RE: Memory leak in Tomcat
--- Ingrid Morterud <[EMAIL PROTECTED]> wrote: > > > Thanks for the quick reply. > > You might be right in us not understanding the java > heap. Still - then > we are even more at a loss on how to fix the problem > than if we really > had understood how it works. > > We are running on a test server with 64 mb total > memory. I know I can > increase that, still increasing it will not solve > the original problem. > To be quite honest I am not quite sure what I would > be expecting to be > using, but I would think that the application up and > running would use > less than what it is using at the moment. > > We are using the following code to determine the > memory used: > > > <%=java.lang.Runtime.getRuntime().freeMemory()/1024%> > KB > > <%=java.lang.Runtime.getRuntime().maxMemory()/1024%> > KB > > <%=java.lang.Runtime.getRuntime().totalMemory()/1024%> > KB > > > The application uses approx 40 - 45 Mb when it is > running. During > runtime (when users access and use the application) > they use from zero > to 25 Mb of memory. I still cannot find any pattern > as to when it uses > the memory. The amount of memory used changes not > accordingly to the > user input, that means that when a user does the > same thing twice, that > does not mean that the same amount of memory is > used. > > We haven't used a memory profiler as of yet, but we > are going to try > that out now. > > If you have any more hints and tips, it would be > highly appreciated. > > > Ingrid and Tommy > Ingrid, I included this on the tomcat users list. Yes, any time you reply to a mail where you asked the question on the list then please include the entire list. It will help everyone help you out as they will get the information you give me, and if it is something they could better help you with then the right person got the info, and you can get helped faster. Yeah, 64mb of memory could be enough depending on what you are doing, but you are using struts and I don't know what other libraries. The jvm itself will use a number of megs of memory simply by loading classes and static information into what is know as Persistent memory. A good link would be: http://java.sun.com/docs/performance/ where you will find a lot of information about memory and performance. Also see: http://java.sun.com/docs/hotspot/gc1.4.2/ Also understand that the info you will see with the commands you are using in your source code are not going to show you the memory being used by the persistent section of the JVM process nor are they going to show you the OS reserved memory for the process or "virtual memory". So, you might have issues trying to use Tomcat on a 64MB machine depending on the number of libraries used to the number of classes loaded to the number of static variables and things of that nature. The OS will use a number of memory along with what ever other applications you are using. After that memory is used you start paging to disk a lot and performance will stink at best. You can also search the list for JProfiler. There have been other mailings about memory and leaks on the list and a lot of information for a starting point has already been provided. I can simply start tomcat with only the admin and manager application running and be using 22mb of memory. Are you memory usage reports after your web application has loaded. Then after it has loaded you are using 40+mb? You can find jstat and install it into your 1.4.2 jvm. If you have 1.5 it will already be available. Then with tomcat running do a jps to locate your PID and then jstat -class PID to explain the classes loaded before you hit your first URL to your web app vs after. Might tell you something as well. Using: http://java.sun.com/docs/hotspot/gc1.4.2/ and JAVA_OPTS you should be able to adjust the memory usage of tomcat, and if not you might have to dig into catalina.bat or if on windows use the configure tomcat GUI for the service. But, that amount of memory is so tiny I don't think you'll have much luck if your web application expects much usage. It will all depend on the number of classes being loaded and used and the number of objects being instantiated. You can limit your entire heap with the -mx option of the JVM. This will not however limit your persistent memory usage. You'll have to use -XX:MaxPermSize to limit that. Limiting your heap and your permsize however will mean you know for a fact or good close estimate that you should be loading x number of classes and using x number of perm memory and limiting your heap means you have calculated your application and tomcats expected memory usage and number of supported users for your needs. Wade - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: Memory leak in Tomcat
