[GitHub] trafficserver issue #1562: dns Negative cache
Github user pubyun commented on the issue: https://github.com/apache/trafficserver/issues/1562 maybe another issue: ATS will try ipv6 dns resolve for non-exist host name every time also i config flowing config: CONFIG proxy.config.hostdb.ip_resolve STRING ipv4;none sudo tcpdump -i em1 -n port 53 <<< tcpdump: verbose output suppressed, use -v or -vv for full protocol decode listening on em1, link-type EN10MB (Ethernet), capture size 65535 bytes 14:53:09.523847 IP 118.184.176.21.26037 > 118.184.176.35.domain: 42650+ A? www.pubyun.com.dns139.com. (43) 14:53:09.524741 IP 118.184.176.35.domain > 118.184.176.21.26037: 42650 NXDomain* 0/1/0 (96) 14:53:09.545515 IP 118.184.176.21.49774 > 118.184.176.38.domain: 27685+ ? www.pubyun.com.dns139.com. (43) 14:53:09.546240 IP 118.184.176.38.domain > 118.184.176.21.49774: 27685 NXDomain* 0/1/0 (96) 14:53:13.125052 IP 118.184.176.21.57917 > 118.184.176.35.domain: 52856+ A? www.pubyun.com.dns139.com. (43) 14:53:13.125884 IP 118.184.176.35.domain > 118.184.176.21.57917: 52856 NXDomain* 0/1/0 (96) 14:53:13.146626 IP 118.184.176.21.51052 > 118.184.176.38.domain: 28095+ ? www.pubyun.com.dns139.com. (43) 14:53:13.147184 IP 118.184.176.38.domain > 118.184.176.21.51052: 28095 NXDomain* 0/1/0 (96) --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1562: dns Negative cache
GitHub user pubyun opened an issue: https://github.com/apache/trafficserver/issues/1562 dns Negative cache tcpdump shows that ATS will query DNS server for a non-exist host name every time. there is no dns Negative cache in ATS. i test ATS 6.2.1. --- 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 feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1399: Paft of tcp_congestion_control code was erased by...
Github user maskit closed the issue at: https://github.com/apache/trafficserver/issues/1399 --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1399: Paft of tcp_congestion_control code was erased by...
Github user maskit commented on the issue: https://github.com/apache/trafficserver/issues/1399 Fixed on 6.2.x, 7.1.x and 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 your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1535: Cleanup new tests/ tree with RAT exclusions / lic...
Github user zwoop commented on the issue: https://github.com/apache/trafficserver/issues/1535 This is done. --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1535: Cleanup new tests/ tree with RAT exclusions / lic...
Github user zwoop closed the issue at: https://github.com/apache/trafficserver/issues/1535 --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1426: issue #1399 add code of tcp_congestion_control er...
Github user zwoop commented on the issue: https://github.com/apache/trafficserver/pull/1426 This is now back ported to 7.1.0. --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1561: A new 7.1 Crash with regex_revalidate config upda...
Github user zwoop commented on the issue: https://github.com/apache/trafficserver/issues/1561 Yeh, I think you have to create the mutex. Before all these changes, the core would create one for you if it needed it, but that had a massive single mutex around that. So instead, we now assert, and the plugins must be updated, because they are using APIs that requires the mutex. It is possible there's a bug somewhere, but odds are that the plugin is using an API that needs a cont mutex. --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1557: brotli support in gzip plugin
Github user myraid commented on the issue: https://github.com/apache/trafficserver/pull/1557 @bryancall Yes the default behavior is br. Are you suggesting that the option should dictate normalizing the AE? What will be the behavior when both gzip and br are selected? --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver pull request #1557: brotli support in gzip plugin
Github user myraid commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/1557#discussion_r105300325
--- Diff: plugins/gzip/gzip.cc ---
@@ -95,18 +104,29 @@ gzip_data_alloc(int compression_type)
}
}
+ if(compression_type == COMPRESSION_TYPE_BROTLI) {
+ debug("gzip-transform: brotli compression. Create Brotli Encoder
Instance.");
--- End diff --
Done.
