Tomcat hangs with 100% CPU

I am using Tomcat 4.0 on HPUX 64bit, JDK 1.4, JDevloper 9.2. After running for 3-4 days Tomcat hangs with CPU at 100%. I tried to do a kill -s 3 on tomcats JVM but it doesnt do anything. I had put lot of trace just to see which JSP page or function it hangs. I donot get any consistent page or function. This has been happening for about 6 months now. Can somebody point me how to debug. What should i do so that i can get extra trace from tomcat or from Oracles library. What should i do to fix this problem. My only solution now is to kill -9 and restart tomcat.
Thanks in advance
Ravi

Getting a stacktrace from tomcat's JVM is the best way to see who is causing the CPU usage. Check the release notes for the HP JVM and see if they have anything noted on alternative ways to get a stacktrace.
Rob
Team JDev

Similar Messages

Firefox hangs with 100% CPU at (obviously) random times

Several times a day, my Firefox suddenly responds pretty slow and within seconds it no longer responds at all. The CPU usage of firefox.exe goes to 100% and the whole browser seems to be frozen. In some cases, the memory consumption in the TaskManager also increases slowly (about 8KB/s), but not always. However, after 30-60s the problem suddenly vanishes and everything is fine again.
An interesting detail: Firefox uses a lot of memory; I am not sure whether this is normal. Currently, I only have this tab opened, but it still uses 166MB! The peak memory usage even says 351MB, although I did not have more than 8 or 10 tabs open at the same time (no YouTube or similar memory intensive sites).
The problem might be caused by an update to an extension; at least I remember that there was an update shortly before the problem first appeared. However, I do not remember which extension it was. I have disabled almost every extension (only TabMixPlus, Download Statusbar, and TinyMenu are still active), and the problem is still there. It is hard to find the reason because the problem only happens three of four times a day and at random times, so it is hard to reproduce.
Some questions:
1) Anyone has got an idea what might cause these problems?
2) Is there a way to see which extensions have been updated recently?
3) Is there a way to see which extension uses how much of the CPU?
Thanks in advance!
gehho.
== This happened ==
A few times a week
== about two weeks ago (maybe after an add-on update, not sure which)

Just some additional info:
I just had the problem again, and it even occurred when I did not have any tab open! I did not even use Firefox, but just wanted to start to use it, but it did not respond. There was _nothing_ open, and when I checked the TaskManager, the memory consumption was up to 300 MB, and the CPU was again at 100%. After the CPU was down to 0% again, the memory slowly decreased as well. First, to ~260, then to ~200, now to ~170 MB. Maybe it will decrease even further...

RH8 hangs with 100% CPU when closing a topic. Resource Manager to blame?

Please refer to the following thread for background information: http://forums.adobe.com/thread/734972?tstart=0
I'm re-listing this issue, as it has rasied its ugly head again, and I've no resolution this time.
Summary:
The original issue was that RH8 would hang whenever I closed a topic for anywhere between 15-30 seconds before allowing me to interact with the application again. Attempting to interact with the GUI during this freeze would result in the 'Not Responding' message at the upper-left of the application window (suffice it to say I'm on a PC with Windows).
After much investigation it was suggested to me that my profile may be corrupt, and in order to test this, another user should log on and see whether they experienced the same problem. As it is, they did not. So I had my Systems Admin recreate my domain profile, and everything worked just fine.
...until now.
Last week we all upgraded from Windows XP to Windows 7 (32 bit). We also changed our anti-virus software. The awful hanging issue in RH has returned.
We've attempted the solution from last time, by recreating my profile on the domain, to no avail. We have in fact gone to the extent of reformatting the hard drive and reinstalling Windows, the anti-virus software and RH8 (plus all available patches) from scratch. In case anyone is wondering, we've tried running RH without the antivirus installed, and when installed have also tried excluding the RH application folder and all my project folders, just in case it was the anti-virus software at fault. Regardless, the RH problem exists.
I can't understand why I'm now having this problem. Truth be told, I never really understood why recreating my domain profile in the first place resolved the problem last time.
However, I have now discovered something else...
In my RH Resource Manager I have a few thousand images in the Image folder. Not much I can do about this - the applications I write Help for have many screens users can access, and I write the Help for half a dozen applications. When I delete the contents of the images folder within the Resources Manager, RH works perfectly - no hanging whatsoever. As soon as I import the images into this folder again, the hanging is back. I have tested this a few times now, and have rebooted and had another user log on to confirm, and it happens every time. I'm dumbfounded of course, particularly because the same set of images has been sitting in the Resource Manager for months (since my original post about this issue, last year), and RH has been working perfectly.
I've also tried Peter Grainge's suggestion to create an empty BaseCSS_res.dll file (http://www.grainge.org/pages/authoring/rh8/using_rh8.htm), but as I've installed all the RH8 updates, this shouldn't be an issue.
I've also tried deleting the CPD file without success.
Thanks in advance.

