It's interesting but which release.revision are you using? Also consider that there are maybe more users connected to the demo than
to your server. For instance we can have
198 instances of "org.apache.tomcat.util.threads.ThreadWithAttributes", loaded by "java.net.URLClassLoader @ 0x7fb4b8f59220" occupy
286 111 256 (45,01%) bytes.
I have tried the link and the help at large locally this afternoon (with last revision but a Postgres DB not cleared for some days)
and it seems the process never end
Same on demo were the trunk process was running near 100% all the afternoon. I
just stopped an restarted it...
This is why we got so much:
*** ASF Nagios ***
Notification Type: PROBLEM
Host: ofbiz-vm.apache.org
Address: 140.211.11.41
Service: HTTP - Ofbiz Demo Trunk
State: CRITICAL
Info: CRITICAL - Socket timeout after 16 seconds
Date/Time: Wed Dec 8 19:48:21 UTC 2010
this afternoon...
*Demo users: please don't use the online help before this problems is solved*
Thanks
Jacques
From: "BJ Freeman" <[email protected]>
it is interesting that I have more instances than on
ofbiz-vm.apache.org
including two trunk instances and can successfully run the full docbook
=========================
BJ Freeman
Strategic Power Office with Supplier Automation
<http://www.businessesnetwork.com/automation/viewforum.php?f=52>
Specialtymarket.com <http://www.specialtymarket.com/>
Systems Integrator-- Glad to Assist
Chat Y! messenger: bjfr33man
BJ Freeman sent the following on 12/8/2010 4:55 AM:
This is the top level ofbiz docbook link.
yes it takes a lot of resources.
it is doing a list of xls's for all two books in the set.
I am focused on going through the docbooks.
one is to split the tech and user docs into different sets.
Jacques Le Roux sent the following on 12/8/2010 1:13 AM:
=========================
BJ Freeman
Strategic Power Office with Supplier Automation
<http://www.businessesnetwork.com/automation/viewforum.php?f=52>
Specialtymarket.com <http://www.specialtymarket.com/>
Systems Integrator-- Glad to Assist
Chat Y! messenger: bjfr33man
OK, When I use this URL locally I get this
2010-12-08 10:02:15,609 (http-0.0.0.0-28080-1)
[ControlEventListener.java:64 :INFO ] Creating session:
65D15F115C41C7719E22FEBE11BFE2F9.jvm1
2010-12-08 10:02:15,609 (http-0.0.0.0-28080-1) [
ControlServlet.java:147:INFO ] [[[APACHE_OFBIZ_HTML] Request Begun,
encoding=[UTF-8]- total:0.0,since last(Begin):0.0]]
2010-12-08 10:02:15,609 (http-0.0.0.0-28080-1) [
VisitHandler.java:232:INFO ] Found visitorId [10000] in cookie
2010-12-08 10:02:15,625 (http-0.0.0.0-28080-1) [
ConfigXMLReader.java:120:INFO ] controller loaded: 0.0s, 1 requests, 0
views in jndi:/0.0.0.0/cmssite/WEB-INF/controller.xml
2010-12-08 10:02:15,625 (http-0.0.0.0-28080-1) [
ConfigXMLReader.java:120:INFO ] controller loaded: 0.0s, 29 requests, 17
views in
file:/D:/workspace/ofbizDev/framework/common/webcommon/WEB-INF/common-cont
roller.xml
2010-12-08 10:02:15,625 (http-0.0.0.0-28080-1) [ CmsEvents.java:145:INFO
] Path INFO for Alias: APACHE_OFBIZ_HTML
And then nothing in log but the Java process eats as much CPU ressources
as it can. Please don't use this on demo!
Jacques
From: "Jacques Le Roux" <[email protected]>
This URL does not seem to be make much sense. How did you come to it?