---
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 is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] trafficserver issue #1557: brotli support in gzip plugin
Github user bryancall commented on the issue: https://github.com/apache/trafficserver/pull/1557 It looks like brotli is on by default? It would be nice to have an option on what types of compression algorithms that you want to enable. --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver pull request #1557: brotli support in gzip plugin
Github user bryancall commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/1557#discussion_r105279083
--- Diff: plugins/gzip/gzip.cc ---
@@ -95,18 +104,29 @@ gzip_data_alloc(int compression_type)
}
}
+ if(compression_type == COMPRESSION_TYPE_BROTLI) {
+ debug("gzip-transform: brotli compression. Create Brotli Encoder
Instance.");
--- End diff --
Please clang-format over the change. gmake 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 if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] trafficserver issue #1516: Implement Cache-Control: immutable handling
Github user shukitchan commented on the issue: https://github.com/apache/trafficserver/pull/1516 feature flag or feature gate should be fine. i think it is a bug, too. but unfortunately i think we still need to support that till the next major release. --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1544: AddressSanitizer failed to deallocate
Github user zwoop commented on the issue: https://github.com/apache/trafficserver/issues/1544 Duplicate of #1498. --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1544: AddressSanitizer failed to deallocate
Github user zwoop closed the issue at: https://github.com/apache/trafficserver/issues/1544 --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1558: redundant url encoding for logging field "cquuc"
Github user zwoop commented on the issue: https://github.com/apache/trafficserver/issues/1558 What version is this? I could have sworn we fixed 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 wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1561: A new 7.1 Crash
GitHub user shinrich opened an issue: https://github.com/apache/trafficserver/issues/1561 A new 7.1 Crash After running in production for 2.5 days without incident (with a couple fixes now merged back to 7.1.x), I got a core with the following stack trace ``` (gdb) bt #0 0x2ba487c1c625 in raise () from /lib64/libc.so.6 #1 0x2ba487c1dd8d in abort () from /lib64/libc.so.6 #2 0x2ba48576b5a9 in ink_abort (message_format=0x2ba48577d75c "%s:%d: failed assertion `%s`") at ink_error.cc:99 #3 0x2ba485768cda in _ink_assert (expression=0x7b3d70 "((INKContInternal *)contp)->mutex", file=0x7b2dad "InkAPI.cc", line=4372) at ink_assert.cc:37 #4 0x0052bab7 in _TSReleaseAssert (text=0x7b3d70 "((INKContInternal *)contp)->mutex", file=0x7b2dad "InkAPI.cc", line=4372) at InkAPI.cc:401 #5 0x00534f77 in TSContSchedule (contp=0x2aad1801bd60, timeout=30, tp=TS_THREAD_POOL_TASK) at InkAPI.cc:4372 #6 0x2aaabbc39132 in config_handler (cont=0x1fe9ca0, event=Unhandled dwarf expression opcode 0xf3 ) at regex_revalidate/regex_revalidate.c:371 #7 0x0052c9fb in INKContInternal::handle_event (this=0x1fe9ca0, event=2, edata=0x2aad10079b80) at InkAPI.cc:1049 #8 0x005184f0 in Continuation::handleEvent (this=0x1fe9ca0, event=2, data=0x2aad10079b80) at /home/shinrich/yats_build/trafficserver/iocore/eventsystem/I_Continuation.h:153 #9 0x007aac3f in EThread::process_event (this=0x109abc90, e=0x2aad10079b80, calling_code=2) at UnixEThread.cc:143 #10 0x007aaf57 in EThread::execute (this=0x109abc90) at UnixEThread.cc:225 #11 0x007aa303 in spawn_thread_internal (a=0x105b0280) at Thread.cc:84 #12 0x2ba487f86aa1 in start_thread () from /lib64/libpthread.so.0 #13 0x2ba487cd293d in clone () from /lib64/libc.so.6 (gdb) ``` The problem is that TSContSchedule calls FORCE_PLUGIN_SCOPED_MUTEX which asserts that the continuation has a mutex. In this case, the continuation does not have a mutex. This check was added last May to solve TS-4387. The offending plugin is regex_revalidate. If the configuration file is updated, it creates a continuation and spins it off on a task thread. But it is calling TSContCreate with a second argument of NULL. Indeed the timestamp of the config file is a minute before the timestamp on the core file. The straightforward fix would be create a mutex to pass into TSContCreate. Not sure what the downside of that would be. It appears that this is not a frequently run path, so I doubt the creation of an additional object would affect much one way or the other. --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1560: Replace all usage (and implementation) of Vec<> w...
GitHub user zwoop opened an issue: https://github.com/apache/trafficserver/issues/1560 Replace all usage (and implementation) of Vec<> with std::vector @SolidWallOfCode says that it's better, and I like consistency, so we should unify the code around using one vector implementation. --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1557: brotli support in gzip plugin
Github user myraid commented on the issue: https://github.com/apache/trafficserver/pull/1557 @bryancall @zwoop I am working on a enhancement to force brotli compression even if content-encoding header is present. Do you think it will be a good idea to have a separate plugin for unzip and combine it with this plugin if there is a need for force encoding brotli? --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1426: issue #1399 add code of tcp_congestion_control er...