Hi, On my test environment I am just on 64 Mb of memory. I know I can increase that - but that still will not fix my initial problem. My application is using 40 - 45 Mb - and that is more than I thought it should use. At the moment I have no JAVA_OPTS. Thanks for trying to help. :-) Ingrid and Tommy -Original Message- From: Mark [mailto:[EMAIL PROTECTED] Sent: 12. september 2005 22:36 To: Tomcat Users List; [EMAIL PROTECTED] Subject: RE: Memory leak in Tomcat Hi, Can you share how much memory do you have and how much used by tomcat and what JAVA_OPTs do you have. Thanks a lot, Mark. --- Michael Oliver <[EMAIL PROTECTED]> wrote: > Ingrid, > > I am not on the tomcat developer committer list so my reply is just an > FYI from my own experience. > > I saw unstable performance myself in a very similar deployment of > Struts applications similar to yours. I too thought there was a > memory leak and > there may be, but I don't think it is in the applications > themselves. The > behavior I saw, led me to think it was related to socket allocation > as after > a period of time my system began to complain and slow down and > other socket > related programs began to complain about timeouts, etc. > > I found that my tomcat needed to use virtual memory to avoid out of > memory exceptions. I added physical memory and the problems all but > went away, > however it still occurs just less frequently. > > I am using > > j2sdk1.4.2_09 > Tomcat-5.0.28 > > On Windows XP Pro sp1 > > > > Michael Oliver > CTO > Alarius Systems LLC > 6800 E. Lake Mead Blvd, #1096 > Las Vegas, NV 89156 > Phone:(702)643-7425 > Fax:(702)974-0341 > *Note new email changed from [EMAIL PROTECTED] > > -Original Message- > From: Ingrid Morterud Rosvall [mailto:[EMAIL PROTECTED] > Sent: Monday, September 12, 2005 1:00 PM > To: tomcat-user@jakarta.apache.org > Subject: Memory leak in Tomcat > > Hello. > > We are running an application on Tomcat 4.1.30, and java 1.4.2. > > Our application is using the struts framework with jsp's, and cocoon > to render the xml's. > > There seems to be a major memory leak at startup - the application > seems to constantly be using between 40 - 45 mb of the memory. We also > have some > memory leak during runtime, when users log on and starts using the > application. > > So far we have not been able to find anything in our code review that > will explain these memory leaks, and when we monitor the memory used, > there is no > obvious reason, nor is there any connection with how the users use > our > application and the amount of memory being used. > > We would highly appreciate any help on this topic, and any tips and > hints you can provide us with. > > Ingrid and Tommy > > > -- > I am using the free version of SPAMfighter for private users. It has > removed 4102 spam emails to date. Paying users do not have this > message in their emails. Try www.SPAMfighter.com for free now! > > > > - > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > > > > - > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > > __ Yahoo! Mail - PC Magazine Editors' Choice 2005 http://mail.yahoo.com - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] -- I am using the free version of SPAMfighter for private users. It has removed 4102 spam emails to date. Paying users do not have this message in their emails. Try www.SPAMfighter.com for free now! - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: Memory leak in Tomcat
Hi, Can you share how much memory do you have and how much used by tomcat and what JAVA_OPTs do you have. Thanks a lot, Mark. --- Michael Oliver <[EMAIL PROTECTED]> wrote: > Ingrid, > > I am not on the tomcat developer committer list so my reply is just > an FYI > from my own experience. > > I saw unstable performance myself in a very similar deployment of > Struts > applications similar to yours. I too thought there was a memory > leak and > there may be, but I don't think it is in the applications > themselves. The > behavior I saw, led me to think it was related to socket allocation > as after > a period of time my system began to complain and slow down and > other socket > related programs began to complain about timeouts, etc. > > I found that my tomcat needed to use virtual memory to avoid out of > memory > exceptions. I added physical memory and the problems all but went > away, > however it still occurs just less frequently. > > I am using > > j2sdk1.4.2_09 > Tomcat-5.0.28 > > On Windows XP Pro sp1 > > > > Michael Oliver > CTO > Alarius Systems LLC > 6800 E. Lake Mead Blvd, #1096 > Las Vegas, NV 89156 > Phone:(702)643-7425 > Fax:(702)974-0341 > *Note new email changed from [EMAIL PROTECTED] > > -Original Message- > From: Ingrid Morterud Rosvall [mailto:[EMAIL PROTECTED] > Sent: Monday, September 12, 2005 1:00 PM > To: tomcat-user@jakarta.apache.org > Subject: Memory leak in Tomcat > > Hello. > > We are running an application on Tomcat 4.1.30, and java 1.4.2. > > Our application is using the struts framework with jsp's, and > cocoon to > render the xml's. > > There seems to be a major memory leak at startup - the application > seems to > constantly be using between 40 - 45 mb of the memory. We also have > some > memory leak during runtime, when users log on and starts using the > application. > > So far we have not been able to find anything in our code review > that will > explain these memory leaks, and when we monitor the memory used, > there is no > obvious reason, nor is there any connection with how the users use > our > application and the amount of memory being used. > > We would highly appreciate any help on this topic, and any tips and > hints > you can provide us with. > > Ingrid and Tommy > > > -- > I am using the free version of SPAMfighter for private users. > It has removed 4102 spam emails to date. > Paying users do not have this message in their emails. > Try www.SPAMfighter.com for free now! > > > > - > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: > [EMAIL PROTECTED] > > > > - > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: > [EMAIL PROTECTED] > > __ Yahoo! Mail - PC Magazine Editors' Choice 2005 http://mail.yahoo.com - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: Memory leak in Tomcat