It seems this has been resolved!
Many thanks for all the suggestions. Peter ultimately forwarded the solution - something that he'd sourced from a rep at Adobe, as follows:
Every file/folder change in the root folder of RH8 Resource Manager triggers a rescan of all the sub-folder/files. That is why in case of thousand image it will take time. Like xxxx said we may ask user for root folder, it should not be project folder or frequently changing folder.
Probably he is using the default path given in the root folder and RoboHelp changes a temporary css file in that folder very frequently, in fact with every topic close.
That is causing the slowness and it should be easy to reproduce in the default scenario. This is RH8 only issue.
So, to fix my issue:
Open the Resource Manager Pod.
Click the Add/Edit Categories icon. The Add/Edit Categories window appears.
Within this window, change the 'Root Folder' path to something other than the default. In my case, I create a new folder on my root hard drive, outside of the Robohelp folder, just to be sure.
So far this is working perfectly - no 'hanging'. I'll keep an eye on it.
I love the comment "This is RH8 only issue."
Oh?! Is it? Was it fixed in a patch, Adobe? Please don't tell me that it was fixed in RH9, because there's no way known I should have to pay for an upgrade to resolve something that certainly looks like a bug. As happy and grateful I am to have this resolved, it has cost me a week's work.

Audio Bin - Select unused - Delete - Hangs on 100% CPU for long time

Hi
I am running Logic Studio 9.0.2 on OS X 10.5.8 on a Mac Pro. Logic was upgraded from version 7.
In Audio Bin and "Select Unused" and then "Delete", the application might hang on 100% CPU for hours on large projects.
Anyone with an idea of what the problem might be? This was not a problem i Logic Pro 7.
Thanks!
GD

I have this same problem, though I'm on 10.5.7. I was hoping maybe updating the OS would help but apparently not. I also get very slow file-selection windows sometimes, like when I go to bounce a file. Navigating to a specific folder is painfully slow. Would like to hear if any Snow Leopard users are experiencing this.
Bull

Queue consumer stops with 100% cpu usage

I'm trying to use Berkeley DB queue with transactions. When I tested what happens when transactions with DB_APPEND are aborted I found that while it works and DB_CONSUME correctly skips over rolled back records, unfortunately extents that have those records are never deleted, which causes database to always grow. Next I tried DB_CONSUME with database opened using DB_INORDER flag and it seems there's a serious regression in Berkeley DB that causes it to loop indefinitely with 100% cpu usage when it encounters a rolled back record. I tested various versions and found that this bug doesn't happen with 5.1.29, but it is reproducible with 5.2.42, so this regression might have been introduced in 5.2. I have also tested 5.3 and 6.0, and both have this behavior. There may be something wrong with the way queue records are rolled back, one indication of that would be that in 5.1.29 doesn't have neither of the two problems I found with DB_QUEUE: extents are deleted after being consumed, and there are no issues when consuming with DB_INORDER either.
You can find Python code to reproduce this issue here:
https://gist.github.com/snaury/027a3c546f5b0a62a440
Sorry for using Python and not e.g. C++, but it's a lot shorter that way.

