its items whenever it exceeds the configurable limit. Additionally, in the current implementation it returns items

Re: [API REVIEW] TS-1432 API is missing TSMutexDestroy

Alan, Take a look at the final commit, we're on the same page. Brian On Thu, Nov 6, 2014 at 1:38 PM, Alan M. Carroll wrote: > We are simply incapable of doing anything in a simple way. > > I see James' point but I disagree with his implementation. > > I would recommend setting the refcount to 1

[API REVIEW] TS-1432 API is missing TSMutexDestroy

This should be self explanatory and IMO we should land for 5.2, the following patch is also attached to TS-1432. Brian diff --git a/proxy/InkIOCoreAPI.cc b/proxy/InkIOCoreAPI.cc index 26f3298..091d7a2 100644 --- a/proxy/InkIOCoreAPI.cc +++ b/proxy/InkIOCoreAPI.cc @@ -189,6 +189,11 @@ TSMutexCre

ATS Integration Testing

Hi All, thanks for your patience I know many people are eager to start pooling resources to make this happen. (Thanks Susan for helping with notes during the summit) To briefly summarize what was discussed at the summit, we have an existing framework called TSQA which is based on bash, while this

Re: Working on functional test framework

Sorry guys I've been traveling and haven't had the chance to start the discussions on this. We more or less have our requirements as gathered at the summit I'll try to send them out today so we can begin discussing a roadmap. Brian On Monday, November 3, 2014, Joshua Blatt wrote: > Good feedbac

ATS Summit Streaming Link

All, To stream the summit for day two please use the following link: https://bluejeans.com/498670616/browser -- Brian

Summit Stream

Hi All, The summit should start around 9AM PDT (-8 GMT), you can remotely participate using the following link: https://bluejeans.com/157121879/browser If you have any issues just ask on irc. Brian

Re: POODLE and ATS configs

cc: users@ For users who want to immediately disable SSLv3 you should only need to change proxy.config.ssl.SSLv3 in records.config to 0 and bounce traffic_server. Brian On Tue, Oct 14, 2014 at 4:13 PM, Leif Hedstrom wrote: > Now that the POODLE is out of the bag, I think we should consider cha

Re: [3/3] git commit: TS-3127: Update documentation

This setting has no affect on the ATS session cache implementation, only the OpenSSL implementation. But you're correct in your assessment. So with this option someone will need to force flush. This is something we plan on exposing via a new API. "Normally the session cache is checked for expired

Re: C++ API shared_ptr dependecy

James, I think the problem of requiring std::shared_ptr is that it wont work with some older compilers without adding --std=c++0x, such as g++ 4.4.x Brian On Mon, Oct 6, 2014 at 8:21 AM, James Peach wrote: > On Oct 3, 2014, at 9:06 PM, Brian Geffon wrote: > > > Hey all, >

Re: C++ API shared_ptr dependecy

ink_autoconf.h? namespace tr1 { } namespace std_tr1 { using namespace std; using namespace tr1; } namespace atscppapi { using std_tr1::shared_ptr; } Brian On Fri, Oct 3, 2014 at 9:03 PM, Brian Geffon wrote: > Hey all, > So there is a weird situation we're in with the c++ api where it h

C++ API shared_ptr dependecy

Hey all, So there is a weird situation we're in with the c++ api where it has a dependency on shared_ptr which can exist in memory or tr1/memory and similarly in the namespace std:: and std;:tr1::, so in the cppapi's shared_ptr.h file we have the following mess: #include "ink_autoconf.h" #if HAVE

Re: [CFP and ANNOUNCE] Fall 2014 ATS Summit

All, To make things easier I've created a Google Form that you can use to RSVP for the summit: *http://goo.gl/forms/f4MeC4osyr *. Brian On Wed, Oct 1, 2014 at 10:06 AM, Leif Hedstrom wrote: > Hi all, > > we’re looking for topics and presentations for the Summit

Re: [CFP and ANNOUNCE] Fall 2014 ATS Summit

Would you mind co-chairing with James Peach? I think together you guys would host a great session. Brian On Wednesday, October 1, 2014, Ryo Okubo wrote: > Hi, > > I want to talk about around HTTP/2 feature that’s proposed at TS-2729: > > * Current progress of my HTTP/2 implementation and future