Ingrid, I am not on the tomcat developer committer list so my reply is just an FYI from my own experience. I saw unstable performance myself in a very similar deployment of Struts applications similar to yours. I too thought there was a memory leak and there may be, but I don't think it is in the applications themselves. The behavior I saw, led me to think it was related to socket allocation as after a period of time my system began to complain and slow down and other socket related programs began to complain about timeouts, etc. I found that my tomcat needed to use virtual memory to avoid out of memory exceptions. I added physical memory and the problems all but went away, however it still occurs just less frequently. I am using j2sdk1.4.2_09 Tomcat-5.0.28 On Windows XP Pro sp1 Michael Oliver CTO Alarius Systems LLC 6800 E. Lake Mead Blvd, #1096 Las Vegas, NV 89156 Phone:(702)643-7425 Fax:(702)974-0341 *Note new email changed from [EMAIL PROTECTED] -Original Message- From: Ingrid Morterud Rosvall [mailto:[EMAIL PROTECTED] Sent: Monday, September 12, 2005 1:00 PM To: tomcat-user@jakarta.apache.org Subject: Memory leak in Tomcat Hello. We are running an application on Tomcat 4.1.30, and java 1.4.2. Our application is using the struts framework with jsp's, and cocoon to render the xml's. There seems to be a major memory leak at startup - the application seems to constantly be using between 40 - 45 mb of the memory. We also have some memory leak during runtime, when users log on and starts using the application. So far we have not been able to find anything in our code review that will explain these memory leaks, and when we monitor the memory used, there is no obvious reason, nor is there any connection with how the users use our application and the amount of memory being used. We would highly appreciate any help on this topic, and any tips and hints you can provide us with. Ingrid and Tommy -- I am using the free version of SPAMfighter for private users. It has removed 4102 spam emails to date. Paying users do not have this message in their emails. Try www.SPAMfighter.com for free now! - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Memory leak in Tomcat
--- Ingrid Morterud Rosvall <[EMAIL PROTECTED]> wrote: > Hello. > > We are running an application on Tomcat 4.1.30, and > java 1.4.2. > > Our application is using the struts framework with > jsp's, and cocoon to > render the xml's. > > There seems to be a major memory leak at startup - > the application seems > to constantly be using between 40 - 45 mb of the > memory. We also have > some memory leak during runtime, when users log on > and starts using the > application. > > So far we have not been able to find anything in our > code review that > will explain these memory leaks, and when we monitor > the memory used, > there is no obvious reason, nor is there any > connection with how the > users use our application and the amount of memory > being used. > > We would highly appreciate any help on this topic, > and any tips and > hints you can provide us with. > > Ingrid and Tommy > One, I think you might be having issues by not understanding the java heap...just an observation by the way you phrased the question. Two, do you have any more information about the memory being used? How much were you expecting to be used? Are you seeing the virtual memory usage or the real memory usage? How did you determine the amount of memory used? Do you have any numbers? Have you tried to use a memory profiler? Search the list for memory profiler. Wade - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Memory leak in Tomcat
Hello. We are running an application on Tomcat 4.1.30, and java 1.4.2. Our application is using the struts framework with jsp's, and cocoon to render the xml's. There seems to be a major memory leak at startup - the application seems to constantly be using between 40 - 45 mb of the memory. We also have some memory leak during runtime, when users log on and starts using the application. So far we have not been able to find anything in our code review that will explain these memory leaks, and when we monitor the memory used, there is no obvious reason, nor is there any connection with how the users use our application and the amount of memory being used. We would highly appreciate any help on this topic, and any tips and hints you can provide us with. Ingrid and Tommy -- I am using the free version of SPAMfighter for private users. It has removed 4102 spam emails to date. Paying users do not have this message in their emails. Try www.SPAMfighter.com for free now! - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Suspected memory leak in Tomcat or JVM?