We have looked at the issues and they are valid. We will roll the fixes out for this in our next release of BDB. The test case was very useful and really helped to speed the process up. If you have any questions, please contact me directly at [email protected] Thanks again for bringing this to our attention.
thanks
mike

Gdm is running dbus-deamon with 100% CPU load

Hello Guys,
I have installed an Arch with Gnome yesterday (I am comming from Fedora). I am very happy with my installation, nearly everything works fine. Now i noticed, that my CPU load is super high. So i used top to check whats going on...
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
536 gdm 20 0 13108 1380 1136 R 100 0.0 126:25.88 /usr/bin/dbus-daemon --config-file=/etc/at-spi2/accessibility.conf --nofork --print-address 3
So gdm runs the dbus-daemon with 100% load. Whats goin on here? I have checked the wiki/forum and could not find anything.
Thx for help!
Some notes:
DAEMONS=(dbus networkmanager hwclock syslog-ng netfs crond)
gdm is started by systemd.
UPDATE
Hmm, seems to be a bug in dbus... The web is full of reports.
Last edited by ushi (2012-01-27 15:27:56)

From the sound side try enabling and disabling msi for snd-hda-intel module (4 combinations to test). Another bothersome parameter is often snoop. Look for quirks for your hardware in alsa source code. Also you might want to disable your ~/.asoundrc for the time being. If you want to set MID as default card, you can use the recommended modprobe method in the meantime.
I would test another browser first though.

IMS 52 Imap crash/hang with multi CPU?

Hello,
Running on NT with SP6a, noticed that with multiple cpu imap just hangs after a while. The process is still there, and it is not taking up any cpu time. It just won't respond, even telneting to the port. Looks like for every attempted connection there is a close_wait in the netstat. Anyone else see this problem?

Yes, I'm getting this too. Sun engineering has yet to fix it. I have a case open with them.

Modprobe -r wl causes kdeinit4 to lock up with 100% CPU

Have an new issue that I beleive started with kde 4.6.2. Whenever i do a modprobe -r wl kdeinit4 stays at 100% cpu until i force kill it. The reason for the removal of the module is the issue where after resuming from sleep wireless doesnt work untill the module is removed and reloaded. I have reproduced this issue after several reboots. In the mean time I have stopped using the sleep function on my laptop.

Have an new issue that I beleive started with kde 4.6.2. Whenever i do a modprobe -r wl kdeinit4 stays at 100% cpu until i force kill it. The reason for the removal of the module is the issue where after resuming from sleep wireless doesnt work untill the module is removed and reloaded. I have reproduced this issue after several reboots. In the mean time I have stopped using the sleep function on my laptop.

[SOLVED] mandb hangs, taking 100% CPU

I have a VPS running Arch (limited to kernel 3.9.4-1 by the provider). The system was unceremoniously rebooted last night due to a power outage at their facility. When it came back up, the mandb process being run by cron was taking 100% of the CPU and had been running for ~10 hours by the time I noticed it.
I eventually killed the mandb process and re-tried it manually with debugging turned on: `mandb -d`. It appears that processing stops at /usr/share/man/man5/core.5.gz (this is the last output I see, even if I leave it running for an hour or two). Even after the output stops, it is still taking 100% CPU. Anyone have any ideas or other things to try debugging it? I'm all out.
Last edited by aclindsa (2013-11-26 01:02:42)

I re-installed man-pages with `pacman -S man-pages` and then received the following messages:
Purging old database entries in /usr/share/man...
Processing manual pages under /usr/share/man...
Updating index cache for path `/usr/share/man/man2'. Wait...mandb: bad fetch on multi key alarm 2
mandb: index cache /var/cache/man/3354 corrupt
Then, after running
mandb -c
subsequent calls to `mandb` executed successfully. Problem solved. I guess maybe mandb was running when the power outage occurred, corrupting the man index cache... makes me wonder what else got corrupted that I haven't noticed yet.

FCPX runs always with 100% CPU usage

Hi guys,
I have following problem or issue:
When i work with FCPX (10.1.4) i have some extreme performance problems with my machine.
iStats Menus always displays an 100% usage of CPU even if I am not working or doing anything, even my background tasks window does not display anything.
The project I am working on is 1080p 30fps in h.264 codec.
I tried projects in ProRes and i had the same problem.
I closed all windows and programs running: same problem.
Its not just the Mac runs slow its like unusable in some cases, i cant even close FCPX properly.
The system I am running is OSX 10.9.5 on a MacPro 5,1 with:
12 Core at 2,4 Ghz
24Gb 1333 Mhz Ram
500 Gb original Apple SSD
2TB Raid 1 HDD with 200GB free
3TB HDD WD Black 1TB free
GTX680 Mac Edition 2GB VRam
I hope you can tell me how i can fix this problem or minimize it as far as possible.
Cheers!

