https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #29 from gregh gregh3...@gmail.com ---
Seems fixed with the AjpAprProcessor patch, been over a week without the fans
starting!!! Thanks @Milo ( Mark).
Greg :-)
--
You are receiving this mail because:
You are the assignee for the
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #27 from gregh gregh3...@gmail.com ---
Excellent, I will apply the change in AjpAprProcessor locally to see if its the
same issue (I do not stop and restart tomcat for it to happen). Thanks!
Greg
--
You are receiving this mail
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #28 from Milo m...@vanderzee.org ---
@Gregh: not restarting Tomcat but restarting Apache triggers the bug in my
setup. Stopping Apache kills one end of the socket and so Tomcat starts the
loop.
--
You are receiving this mail
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #23 from Milo m...@vanderzee.org ---
I found that completely stopping apache triggers the bug instantly. The number
of context switches rises from about 5000 cs/sec to 50 cs/sec.
When I then start apache cs/sec goes down to
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #24 from Mark Thomas ma...@apache.org ---
Thanks for all the pointers. I'm able to reproduce behavior that looks like the
behavior described with high CPU on a single java thread. I can't be sure that
I am seeing the same problem
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
Mark Thomas ma...@apache.org changed:
What|Removed |Added
Status|NEW |RESOLVED
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #26 from Milo m...@vanderzee.org ---
Thankyou very much.
--
You are receiving this mail because:
You are the assignee for the bug.
-
To unsubscribe, e-mail:
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #18 from gregh gregh3...@gmail.com ---
Mark,
I do not think I will be able to debug the run() {..} Trying to check whether
the odd/bad state when I really have no idea what it is supposed to do.
if (((desc[n*2]
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #20 from Mark Thomas ma...@apache.org ---
Don't worry too much about why the loop is happening. Once we know the route
execution is taking we'll probably be able to figure out the why. As a starting
point, would it be possible for
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #19 from gregh gregh3...@gmail.com ---
if you check the private void add{..}
what did the ok do?
boolean ok = false;
synchronized (this) {
// Add socket to the list. Newly added sockets will wait
// at most for pollTime
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #21 from Milo m...@vanderzee.org ---
Mark, is it helpfull if I try to setup desktop sharing while the bug is
happening? Then you could debug whatever you like :)
But I'll also add the stacktraces as soon as I'm able to grap them.
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #22 from Milo m...@vanderzee.org ---
Breakpoint 1:
AprEndpoint$Poller.add(139982580808944, -1, 1) line: 1485
AprEndpoint$Poller.access$500(AprEndpoint$Poller, long, int, int) line: 1251
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #17 from Mark Thomas ma...@apache.org ---
The stack trace only shows where execution is at a single point in time. It
doesn't show how execution reached that point or where execution will go
afterwards.
I suspect that something is
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
gregh gregh3...@gmail.com changed:
What|Removed |Added
Status|NEEDINFO|NEW
--- Comment #14 from
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #15 from gregh gregh3...@gmail.com ---
Created attachment 32951
-- https://bz.apache.org/bugzilla/attachment.cgi?id=32951action=edit
Thread 3187 dump
--
You are receiving this mail because:
You are the assignee for the bug.
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #16 from gregh gregh3...@gmail.com ---
Checking the source from 7.0.61 which works ok with 8.0.24 there are some
differences in the run method, maybe these need checking???
removeFromPoller(info.socket); added
rv =
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #12 from gregh gregh3...@gmail.com ---
I have setup eclispe to debug tomcat running on the server, (not done much
remote debugging, so bear with me) when I attach the debugger the with a
breakpoint it stops as expected, if I then
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #13 from Milo m...@vanderzee.org ---
What do you mean with trace the code path. Isn't that the stacktrace? But
that is already given so what exactly do you mean?
--
You are receiving this mail because:
You are the assignee for the
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #9 from gregh gregh3...@gmail.com ---
It happens randomly on all three of my Centos boxes 5/6/7. Maybe its only
centos/rh? Once it starts you have to stop tomcat and restart it.
from the Dump for 8.0.24
ajp-apr-8009-Poller #26
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #10 from Milo m...@vanderzee.org ---
Hello,
I'm using Gentoo so not only Centos is affected. I can trigger the bug by
restarting apache while there are sessions active.
Load on the server does not seem to influence the occurance
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #11 from Mark Thomas ma...@apache.org ---
A couple of further questions.
What, exactly, do you mean by sessions acive? An actively used AJP connection
to Tomcat?
Also, what do you mean by restart? Stop/start? Graceful restart?
It
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
Mark Thomas ma...@apache.org changed:
What|Removed |Added
Status|NEW |NEEDINFO
--- Comment #8
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #7 from gregh gregh3...@gmail.com ---
Centos 7 running via jsvc
PID USER PR NIVIRTRESSHR S %CPU %MEM TIME+ COMMAND
1459 runner20 0 4588960 578536 8312 S 263.0 15.4 8:00.51 jsvc
:{
--
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #6 from Milo m...@vanderzee.org ---
Works fine in Tomcat 7. In Tomcat 8 I see a massive amount of context switching
and so I assume Tomcat tries to create a connection and fails but tries again
immediately.
--
You are receiving
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #4 from Milo m...@vanderzee.org ---
Hello,
I'm having trouble with the following setup:
- Gentoo / mariaDB
- Tomcat 8.0.23
- Tomcat native 1.1.33
- Apache http 2.4.12 with modproxy_ajp
server.xml:
...
Connector port=8208
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #2 from gregh gregh3...@gmail.com ---
Well, same again
PID USER PR NI VIRT RES SHR S %CPU %MEMTIME+ COMMAND
2707 runner20 0 3728m 599m 7356 S 36.5 15.6 52:28.85 java
--
You are receiving this mail because:
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #3 from gregh gregh3...@gmail.com ---
Created attachment 32936
-- https://bz.apache.org/bugzilla/attachment.cgi?id=32936action=edit
Dump for 8.0.24
--
You are receiving this mail because:
You are the assignee for the bug.
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #5 from gregh gregh3...@gmail.com ---
Does it work ok on tomcat 7?
--
You are receiving this mail because:
You are the assignee for the bug.
-
To unsubscribe,
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
--- Comment #1 from gregh gregh3...@gmail.com ---
Will give version 8.0.24 a test. TBA.
--
You are receiving this mail because:
You are the assignee for the bug.
-
To
https://bz.apache.org/bugzilla/show_bug.cgi?id=58151
Ari Luoma arilu...@gmail.com changed:
What|Removed |Added
CC||arilu...@gmail.com
--
30 matches
Mail list logo