Github user maskit commented on the issue: https://github.com/apache/trafficserver/pull/1426 @zwoop Yes, we should. I wrote the reason on #1399. --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1335: Deadlock in HostDB
Github user zwoop commented on the issue: https://github.com/apache/trafficserver/issues/1335 @jacksontj @vmamidi is working on this, and has a fix (hopefully). --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1412: Assertion in HttpSM::tunnel_handler_post
Github user zwoop commented on the issue: https://github.com/apache/trafficserver/issues/1412 @oknet Did any of your commits address this issue? --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1527: More and slower active connections in 7.1.x
Github user zwoop commented on the issue: https://github.com/apache/trafficserver/issues/1527 On our boxes, it was very noticeable. I can try again, but the net result was roughly 2-3x the number of active connections, and each transaction was 1/2 - 1/3 the speed (throughput in Mbps). This was measured for large downloads, greater than 1MB in object size. --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1446: Need dedicated TS_SSL_SERVERNAME_HOOK
Github user zwoop commented on the issue: https://github.com/apache/trafficserver/pull/1446 @shinrich @oknet Is this still a 7.1.0 candidate? Or do we move out to 7.2.0? --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1426: issue #1399 add code of tcp_congestion_control er...
Github user zwoop commented on the issue: https://github.com/apache/trafficserver/pull/1426 @maskit should we backport this to 7.1.0 ? --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1526: When SSL connect fails, we return 502 success
Github user zwoop commented on the issue: https://github.com/apache/trafficserver/pull/1526 I'd like to put this in for 7.1.0 I think. Any objections? --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1531: Assertion in state_read_server_response_header (v...
Github user zwoop closed the issue at: https://github.com/apache/trafficserver/issues/1531 --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver pull request #1559: Do not callback one SM EVENT_ERROR twice.
Github user zwoop closed the pull request at: https://github.com/apache/trafficserver/pull/1559 --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1516: Implement Cache-Control: immutable handling
Github user danobi commented on the issue: https://github.com/apache/trafficserver/pull/1516 Ok. Does a feature gate sound good? We can flip the flag on the next release. --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1481: Log Field % always outputs "-"
Github user zwoop commented on the issue: https://github.com/apache/trafficserver/issues/1481 Yeh, I suck. Patch coming. --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1175: socks.config - dest_ip rule issue
Github user 06chaynes commented on the issue: https://github.com/apache/trafficserver/issues/1175 I apologize for the lack of updates on this, i Just now got a chance to pull this back out and start testing. I verified that the setup works properly in Debian. I also agree that my issue on the FreeBSD box is related to dependency issues. I will be working those issues out soon and will close the ticket once I have verified the issue. Thanks again for all your help. --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver pull request #1559: Do not callback one SM EVENT_ERROR twice.
Github user zwoop commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/1559#discussion_r105187896
--- Diff: iocore/net/UnixNetVConnection.cc ---
@@ -271,9 +271,22 @@ read_from_net(NetHandler *nh, UnixNetVConnection *vc,
EThread *thread)
// if it is a non-temporary error, we should die appropriately
if (err && err != EAGAIN && err != EINTR) {
- read_signal_error(nh, vc, err);
+ Continuation *reader_cont = vc->read.vio._cont;
+
+ if (read_signal_error(nh, vc, err) == EVENT_DONE) {
+return;
+ }
+ // If vc is closed or shutdown(WRITE) in last read_signal_error
callback,
+ // or reader_cont is same as write.vio._cont.
--- End diff --
Great comment, this really helps explaining the issue / solution.
---
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 is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] trafficserver issue #1559: Do not callback one SM EVENT_ERROR twice.
Github user zwoop commented on the issue: https://github.com/apache/trafficserver/pull/1559 Nice! Testing this PR on docs.trafficserver now. --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1559: Do not callback one SM EVENT_ERROR twice.
Github user zwoop commented on the issue: https://github.com/apache/trafficserver/pull/1559 @oknet Yes, ignore AU tests for now. --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1531: Assertion in state_read_server_response_header (v...
Github user oknet commented on the issue: https://github.com/apache/trafficserver/issues/1531 I created a PR #1559 to resolve this issue. --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
[GitHub] trafficserver issue #1559: Do not callback one SM EVENT_ERROR twice.
Github user scw00 commented on the issue: https://github.com/apache/trafficserver/pull/1559 @oknet I it work well on my test env, and jtest can not re-product any more ! --- 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 is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---
