Github user zwoop closed the pull request at:
https://github.com/apache/trafficserver/pull/763
---
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 feature
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/763
This might be above my Git skill level. I did try to pull the PR into my
tree, but it's so far behind that it made some crazy changes / conflicts :-/.
Manually applying the "diff&quo
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/763
I resolved the conflicts, I think:
http://paste.fedoraproject.org/389174/80381301/
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/763
It does have merge conflicts, so maybe I should resolve that locally, and
commit it with --author=susan ? @PSUdaemon wdyt?
---
If your project is set up for it, you can reply to this email
Github user zwoop closed the pull request at:
https://github.com/apache/trafficserver/pull/790
---
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 feature
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/763
@shinrich @masaori335 Should we land this patch though? I'm landing my
other fix for the active client conns.
---
If your project is set up for it, you can reply to this email and have
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/770
:+1:
---
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
GitHub user zwoop opened a pull request:
https://github.com/apache/trafficserver/pull/790
TS-4472 Fixes http_current_active_client_connections_stat metric
The issue here is that there are two m_active member variables, in
both Http1ClientSession and Http1ClientTransaction
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/789
:+1:
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/788
:+1:
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/788
Hmmm, this code looks overly complicated already to begin with :-/. Why is
RefCountCacheSerializer a template class? It's tightly coupled with HostDB it
seems, and is only used to co
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/763
Some sort of odd race condition here maybe. If I run single requests using
curl, I'm not seeing the problem. But if I blast it with "ab -c 10 -n 1"
the counter permanen
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/763
So, I tried this patch, and it does not fix the problem I'm running into,
where proxy.process.http.current_active_client_connections is seemingly never
decremented. I had a merge conflict
Github user zwoop commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/763#discussion_r69666098
--- Diff: proxy/http/HttpSM.cc ---
@@ -200,7 +200,8 @@ HttpVCTable::cleanup_entry(HttpVCTableEntry *e)
// Update stats
switch (e
Github user zwoop commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/783#discussion_r69650806
--- Diff: proxy/http/HttpTransact.cc ---
@@ -63,7 +63,12 @@ static char range_type[] = "multipart/byteranges;
boundary=RANGE_SEPA
Github user zwoop commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/783#discussion_r69649977
--- Diff: proxy/http/HttpTransact.cc ---
@@ -63,7 +63,12 @@ static char range_type[] = "multipart/byteranges;
boundary=RANGE_SEPA
Github user zwoop commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/783#discussion_r69647327
--- Diff: proxy/StatPages.cc ---
@@ -83,7 +83,7 @@ StatPagesManager::handle_http(Continuation *cont, HTTPHdr
*header)
host_len
Github user zwoop commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/783#discussion_r69640193
--- Diff: proxy/StatPages.cc ---
@@ -83,7 +83,7 @@ StatPagesManager::handle_http(Continuation *cont, HTTPHdr
*header)
host_len
Github user zwoop commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/783#discussion_r69611852
--- Diff: proxy/http/HttpTransact.cc ---
@@ -762,8 +762,8 @@ HttpTransact::StartRemapRequest(State *s)
const char syntxt[] = "syntheti
Github user zwoop closed the pull request at:
https://github.com/apache/trafficserver/pull/743
---
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 feature
Github user zwoop commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/772#discussion_r69379523
--- Diff: proxy/api/ts/ts.h ---
@@ -2336,6 +2336,35 @@ tsapi uint64_t TSHttpTxnIdGet(TSHttpTxn txnp);
tsapi TSReturnCode TSBase64Decode(const char
Github user zwoop closed the pull request at:
https://github.com/apache/trafficserver/pull/777
---
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 feature
Github user zwoop commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/777#discussion_r69379428
--- Diff: tools/pre-commit ---
@@ -0,0 +1,66 @@
+#! /usr/bin/env bash
+#
+# Simple pre-commit hook script, enforcing clang-format on our
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/782
Oh I see, the 6.2.0 PR is in #781. Great.
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/781
:+1:
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/782
:+1:
Lets back port this to 6.2.0 as well, since we have to re-roll because of
the other fix.
---
If your project is set up for it, you can reply to this email and have your
reply
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/784
:+1:
---
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
Github user zwoop closed the pull request at:
https://github.com/apache/trafficserver/pull/775
---
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 feature
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/775
[approve ci]
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/765
Yeah, maybe we should close this PR ?