There are some memory leaks in the AJP/1.3 Connector (e.g. http://issues.apache.org/bugzilla/show_bug.cgi?id=32141), but the CVS logs say that these were introduced after 4.1.27. "Jochen Wiedmann" <[EMAIL PROTECTED]> wrote in message news:[EMAIL PROTECTED] Hi, we have an elder application running on Tomcat 4.1.27 with Java 1.4.2_08 on Sparc Solaris 8. Recently we moved the application to a new machine running on Sparc Solaris 9. Since then we have a serious memory problem and need to restart the same application twice a day. One minor change: We are now using mod_jk 1.2.14 and no longer mod_proxy. As I am unaware of any serious heap profiler that is fast enough for production use, I decided to try the very simple heap profiler from http://www.virtualmachine.de/. I made a dump after starting the application and before shutting down. The results were (to me) quite surprising: The main difference was in the following numbers: At startBefore shutdown Objects Size Objects Size [byte5638 6M160796569M [char 3006402 195M 3211457418M ByteChunk62 2K 345450 14M CharChunk 46 2K 324080 13M (With ByteChunk and CharChunk being from the org.apache.tomcat.util.buf package.) To me, this numbers seem to be related. Any ideas? Regards, Jochen -- Having experienced 7 years of labour/green government, I now know the reason, why a conservative government is good for the economy: The economy's unable to imagine anything else ... - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Suspected memory leak in Tomcat or JVM?
Hi, we have an elder application running on Tomcat 4.1.27 with Java 1.4.2_08 on Sparc Solaris 8. Recently we moved the application to a new machine running on Sparc Solaris 9. Since then we have a serious memory problem and need to restart the same application twice a day. One minor change: We are now using mod_jk 1.2.14 and no longer mod_proxy. As I am unaware of any serious heap profiler that is fast enough for production use, I decided to try the very simple heap profiler from http://www.virtualmachine.de/. I made a dump after starting the application and before shutting down. The results were (to me) quite surprising: The main difference was in the following numbers: At startBefore shutdown Objects Size Objects Size [byte5638 6M160796569M [char 3006402 195M 3211457418M ByteChunk62 2K 345450 14M CharChunk 46 2K 324080 13M (With ByteChunk and CharChunk being from the org.apache.tomcat.util.buf package.) To me, this numbers seem to be related. Any ideas? Regards, Jochen -- Having experienced 7 years of labour/green government, I now know the reason, why a conservative government is good for the economy: The economy's unable to imagine anything else ... - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Memory leak in tomcat 5.0.28
Marx, Mitchell E (Mitch), ALABS wrote: I see the bugzilla ID: http://issues.apache.org/bugzilla/show_bug.cgi?id=33368 Anyone know if this is present in Tomcat 4.1.30? This is now fixed in CVS for TC4. Mark - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Memory leak in tomcat 5.0.28
Marx, Mitchell E (Mitch), ALABS wrote: I see the bugzilla ID: http://issues.apache.org/bugzilla/show_bug.cgi?id=33368 Anyone know if this is present in Tomcat 4.1.30? Yes. http://issues.apache.org/bugzilla/show_bug.cgi?id=20758 is also present but is fixed in 4.1.31 Mark - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Memory leak in tomcat 5.0.28
Thank you Robert!! Just wanna say thanks alot for sharing all your findings with the rest of us. I start my tomcat 5.0.28 server with -ms252m -mx512m and it was running for about 3-4 days before i got the OutOfMemoryError. Since i removed the swallowOutput from my context, my server has'nt been over 200m of used memory for 5 days!! Once again thank you for finding this leak. Regards Trond - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Memory leak in tomcat 5.0.28
I see the bugzilla ID: http://issues.apache.org/bugzilla/show_bug.cgi?id=33368 Anyone know if this is present in Tomcat 4.1.30? - Original Message - From: "Robert Wille" <[EMAIL PROTECTED]> To: Sent: Wednesday, February 02, 2005 1:25 PM Subject: RE: Memory leak in tomcat 5.0.28 I've figured out my problem. I'm posting what I've discovered for the benefit of others. The SystemLogHandler uses a map called logs where the key is a ThreadWithAttributes and the value is a stack of CaptureLogs. The problem is that when a thread dies, the ThreadWithAttributes object lives forever because the map is never cleaned out. Threads come and go in the thread pool, so stuff keeps accumulating there forever. You can prevent the problem by turning off swallowOutput. logs should be a ThreadLocal, not a map. That way the ThreadWithAttributes objects can be collected (as well as the stack of CaptureLogs). - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Memory leak in tomcat 5.0.28