Jacques
From: "BJ Freeman" <[email protected]>
seems I blew it up again doing:
http://demo-trunk.ofbiz.apache.org/cmssite/cms/APACHE_OFBIZ_HTML
HTTP - Ofbiz Demo Trunk
This service has 1 comment associated with it This service is
flapping between states
CRITICAL 2010-12-08 02:01:58 0d 0h 33m 3s 10/10 CRITICAL - Socket
timeout after 16 seconds
=========================
BJ Freeman
Strategic Power Office with Supplier Automation
<http://www.businessesnetwork.com/automation/viewforum.php?f=52>
Specialtymarket.com <http://www.specialtymarket.com/>
Systems Integrator-- Glad to Assist
Chat Y! messenger: bjfr33man
Jacques Le Roux sent the following on 12/7/2010 1:51 PM:
We have already 2 memory dumps and the suspects so far are
1)
110 542 instances of "org.apache.derby.impl.jdbc.EmbedConnection40",
loaded by "java.net.URLClassLoader @ 0x7f3681c111f0" occupy 415
680 440 (79,89%) bytes. These instances are referenced from one
instance
of "java.util.WeakHashMap$Entry[]", loaded by "<system
class loader>"
More at https://issues.apache.org/jira/browse/OFBIZ-4043
2)
198 instances of
"org.apache.tomcat.util.threads.ThreadWithAttributes",
loaded by "java.net.URLClassLoader @ 0x7fb4b8f59220" occupy
286 111 256 (45,01%) bytes.
Biggest instances:
a.. org.apache.tomcat.util.threads.ThreadWithAttributes @
0x7fb4c47e7818
TP-Processor90 - 8 558 720 (1,35%) bytes.
b.. org.apache.tomcat.util.threads.ThreadWithAttributes @
0x7fb4c35bb570
TP-Processor70 - 8 197 640 (1,29%) bytes.
c.. org.apache.tomcat.util.threads.ThreadWithAttributes @
0x7fb4c9f1b320
TP-Processor122 - 8 195 856 (1,29%) bytes.
d.. org.apache.tomcat.util.threads.ThreadWithAttributes @
0x7fb4c0719340
TP-Processor27 - 7 108 272 (1,12%) bytes.
e.. org.apache.tomcat.util.threads.ThreadWithAttributes @
0x7fb4ca6fd200
TP-Processor129 - 6 912 832 (1,09%) bytes.
f.. org.apache.tomcat.util.threads.ThreadWithAttributes @
0x7fb4cbfe2f60
TP-Processor152 - 6 639 528 (1,04%) bytes.
g.. org.apache.tomcat.util.threads.ThreadWithAttributes @
0x7fb4c80a7b98
TP-Processor114 - 6 494 872 (1,02%) bytes.
h.. org.apache.tomcat.util.threads.ThreadWithAttributes @
0x7fb4ca8bf9a8
TP-Processor133 - 6 478 984 (1,02%) bytes.
For the later it seems we simply miss enough memory to handle the
requests (there are no threads taking a huge part, only 1,35% max)
Most of the time one of the demo instances fails because there is not
enough swap space. But this symptom means something is going wrong
anyway. Because we have 2.5GB of memory and 1.5GB of swap space. The
instances use
branch9: -Xms128M -Xmx512M (default MaxPermSize is 64m for server)
trunk: -Xms128M -Xmx768M -XX:MaxPermSize=256m (I have also tried with
-XX:MaxPermSize=192m but it's the same)
I suspect something from users (or bots) and the infra team
suggested to
use mod_log_forensic. I have not looked at it yet
Jacques
From: "BJ Freeman" <[email protected]>
both my demos are working fine on Centos5
one other thing I noticed is that when the one failed the other
server
was up but there was no fail over to the working server.
Will look at the dumps and see if I can see anything and track on my
demos.
=========================
BJ Freeman
Strategic Power Office with Supplier Automation
<http://www.businessesnetwork.com/automation/viewforum.php?f=52>
Specialtymarket.com <http://www.specialtymarket.com/>
Systems Integrator-- Glad to Assist
Chat Y! messenger: bjfr33man
Jacques Le Roux sent the following on 12/7/2010 1:59 AM:
I'm currently struggling to get both demo working smoothly.
Though we
did not change the running configuration, since the last OS
update we
had a lot of problems.
I have changed the running configuration by minoring the memory
used but
there are still issues. I have already 2 heap dumps which show some
leaks suspects. But I have not enough time to look into details
Jacques
From: "BJ Freeman" <[email protected]>
did a
http://demo-trunk.ofbiz.apache.org/cmssite/cms/APACHE_OFBIZ_HTML
and got
Service Temporarily Unavailable
checked
http://monitoring.apache.org/status/
HTTP - Ofbiz Demo Trunk
WARNING 2010-12-06 21:08:58 0d 0h 6m 3s 6/10 HTTP WARNING: -
HTTP/1.1
200 OK - 13.380 second response time
=========================
BJ Freeman
Strategic Power Office with Supplier Automation
<http://www.businessesnetwork.com/automation/viewforum.php?f=52>
Specialtymarket.com <http://www.specialtymarket.com/>
Systems Integrator-- Glad to Assist
Chat Y! messenger: bjfr33man
