https://bz.apache.org/bugzilla/show_bug.cgi?id=9
Mark Thomas changed:
What|Removed |Added
Resolution|--- |FIXED
Status|NEW
https://bz.apache.org/bugzilla/show_bug.cgi?id=9
--- Comment #15 from Mark Thomas ---
Note 2.0.x is also affected
--
You are receiving this mail because:
You are the assignee for the bug.
-
To unsubscribe, e-mail: dev-unsub
https://bz.apache.org/bugzilla/show_bug.cgi?id=9
Mark Thomas changed:
What|Removed |Added
Component|Catalina|Library
Target Milestone|-
https://bz.apache.org/bugzilla/show_bug.cgi?id=9
Mark Thomas changed:
What|Removed |Added
Status|NEEDINFO|NEW
--- Comment #13 from Mark Thomas --
https://bz.apache.org/bugzilla/show_bug.cgi?id=9
--- Comment #12 from DigitalCat ---
We are anticipating your reply
--
You are receiving this mail because:
You are the assignee for the bug.
-
To unsubscribe, e-mail: dev-uns
https://bz.apache.org/bugzilla/show_bug.cgi?id=9
--- Comment #11 from DigitalCat ---
Yes branch 1.2.X can match line no.
We use wrk to reproduce this issue.
"./wrk -t 20 -c 1000 -d 5s --latency --timeout 1s https://XX/V3/LoginRoute";
wrk is an opensource project, you can try with this
https://bz.apache.org/bugzilla/show_bug.cgi?id=9
--- Comment #10 from Mark Thomas ---
The line numbers match up for the quoted version (1.2.33) and the relevant code
is unchanged in 1.2.x.
The description of a potential memory leak looks plausible but I am unable to
reproduce it with 9.0.x a
https://bz.apache.org/bugzilla/show_bug.cgi?id=9
--- Comment #9 from Christopher Schultz ---
(In reply to DigitalCat from comment #7)
> When line 476 is commented out in native/src/sslcontext.c, the JVM crash
> does not occur.
> We guess the content in servername is not released in line 192 i
https://bz.apache.org/bugzilla/show_bug.cgi?id=9
--- Comment #8 from DigitalCat ---
log in jvm hs_err_.log
--- T H R E A D ---
Current thread (0x2b7a41865800): JavaThread "catalina-exec-18000-120"
daemon [_thread_in_native, id=16233,
stack(0x2b7b724e7000,0
https://bz.apache.org/bugzilla/show_bug.cgi?id=9
--- Comment #7 from DigitalCat ---
We found the way to reproduce this issue。
Upgrading Tomcat Native Library 1.2.23 to Tomcat Native Library 1.2.33, we set
the tomcat in apr mode, In a high-concurrency scenario, the Tomcat process
crashes.
Whe
https://bz.apache.org/bugzilla/show_bug.cgi?id=9
--- Comment #6 from Christopher Schultz ---
Also, the native backtrace would be helpful (it should be found in the
hs_pid_.txt file generated on crash). If you were able to inspect with gdb,
anything you found in there would be helpful as w
https://bz.apache.org/bugzilla/show_bug.cgi?id=9
Mark Thomas changed:
What|Removed |Added
Severity|critical|normal
--- Comment #5 from Mark Thomas
https://bz.apache.org/bugzilla/show_bug.cgi?id=9
Mark Thomas changed:
What|Removed |Added
Status|NEW |NEEDINFO
--- Comment #4 from Mark Thomas
https://bz.apache.org/bugzilla/show_bug.cgi?id=9
--- Comment #3 from Michael Osipov ---
Can you reliably reproduce the issue? What OS are you using? I haven't seen so
many crashes with APR for the past 10 years in such a short time frame.
Especially #setSocketOptions() is weird.
--
You are
https://bz.apache.org/bugzilla/show_bug.cgi?id=9
--- Comment #2 from Mark Thomas ---
Note:
The APR/Tomcat Native HTTP and AJP connectors are deprecated in Tomcat 9 and
have been removed in Tomcat 10.1.x onwards. You have plenty of time before
9.0.x reaches End-Of-Life but you might want to sw
https://bz.apache.org/bugzilla/show_bug.cgi?id=9
--- Comment #1 from DigitalCat ---
add crash stacktrace
Thread 8117: (state = IN_NATIVE)
- org.apache.tomcat.jni.SSLSocket.handshake(long) @bci=0 (Compiled frame;
information may be imprecise)
-
org.apache.tomcat.util.net.AprEndpoint.setSo
16 matches
Mail list logo