feel free to open a bug report, so that this issue can be tracked. - Original Message - From: "Robert Wille" <[EMAIL PROTECTED]> To: Sent: Wednesday, February 02, 2005 1:25 PM Subject: RE: Memory leak in tomcat 5.0.28 I've figured out my problem. I'm posting what I've discovered for the benefit of others. The SystemLogHandler uses a map called logs where the key is a ThreadWithAttributes and the value is a stack of CaptureLogs. The problem is that when a thread dies, the ThreadWithAttributes object lives forever because the map is never cleaned out. Threads come and go in the thread pool, so stuff keeps accumulating there forever. You can prevent the problem by turning off swallowOutput. logs should be a ThreadLocal, not a map. That way the ThreadWithAttributes objects can be collected (as well as the stack of CaptureLogs). >From: "Robert Wille" <[EMAIL PROTECTED]> >Reply-To: "Tomcat Users List" >To: tomcat-user@jakarta.apache.org >Subject: Memory leak in tomcat 5.0.28 >Date: Wed, 02 Feb 2005 09:16:49 -0700 > >I'm running tomcat 5.0.28 on Linux with JRE 1.4.2_04 and I seem to have a >memory leak. I am not using Apache, but am using the Coyote connector. The >server has been running under heavy load, being accessed by about 150 >computers running automated tests. I took heap snapshots about 8 and 20 >hours into the test using YourKit Java Profiler. When taking the snapshots, >I first paused the system for several minutes, attempted to allocate more >memory than was available to cause all collectable objects to be collected, >and then took the snapshot. Therefore, the snapshots should contain very >few collectable objects, and there should be very few open http >connections. The following seems very suspicious: > >The last snapshot shows 419 Http11Processor objects referencing 41M of >memory. That is an increase of 232 Http11Processor objects. > >It also shows 81,829 objects in the org.apache.tomcat.util.buf package, >which reference 37M of memory. This is an increase of 44,874 objects. > >The buffers and Http11Processor objects appear to be referenced by >org.apache.tomcat.util.threads.ThreadWithAttributes. I show 425 instances, >which is an increase of 225. > >The first snapshot was 8 hours into the test, and in reality, I think the >system should have reached steady state just a few minutes into the test. >But I am obviously accumulating a lot of stuff. > >Can somebody help? > >Robert Wille > >_ >Don’t just search. Find. Check out the new MSN Search! >http://search.msn.click-url.com/go/onm00200636ave/direct/01/ > > >- >To unsubscribe, e-mail: [EMAIL PROTECTED] >For additional commands, e-mail: [EMAIL PROTECTED] > _ Don’t just search. Find. Check out the new MSN Search! http://search.msn.click-url.com/go/onm00200636ave/direct/01/ - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: Memory leak in tomcat 5.0.28
I've figured out my problem. I'm posting what I've discovered for the benefit of others. The SystemLogHandler uses a map called logs where the key is a ThreadWithAttributes and the value is a stack of CaptureLogs. The problem is that when a thread dies, the ThreadWithAttributes object lives forever because the map is never cleaned out. Threads come and go in the thread pool, so stuff keeps accumulating there forever. You can prevent the problem by turning off swallowOutput. logs should be a ThreadLocal, not a map. That way the ThreadWithAttributes objects can be collected (as well as the stack of CaptureLogs). From: "Robert Wille" <[EMAIL PROTECTED]> Reply-To: "Tomcat Users List" To: tomcat-user@jakarta.apache.org Subject: Memory leak in tomcat 5.0.28 Date: Wed, 02 Feb 2005 09:16:49 -0700 I'm running tomcat 5.0.28 on Linux with JRE 1.4.2_04 and I seem to have a memory leak. I am not using Apache, but am using the Coyote connector. The server has been running under heavy load, being accessed by about 150 computers running automated tests. I took heap snapshots about 8 and 20 hours into the test using YourKit Java Profiler. When taking the snapshots, I first paused the system for several minutes, attempted to allocate more memory than was available to cause all collectable objects to be collected, and then took the snapshot. Therefore, the snapshots should contain very few collectable objects, and there should be very few open http connections. The following seems very suspicious: The last snapshot shows 419 Http11Processor objects referencing 