Re: Minimum OpenSSL version

Ok, so I just found this in SSLUtils.cc: #if (OPENSSL_VERSION_NUMBER < 0x00090400L)# error Traffic Server requires an OpenSSL library version 0.9.4 or greater #endif On Sun, Sep 21, 2014 at 5:31 PM, James Peach wrote: > > On Sep 21, 2014, at 5:23 PM, Brian Geffon wrote: >

Minimum OpenSSL version

Hi All, This is something that apparently has never been brought up, we don't have a strict OpenSSL minimum version. By explicitly stating our minimum supported OpenSSL version we can clean up our SSL code, specifically around the SNI #ifdefs. I'm going to propose that we make *0.9.8f* our minimum

Pending SSL changes?

Does anyone have pending SSL changes? In the next few days I'm going to land a patch that adds SSL session caching that doesn't use OpenSSL's session cache. I'd prefer to not have a massive conflict. I know Susan has been working on some SSL fixes, any idea when that will make it in? Brian

Re: Cache producer with non-NULL vc and NULL buffer_start?

I wonder if this is related to: https://issues.apache.org/jira/browse/TS-2497 On Wed, Sep 17, 2014 at 6:32 AM, Acácio Centeno wrote: > Folks, > > I have a situation that happens about once a week and causes a segfault on > ATS (5.0.1). The stack trace always shows Ptr::operator= and > the last f

Re: [1/2] git commit: [TS-3069] Add mysql_remap to autoconf

> http://git-wip-us.apache.org/repos/asf/trafficserver/commit/ae60ad7a > > Tree: http://git-wip-us.apache.org/repos/asf/trafficserver/tree/ae60ad7a > > Diff: http://git-wip-us.apache.org/repos/asf/trafficserver/diff/ae60ad7a > > > > Branch: refs/heads/master > > Com

Re: TSMimeHdrFieldNext

rove for looping over all fields ? Or even a new API > with functor semantics? > > TSMimeHdrFieldGet at least avoids one loop to avoid finding the slotnum > > > On Aug 16, 2014, at 5:29 AM, Brian Geffon wrote: > > > > If no one has already started on a fix I can ta

Re: TSMimeHdrFieldNext

If no one has already started on a fix I can take this. Brian On Aug 16, 2014 6:13 AM, "Nick Kew" wrote: > > On 15 Aug 2014, at 21:55, James Peach wrote: > > > What's a real use case for iterating over all the headers? > > Ironbee. Or any WAF, or WAF framework. > > Or performance analysis tools

Re: [1/2] git commit: TS-2986: Adding stats to TLS errors

gt; > Tree: http://git-wip-us.apache.org/repos/asf/trafficserver/tree/a8070bbb > > Diff: http://git-wip-us.apache.org/repos/asf/trafficserver/diff/a8070bbb > > > > Branch: refs/heads/master > > Commit: a8070bbb80f2e709f91d4c8b47d9ded4d55bdcdc > > Parents: b434317 &

Re: ATS 5.0.0 issue with TSHttpConnect

Sounds like a bug related to spdy chunking changes, possibly I'm fetchsm. On Wednesday, June 25, 2014, Walsh, Peter wrote: > Hello, > I just upgraded from 4.2.1 to ATS 5.0.0 but am having issues in my plugins > that use TSHttpConnect. > > When the response body is chunked, the response body bei

Re: Lifetime of a Continuation

I think the one we were seeing was slightly different, nick would you mind sharing an example of what you changed with a simple transformation plugin? On Thu, Jun 19, 2014 at 8:20 AM, Alan M. Carroll < a...@network-geographics.com> wrote: > Nick, > > You should talk to Brian G. He had a bug last

Re: [VOTE] Release Apache Traffic Server 5.0.0 (RC0)

+ 1, on RHEL 6.4. On Friday, June 13, 2014, Igor Galić wrote: > +1 compiles and runs fine on Ubuntu 12.04 (old LTS) on amd64 in forward > proxy mode > (will soon test as reverse proxy, if it didn't happen already > automatically ;) > > o/~ i > > - Original Message - > > I have updated 5.