---
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
Github user zwoop closed the pull request at:
https://github.com/apache/trafficserver/pull/778
---
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 feature
Github user zwoop closed the pull request at:
https://github.com/apache/trafficserver/pull/747
---
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 feature
GitHub user zwoop opened a pull request:
https://github.com/apache/trafficserver/pull/778
TS-4625 epic: Adds a mutex to the continutation
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/zwoop/trafficserver TS-4625
Alternatively
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/775
It seems there are merge conflicts here though, Tyler, can you please
rebase your tree (which probably means pushing up the main master into your
fork first) ?
---
If your project is set up
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/775
Yeah, this seems right, TSEventDestroy() checks its argument, and we use it
like this in several other places (which is likely why coverity gets confused).
Consistency is good!
---
If your
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/776
I'll have to review this properly, but an initial thought is that maybe
this should be part of the gzip plugin? Maybe that plugin gets renamed to
"compression" or something? It h
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/770
Unless there's a penalty (or error) to always use $(sort ...), I'd
recommend doing so consistently. That avoids future kids from having to know
when, or when not to, use it.
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/747
wth, how did FreeBSD core dump ... [approve ci]
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/777
Need input from the git expert, too bad @bgaff doesn't have a computer
still :-/. So will have to settle for @jpeach and @PSUdaemon .
---
If your project is set up for it, you can rep
GitHub user zwoop opened a pull request:
https://github.com/apache/trafficserver/pull/777
TS-4623 Adds a pre-commit script for git
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/zwoop/trafficserver TS-4623
Alternatively you
Github user zwoop commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/747#discussion_r69205590
--- Diff: plugins/experimental/remap_purge/remap_purge.c ---
@@ -202,34 +202,38 @@ handle_purge(TSHttpTxn txnp, PurgeInstance *purge)
bool
Github user zwoop commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/747#discussion_r69167535
--- Diff: plugins/experimental/remap_purge/remap_purge.c ---
@@ -0,0 +1,335 @@
+/** @file
+
+ Per-remap purge RESTful API for stateful
Github user zwoop commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/747#discussion_r69163153
--- Diff: plugins/experimental/remap_purge/remap_purge.c ---
@@ -0,0 +1,335 @@
+/** @file
+
+ Per-remap purge RESTful API for stateful
Github user zwoop commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/747#discussion_r69163780
--- Diff: plugins/experimental/remap_purge/remap_purge.c ---
@@ -0,0 +1,335 @@
+/** @file
+
+ Per-remap purge RESTful API for stateful
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/723
@jpeach Hmmm, what happened here, did we not land this? I have lost my
mind, and do not remember ...
---
If your project is set up for it, you can reply to this email and have your
reply
Github user zwoop closed the pull request at:
https://github.com/apache/trafficserver/pull/745
---
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 feature
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/747
Any thoughts on this ?
---
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
Github user zwoop closed the pull request at:
https://github.com/apache/trafficserver/pull/744
---
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 feature
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/749
Yeah, I need to work on this some more, not sure exactly yet what path to
take yet. We have some options :). I'd like to keep this open, so we retain the
feedback.
---
If your project i
Github user zwoop commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/748#discussion_r69008502
--- Diff: plugins/healthchecks/healthchecks.c ---
@@ -301,8 +301,22 @@ parse_configs(const char *fname)
char buf[2 * 1024];
HCFileInfo
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/762
clang-format
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/765
Bike shedding and side-tracking a bit, but we should encourage to not use
the pattern
if (foo)
delete foo;
it should be just
delete foo
Github user zwoop closed the pull request at:
https://github.com/apache/trafficserver/pull/739
---
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 feature
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/739
@SolidWallOfCode decided that we should keep this, so closing.
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/766
[approve ci]
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/767
:+1:
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/749
Another option would be to just not remove the Last-Modified here, but
unclear what exactly implications that has. I guess I really ought to look
through all possible status codes, and see if
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/749
So, do we want to change it such that this isn't used for 304's? Or do we
want to make it conditional in here for 304's? Maybe build_error_response()
should avoid setting or mo
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/761
Question: Where does this fit in with the new order of the LRU for
keep-alive connections? It feels that if we're about to refuse connections (or
close it immediately after accept), we s
Github user zwoop commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/761#discussion_r68876660
--- Diff: doc/admin-guide/files/records.config.en.rst ---
@@ -643,6 +644,8 @@ tr-out
tr-pass
Transparent pass through. This option is useful
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/761
clang-format
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/743
Also, I'm not married to the new log tags. It seems Github mangled my
comment badly, which I fixed right now. If someone has better suggestions for
naming the tags, please voice your op
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/663
clang-format
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/757
Hmmm, wasn't the idea that we'd use :program:`traffic_ctl`? :)
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If yo
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/751
This seems good, +1.