41M of memory. That is an increase of 232 Http11Processor objects. It also shows 81,829 objects in the org.apache.tomcat.util.buf package, which reference 37M of memory. This is an increase of 44,874 objects. The buffers and Http11Processor objects appear to be referenced by org.apache.tomcat.util.threads.ThreadWithAttributes. I show 425 instances, which is an increase of 225. The first snapshot was 8 hours into the test, and in reality, I think the system should have reached steady state just a few minutes into the test. But I am obviously accumulating a lot of stuff. Can somebody help? Robert Wille _ Dont just search. Find. Check out the new MSN Search! http://search.msn.click-url.com/go/onm00200636ave/direct/01/ - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] _ Dont just search. Find. Check out the new MSN Search! http://search.msn.click-url.com/go/onm00200636ave/direct/01/ - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Memory leak in tomcat 5.0.28
I'm running tomcat 5.0.28 on Linux with JRE 1.4.2_04 and I seem to have a memory leak. I am not using Apache, but am using the Coyote connector. The server has been running under heavy load, being accessed by about 150 computers running automated tests. I took heap snapshots about 8 and 20 hours into the test using YourKit Java Profiler. When taking the snapshots, I first paused the system for several minutes, attempted to allocate more memory than was available to cause all collectable objects to be collected, and then took the snapshot. Therefore, the snapshots should contain very few collectable objects, and there should be very few open http connections. The following seems very suspicious: The last snapshot shows 419 Http11Processor objects referencing 41M of memory. That is an increase of 232 Http11Processor objects. It also shows 81,829 objects in the org.apache.tomcat.util.buf package, which reference 37M of memory. This is an increase of 44,874 objects. The buffers and Http11Processor objects appear to be referenced by org.apache.tomcat.util.threads.ThreadWithAttributes. I show 425 instances, which is an increase of 225. The first snapshot was 8 hours into the test, and in reality, I think the system should have reached steady state just a few minutes into the test. But I am obviously accumulating a lot of stuff. Can somebody help? Robert Wille _ Dont just search. Find. Check out the new MSN Search! http://search.msn.click-url.com/go/onm00200636ave/direct/01/ - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Horrible memory leak in tomcat 5.0.19
I'm running out of memory with the msg: ... WARNING: Error registering request May 25, 2004 5:36:44 PM org.apache.tomcat.util.threads.ThreadPool$ControlRunnable run SEVERE: Caught exception (java.lang.OutOfMemoryError) executing [EMAIL PROTECTED], terminating thread May 25, 2004 5:37:05 PM org.apache.jk.common.HandlerRequest invoke INFO: Unknown message 0 java.lang.OutOfMemoryError java.lang.OutOfMemoryError java.lang.OutOfMemoryError ... I've seen a sugestion of the "request.registerRequests=false" to put in jk2.properties. I'm using mod_jk, but as I see a similar message, I'll try this too. But... where do I put this conf, as mod_jk does not have a jk.properties file. could be it worker.properties? :) thanks in advance Joseph Shraibman wrote: Robert Krüger wrote: Hi, we had the same problem (enormous memory leak which frequently made our production system crash), downgraded to 5.0.18 and everything went back to normal. Just yesterday a colleague of mine came to the conclusion that it is not too unlikely that it is the problem described in the message http://www.mail-archive.com/[EMAIL PROTECTED]/msg53035.html It that is so, you can simply disable the JMX registration of requests to get rid of the problem. We will try that probably later today but of course you can give it a shot yourself. There is not much to lose. I was able to figure out how to reproduce this problem on my test machine (using multiple concurrent requests). I added to my jk2.properties: request.registerRequests=false This got rid of some of the messages in catalina.out but left these: Mar 19, 2004 1:47:51 PM org.apache.jk.common.HandlerRequest invoke INFO: Unknown message 0 Mar 19, 2004 1:47:52 PM org.apache.jk.common.HandlerRequest invoke INFO: Unknown message 0 ... and the memory leak did not go away. The message that I'm not getting anymore are: Mar 19, 2004 1:25:34 PM org.apache.jk.common.HandlerRequest decodeRequest WARNING: Error registering request Mar 19, 2004 1:25:34 PM org.apache.jk.common.HandlerRequest decodeRequest WARNING: Error registering request - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] -- Emerson Cargnin Analista de Sistemas Setor de Desenvolvimento de Sistemas - TRE-SC tel : (048) - 251-3700 - Ramal 3181 - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Horrible memory leak in tomcat 5.0.19