Re: [API REVIEW] TSHttpIsInternalSession

Makes sense +1. Brian On Wed, May 28, 2014 at 9:12 PM, James Peach wrote: > Hi all, > > I'd like to add a session-oriented counterpart to TSHttpIsInternalRequest, > named TSHttpIsInternalSession: > > TSReturnCode TSHttpIsInternalSession(TSHttpSsn ssnp); > > This API has identical seman

Re: [RFC] coding style for member variable access

I think this-> is incredibly ugly an unnecessary, what does it really get us? It will be become convention just as appending a underscore or adding m_ in front of the name. In my opinion this-> should only be used in situations where you need to disambiguate between variable names and not as a conv

Re: git commit: TS-2821 Add to default records.config and tweak the name to be more consistent.

In response to this default value, most spdy servers I've seen use 256 for this value. Take a look at net internals in chrome. Brian On Monday, May 19, 2014, Leif Hedstrom wrote: > Agree with James, let's be consistent. > > -- Leif > > > On May 19, 2014, at 8:45 PM, James Peach wrote: > > > >>

Re: [VOTE] Release Apache Traffic Server 4.2.1 (RC1)

+1 RHEL 6.4 On Monday, April 21, 2014, Adam W. Dace wrote: > P.S. My apologies for not actually running these releases on my personal > cache, but I'm finishing up with more tuning work. On the upside, ATS is > doing an amazing job on my now 100Mbps link. > > > On Mon, Apr 21, 2014 at 11:37 AM

Re: [VOTE] Release Apache Traffic Server 4.2.1 (RC0)

+1 on RHEL6.4 On Sat, Apr 12, 2014 at 6:14 PM, Phil Sorber wrote: > Hello All, > > I've prepared a release for v4.2.1 (RC0) which is the latest stable release > in the 4.2.x series. This is the second release in our Long Term Support > (LTS) version as detailed in our Release Management documen

Re: Importing libck

Phil is correct, while discussing ck and the architectures it supports we proposed dropping 32bit support independent of ck. Brian On Saturday, April 12, 2014, Phil Sorber wrote: > On Sat, Apr 12, 2014 at 2:00 PM, James Peach > > wrote: > > > On Apr 11, 2014, at 11:34 AM, Phil Sorber > > wrote:

Re: Importing libck

+ 1 On Friday, April 11, 2014, Leif Hedstrom wrote: > +1 > > -- Leif > > On Apr 11, 2014, at 12:34 PM, Phil Sorber > > wrote: > > > I'd like to propose that we pull libck into our tree and use it to > replace > > some of our stuff like the freelist, ink_atomic_list and hash tables. > > > > http:

Re: Proposal to remove 32bit support.

This seems like a logical move for a 5.0 release, +1 from me. On Fri, Apr 11, 2014 at 8:51 AM, Theo Schlossnagle wrote: > Given the nature of ATS and its focus exclusively on high performance > environments, I suggest we throw off the bonds of 32bit support going > forward. I'm unaware of anyon

ATS summit dinner

Hi All, We're doing a sponsored dinner tonight at Wynkoop Brewery, at 18th and Wynkoop, at 7pm. It would be great to see you all there. Brian

Re: Question about developing plugins using the C++ API

Hi, This sounds like a bug in the C++ API, would you mind creating a Jira ticket and assigning it to me? Thanks. Brian On Tue, Apr 1, 2014 at 6:36 AM, Acácio Centeno wrote: > Hello, > > We're creating a plugin that blocks/redirects some requests based on its > properties. We're using the C++ AP

Re: Remap/regex_remap consolidation

not be > premature optimization) is that if you have seperate regexes we can create > hash maps similar to how maps work (a hashmap of domain_regex -> list of > path regexes) which would make overall remap performance faster/better. > > > On Tue, Mar 25, 2014 at 9:50 PM, Brian Geffo

Re: Remap/regex_remap consolidation

Right. On Tuesday, March 25, 2014, Leif Hedstrom wrote: > > On Mar 25, 2014, at 7:51 PM, Brian Geffon > > > wrote: > > > What Thomas called Question #1 -- 1 Regex. > > > > Makes sense to have them combined. Assuming groups etc. works, that allows &g