Then I would simply remove it and reinstall it. If FCPX comes with an uninstaller then use that. If not then see the following:
Uninstalling Software: The Basics
Most OS X applications are completely self-contained "packages" that can be uninstalled by simply dragging the application to the Trash. Applications may create preference files that are stored in the /Home/Library/Preferences/ folder. Although they do nothing once you delete the associated application, they do take up some disk space. If you want you can look for them in the above location and delete them, too.
Some applications may install an uninstaller program that can be used to remove the application. In some cases the uninstaller may be part of the application's installer, and is invoked by clicking on a Customize button that will appear during the install process.
Some applications may install components in the /Home/Library/Applications Support/ folder. You can also check there to see if the application has created a folder. You can also delete the folder that's in the Applications Support folder. Again, they don't do anything but take up disk space once the application is trashed.
Some applications may install a startupitem or a Log In item. Startupitems are usually installed in the /Library/StartupItems/ folder and less often in the /Home/Library/StartupItems/ folder. Log In Items are set in the Accounts preferences. Open System Preferences, click on the Accounts icon, then click on the LogIn Items tab. Locate the item in the list for the application you want to remove and click on the "-" button to delete it from the list.
Some software use startup daemons or agents that are a new feature of the OS. Look for them in /Library/LaunchAgents/ and /Library/LaunchDaemons/ or in /Home/Library/LaunchAgents/.
If an application installs any other files the best way to track them down is to do a Finder search using the application name or the developer name as the search term. Unfortunately Spotlight will not look in certain folders by default. You can modify Spotlight's behavior or use a third-party search utility, EasyFind, instead.
Some applications install a receipt in the /Library/Receipts/ folder. Usually with the same name as the program or the developer. The item generally has a ".pkg" extension. Be sure you also delete this item as some programs use it to determine if it's already installed.
There are many utilities that can uninstall applications. Here is a selection:
    1. AppZapper
    2. AppDelete
    3. Automaton
    4. Hazel
    5. AppCleaner
    6. CleanApp
    7. iTrash
    8. Amnesia
    9. Uninstaller
10. Spring Cleaning
For more information visit The XLab FAQs and read the FAQ on removing software.

Installed Firefox 8.0 and plugin container.exe makes pc hang with 99% cpu usage, how can I fix this?

All extensions and add ons are disables other then flash player. Flash player is needed for the webpage.

Reboot. Uninstall Firefox, then use another browser to download and install a new copy of Firefox from mozilla.org/firefox - see [[Installing Firefox on Windows]] for more help.
See also the post [https://support.mozilla.com/en-US/questions/901329 Firefox 8 Memory Leak]
Please mark this issue as "Solved" if it solved your issue, or post back if it did not.
Edited to add link to similar thread

MobileClient.exe holds 100% CPU usage in Citrix

Hi
We have successfully got Mobile Sales 5.0 SP11 working on Citrix but unfortunately, as soon as we have more than 10 people working on it, the application hangs with 100% CPU usage
This server has 4 CPU's and should not max out with so few users.
The infrastructure is set out with the application and the database sitting on different virtual servers with a gigabit connection between them.
The locking point always seems to be on the application server.
Any ideas?
James

Another question:
How did you get it work?
I want to do this for CRM7.0 and I don't know how...
Thank you in advance
Simone