One comment, as discussed on the IRC:
It seems unfortunate that we have two state variables (closed=true, and the
_state= HTTP2_STREAM_STATE_CLOSED for H2
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/624
Yeah, examples are great. So please create a Jira, if you are going to work
on it. Any changes to APIs would need to go through API review though, so
depending on what you are doing, you might
Github user zwoop closed the pull request at:
https://github.com/apache/trafficserver/pull/755
---
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 feature
GitHub user zwoop opened a pull request:
https://github.com/apache/trafficserver/pull/755
TS-4518 Eliminate memory leaks in UUID regressions
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/zwoop/trafficserver TS-4518
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/749
@masaori335 HttpTransact::build_response() seems correct, doesn't it? But,
seeing that, also makes me think that the place where we modify Cache-Control:
(see your comment above) might act
Github user zwoop commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/749#discussion_r68776903
--- Diff: proxy/http/HttpTransact.cc ---
@@ -8199,13 +8199,15 @@ HttpTransact::build_error_response(State *s,
HTTPStatus status_code, const char
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/749
I updated the PR as per William's post on the Jira. I had a bunch of typos
and errors here, I hope this fixes those issues.
---
If your project is set up for it, you can reply to this
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/754
[approve ci].
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/752
[approve ci]
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/753
[approve ci].
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/642
Updating this PR is fine too
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/739
@SolidWallOfCode Can you decide what to do here? I'm going back and forth
on this :-/. You get to put the final word on this, I'm honestly ok keeping the
current state and just
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/750
Also fix the merge conflicts and squash the commits down to one please :)
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/750
Also, if the old default was "1", why did we change to "0" now?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as w
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/750
Minor details, but
1) Can this be squashed into one commit?
2) Please update the Subject / Title of this PR accordingly.
---
If your project is set up for it, you can reply
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/750
Also, the branch has merge conflicts, so please rebase with current master.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/642
Assigned a number of people for review, this is a change with potentially
dangerous implications. Also, @oknet, is it possible to squash this down to
fewer commits?
---
If your project is set
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/642
@oknet There's now a merge conflict, can you rebase please?
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/647
Adding @shinrich for review too please.
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/664
@jpeach you had some concerns on this concept ?
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/673
@oknet This needs a rebase please. @shinrich Can you review please?
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/624
Also, repeating the question above, what is the Jira for this?
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/624
This needs a rebase as well before we can proceed.
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/593
@mingzym Is this going to land? Needs a rebase before you can merge this
too.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/351
@bgaff What do you want to do with this? It's been sitting here for quite a
while.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitH
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/210
We still need an update to this. @oschaaf and @jpeach , assigning to you
for final reviews, but we should try to land this assuming it seems reasonable
(once clang-format etc. has been
Github user zwoop closed the pull request at:
https://github.com/apache/trafficserver/pull/123
---
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 feature
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/123
Going to close this, since there's been no traction on this in a long time.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as wel
Github user zwoop closed the pull request at:
https://github.com/apache/trafficserver/pull/199
---
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 feature
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/199
I'm going to close this one. I have landed the plumbing for UUID in various
components. @acaciocenteno if you want these changes on the HttpSM debug
statements, can you open a Jira p
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/749
@bryancall @masaori335 Please review.
---
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
GitHub user zwoop opened a pull request:
https://github.com/apache/trafficserver/pull/749
TS-4506 Should not remove Expires/Last-Modifed on 304 response
When generating a 304 page, we used to remove Expires and Last-Modified
always. This change makes sure we never touch Expires
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/743
@PSUdaemon Please review? :)
---
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
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/744
@shukitchan Did you file a Jira?
---
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
GitHub user zwoop opened a pull request:
https://github.com/apache/trafficserver/pull/748
TS-4504 Use the default config dir for configs
When a config file is given, and it's not an absolute path, we
should default to the TSConfigDirGet() base dir.
You can merge this
Github user zwoop commented on the issue:
https://github.com/apache/trafficserver/pull/746
Seems benign, +1.
:+1:
---
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
1 - 100 of 578 matches
Mail list logo