Re: Remap/regex_remap consolidation

What Thomas called Question #1 -- 1 Regex. Brian On Tuesday, March 25, 2014, Leif Hedstrom wrote: > > On Mar 25, 2014, at 7:42 PM, Brian Geffon > > wrote: > > > I personally vote for regexing everything until we get around to > finalizing > > lua config, I also v

Re: Remap/regex_remap consolidation

I personally vote for regexing everything until we get around to finalizing lua config, I also vote for this being the one who will end up writing the code :) Brian On Tuesday, March 25, 2014, James Peach wrote: > On Mar 25, 2014, at 3:28 PM, Leif Hedstrom > > wrote: > > > > > On Mar 25, 2014,

Re: [VOTE] Release Apache Traffic Server 4.2.0 (RC1)

+1, RHEL 6.4. Brian On Fri, Mar 7, 2014 at 5:51 PM, Phil Sorber wrote: > Hello All, > > It's been a while since RC0, and we've tracked down quite a few bugs. So > I've prepared a new release for v4.2.0 (RC1) which is the last stable > release in the 4.x series. This will serve as our Long Term

Stack traces across scheduled events

Hi all, There has been discussion on IRC about techniques for following requests across events. I've slapped together a simple patch that attaches backtraces to Continuations every time an event is scheduled. Since some people may find this useful I wanted to "archive" it via the mailing list. If i

Re: git commit: TS-2579: remove ipv4 limit on FetchSM and TSFetchUrl/TSFetchPages

should we have jira open to fix them? > > > > yes probably > > > >>> On Sun, Feb 23, 2014 at 3:02 PM, James Peach > wrote: > >>> > >>>> On Feb 23, 2014, at 10:47 AM, Brian Geffon wrote: > >>>> > >>>> Doesn

Re: git commit: TS-2579: remove ipv4 limit on FetchSM and TSFetchUrl/TSFetchPages

Doesn't the http connect parse headers anyway? Isn't the address only used for local host proxying? Brian On Sunday, February 23, 2014, Shu Kit Chan wrote: > ESI plugin unfortunately still uses TSFetchUrl and therefore cannot handle > ipv6 client addresses. > > > On Sun, Feb 23, 2014 at 8:19 AM

Re: [API Review] Add lifecycle hook for SSL_CTX initialization

Do you guys have any example plugins that use these hooks? Brian On Thursday, February 13, 2014, Wei Sun wrote: > Hi all, > > I'd like to add two new lifecycle hooks, > TS_LIFECYCLE_SERVER_SSL_CTX_INITIALIZED_HOOK and > TS_LIFECYCLE_CLIENT_SSL_CTX_INITIALIZED_HOOK, as discussed on this jira: >

Re: [VOTE] Release Apache Traffic Server 4.2.0 (RC0)

+1 from me. Brian On Mon, Feb 10, 2014 at 2:47 PM, Otto van der Schaaf wrote: > +1 > > Sent from my Windows Phone > -- > From: Phil Sorber > Sent: 4-2-2014 23:00 > To: us...@trafficserver.apache.org; dev@trafficserver.apache.org > Subject: [VOTE] Release Apache T

Re: WebSocket support

n checks (they would still be > after the presence checks). It also makes more logical sense to put all > prerequisites in one statement. > > Cheers, > > -- Leif > > > On Jan 31, 2014, at 5:48 PM, Brian Geffon > > wrote: > > > > See responses below. > &

Re: WebSocket support

See responses below. On Fri, Jan 31, 2014 at 4:05 PM, Leif Hedstrom wrote: > > On Jan 29, 2014, at 6:22 PM, Brian Geffon wrote: > > > Hi All, > > I've created a patch adding WebSocket support to ATS, I would appreciate > > community feedback. This is being

WebSocket support

Hi All, I've created a patch adding WebSocket support to ATS, I would appreciate community feedback. This is being tracked in TS-2541, the patch is attached to the jira ticket https://issues.apache.org/jira/browse/TS-2541 Thanks! Brian

  1   2   3   >