Some web pages causing 100% CPU use of iexplore.exe since Flash update

I'm having a problem with certain Amazon.com pages hanging my
Internet Explorer 6 using Flash Player 10,0,12,36, causing 100% CPU
use for iexplore.exe and requiring me to End Task to quit. All the
pages effected have an interactive mouseover Size and Color chart
that intially says "Loading" but then says "Select Size and Color"
once loaded. One mouseovers the various color patches and size
options, each time giving item availability in that option. When
the pages hang the IE browser with 100% CPU use that chart never
gets to the "Select Size and Color" point, instead stuck at the
"Loading" status. This happens most times, but not every time.
Sometimes the pages load correctly, after a brief lag as it loads
the chart.
Amazon pages without this chart load fine.
The start of this problem coincides with updating Flash.
I can prevent the problem from happening almost every time if
I can quickly get my cursor into that chart area while the page
begins loading. If I allow the pages to load fully with no
interaction they will hang almost every time.
I have not noticed any problems on those same web pages using
Firefox using Flash Player 9,0,124,0
I also have not noticed any issues with other sites using
Internet Explorer either. (YouTube, QVC, CNN...etc, all OK.)
I had posted for assistance on the Experts Exchange site and
there was some debate on whether the chart was a Java or
Flash-based chart, so initially it was being approached as a
possible Java issue, so I removed all Java versions on the system
and installed various versions, but the same problem persists.
I've downloaded but not run the Flash Uninstaller, as I'm
reluctant to uninstall the Firefox Flash version without knowing
for sure I could install 9,0,124,0 again.
I've downloaded an archived fp9_archive.zip, which seems to
indicate the contained files flashplayer9r124_winax.exe (Internet
Explorer) and flashplayer9r124_win.exe (Netscape) would be versions
9,0,124,0 to install after running the uninstaller.
Can anyone check a few of the links below and see if the
"Select Size and Color" chart is Flash based? If so, would
executing the files listed above install 9,0,124,0 into my
respective browsers again?
http://www.amazon.com/FCS-Regular-7mm-Surf-Leash/dp/B00149XJ68
http://www.amazon.com/Womens-Classic-Short-UGG-Boots/dp/B00064NH8W/ref=pd_ecc_rvi_cart_3
http://www.amazon.com/Columbia-Sportswear-Womens-Benton-Springs/dp/B0006M6IQ2/ref=pd_ecc_r vi_cart_2
http://www.amazon.com/Ladies-Chenille-Zip-Cardigan-Sweater/dp/B001E0YJ0C/ref=pd_ecc_rvi_ca rt_4
XP IE 6 (error module : MSHTML.DLL)
Flash Player version 10,0,12,36
Firefox
Flash Player version 9,0,124,0
Thank you-
Vickie

I'm attempting to troubleshoot an issue with Amazon web pages
since I've updated my Flash player. I would appreciate if someone
could take a look at my initial post and advise if the issue seems
to be Flash related.
Thank you.
I'm having a problem with certain Amazon.com pages hanging my
Internet Explorer 6 using Flash Player 10,0,12,36, causing 100% CPU
use for iexplore.exe and requiring me to End Task to quit. All the
pages effected have an interactive mouseover Size and Color chart
that intially says "Loading" but then says "Select Size and Color"
once loaded. One mouseovers the various color patches and size
options, each time giving item availability in that option. When
the pages hang the IE browser with 100% CPU use that chart never
gets to the "Select Size and Color" point, instead stuck at the
"Loading" status. This happens most times, but not every time.
Sometimes the pages load correctly, after a brief lag as it loads
the chart.
Amazon pages without this chart load fine.
The start of this problem coincides with updating Flash.
I can prevent the problem from happening almost every time if
I can quickly get my cursor into that chart area while the page
begins loading. If I allow the pages to load fully with no
interaction they will hang almost every time.
I have not noticed any problems on those same web pages using
Firefox using Flash Player 9,0,124,0
I also have not noticed any issues with other sites using
Internet Explorer either. (YouTube, QVC, CNN...etc, all OK.)
I had posted for assistance on the Experts Exchange site and
there was some debate on whether the chart was a Java or
Flash-based chart, so initially it was being approached as a
possible Java issue, so I removed all Java versions on the system
and installed various versions, but the same problem persists.
I've downloaded but not run the Flash Uninstaller, as I'm
reluctant to uninstall the Firefox Flash version without knowing
for sure I could install 9,0,124,0 again.
I've downloaded an archived fp9_archive.zip, which seems to
indicate the contained files flashplayer9r124_winax.exe (Internet
Explorer) and flashplayer9r124_win.exe (Netscape) would be versions
9,0,124,0 to install after running the uninstaller.
Can anyone check a few of the links below and see if the
"Select Size and Color" chart is Flash based? If so, would
executing the files listed above install 9,0,124,0 into my
respective browsers again?
http://www.amazon.com/FCS-Regular-7mm-Surf-Leash/dp/B00149XJ68
http://www.amazon.com/Womens-Classic-Short-UGG-Boots/dp/B00064NH8W/ref=pd_ecc_rvi_cart_3
http://www.amazon.com/Columbia-Sportswear-Womens-Benton-Springs/dp/B0006M6IQ2/ref=pd_ecc_r vi_cart_2
http://www.amazon.com/Ladies-Chenille-Zip-Cardigan-Sweater/dp/B001E0YJ0C/ref=pd_ecc_rvi_ca rt_4
XP IE 6 (error module : MSHTML.DLL)
Flash Player version 10,0,12,36
Firefox
Flash Player version 9,0,124,0

NIO Selector.select() enters a 100% CPU infinite loop on interruption

This issue is only reproducible in Windows.
When a thread is calling Selector.select() and the other thread interrupts the selector thread, Selector.select() enters an infinite loop taking up CPU time. All other JDKs I've tried including IBM 1.5 / 1.6, SUN 1.5 / 1.6 simply returned 0 when its interrupted. I think JRockit needs to behave similarly, without 100% CPU consumption. Here's the test code:
import java.nio.channels.Selector;
public class Main {
public static void main(String[] args) throws Exception {
final Thread mainThread = Thread.currentThread();
final Selector s = Selector.open();
new Thread() {
@Override
public void run() {
try {
Thread.sleep(1000);
} catch (InterruptedException e) {
// Ignore.
System.out.println("interrupt()");
mainThread.interrupt();
}.start();
System.out.println("select()");
s.select(5000);
System.out.println("success");
s.close();
Expected result:
select()
interrupt()
success
Actual result:
select()
interrupt()
<stuck with 100% CPU consumption>
Version:
JRMC-3.0.3-1.5.0 - build R27.6.0-50_o-100423-1.5.0_15-20080626-2105-windows-ia32, compiled mode
JRMC-3.0.3-1.6.0 - build R27.6.0-50_o-100423-1.6.0_05-20080626-2105-windows-ia32, compiled mode

Hi,
This looks like a bug. Please open a support ticket and request a patch if you need a fix asap. If you can wait we will fix this in an upcoming release. Thank you for your help

OAM process takes 100% CPU (strange SQL in it)

Hi!
I have noticed on our PROD that on eprocess is taking more then 7 hours with 100% CPU load. Here is brief detail about it:
03:08:12 apps@PROD> @si 547
Current SQL....have result if session is active....
SQL_FULLTEXT
Select count(distinct(R.Request_ID)) From Fnd_Concurrent_Programs_vl CP, Fnd_User U, Fnd_Concurr
ent_Requests R, Fnd_Responsibility_Tl RES, Fnd_Application_vl A
Elapsed: 00:00:00.06
SID_SERIAL STATE      OSUSER     TERMINAL   LOGON           SEC_WAIT LAST_CALL
547,42977 WAITED SHO prod       unknown    12.01 12:50:54     26423     26425
            RT TIME
Elapsed: 00:00:00.01
SID_SERIAL      AUDSID DBPID APPPID USERNAME   MODULE        ACTION          STATUS   STATE
547,42977   11087548 10585 1344   APPS       OAM                           ACTIVE   WAITED SHO
                                                                                       RT TIME
Elapsed: 00:00:00.01
03:15:41 apps@PROD> As you can see, SQL is prety strange and it's plan is several Gigs(!!!) Because module is OAM I have no guts to kill it just like that. Is there any way to see what is this really and could it be killed (Do I have to restart concurrent) or what?
THX in front.
Damir Vadas
Edited by: Funky on Jan 13, 2009 3:15 AM

After some KGB approaches, one user with System Administrator responsibility has admit that he has run by mistake u query!
So be aware whom do you expose System Administrator responsibility!
Case closed!

Tomcat hangs with 100% CPU

Similar Messages

Maybe you are looking for