Hello Robert, We are having the same problem with mod_jk 1.2.5 and Tomcat 5.0.19. However, we cannot downgrade to Tomcat 5.0.18 because of a bug fix regarding session states. I was wondering, have you guys tried disableing the JMX registration? Was that indeed the problem? Thank you very much. Asim On Mar 18, 2004, at 4:08 AM, Robert Krüger wrote: Hi, we had the same problem (enormous memory leak which frequently made our production system crash), downgraded to 5.0.18 and everything went back to normal. Just yesterday a colleague of mine came to the conclusion that it is not too unlikely that it is the problem described in the message http://www.mail-archive.com/[EMAIL PROTECTED]/msg53035.html It that is so, you can simply disable the JMX registration of requests to get rid of the problem. We will try that probably later today but of course you can give it a shot yourself. There is not much to lose. Regards, Robert Joseph Shraibman wrote: Shapira, Yoav wrote: Hi, Does the memory leak occur when you run tomcat standalone? I can't run the production server in standalone, and I can't seem to hit my test server hard enough to reproduce the problem clearly. I'll keep trying. Can you point out its source e.g. by running with a profiler? I don't have a profiler :( Was the load during these 50 minutes exceptionally heavy? On the heavy side, but not really stressing the server except for this memroy leak. Tomcat 4.1.18 handled loads like this all the time. BTW How can I view the jk2 status? I tried both of these: In httpd.conf: JkUriSet worker status in workers2.properties: [uri:/status/*] worker=status:status In my error_log I see: [Wed Mar 17 14:26:42 2004] [notice] Apache/2.0.48 (Unix) mod_ssl/2.0.48 OpenSSL/0.9.7a mod_jk2/2.0.2 configured -- resuming normal operations [Wed Mar 17 14:26:42 2004] [error] uriEnv.init() map to invalid worker /jkstatus/* status [Wed Mar 17 14:26:42 2004] [error] uriEnv.init() map to invalid worker /status/* status:status [Wed Mar 17 14:26:42 2004] [error] mod_jk child init 1 -2 [Wed Mar 17 14:26:43 2004] [error] jk2_init() Can't find child 2506 in scoreboard ... and going to the url just produces a 500 error. Also The faq at http://jakarta.apache.org/tomcat/faq/connectors.html says that mod_jk is the preferred connector and that jk2 is not production quality. My understanding was that jk was in maintence mode and jk2 is considered the stable connector. Is the faq right or is it just out of date? - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] -- Robert Krüger Signal7 GmbH Brüder Knauss Str. 79 64285 Darmstadt Germany - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Horrible memory leak in tomcat 5.0.19 (JMX bug?)
I know, I ran a few hundred lynx's at a time. Ralph Einfeldt wrote: To your question: I don't know. Having a delay in the jsp (or servlet) isn't enough to enshure concurrent requests. You have to do something on the calling site to enforce this. If you just call lynx as in your example, lynx will wait until the response is there. So if you delay the jsp, you just delay the return of lynx. So you have either to run more than one instace of your script at the same time or to extend the scrip in a way that it runs several lynx instances at the same time. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: Horrible memory leak in tomcat 5.0.19 (JMX bug?)
To your question: I don't know. Having a delay in the jsp (or servlet) isn't enough to enshure concurrent requests. You have to do something on the calling site to enforce this. If you just call lynx as in your example, lynx will wait until the response is there. So if you delay the jsp, you just delay the return of lynx. So you have either to run more than one instace of your script at the same time or to extend the scrip in a way that it runs several lynx instances at the same time. > -Original Message- > From: Joseph Shraibman [mailto:[EMAIL PROTECTED] > Sent: Friday, March 19, 2004 7:21 PM > To: Tomcat Users List > Subject: Re: Horrible memory leak in tomcat 5.0.19 (JMX bug?) > > Why are those message generated by a thread starting up or something? > If so I already have a jsp with a built in delay I can use to > make sure there are a bunch of requests at the same time > - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Horrible memory leak in tomcat 5.0.19
Robert Krüger wrote: Hi, we had the same problem (enormous memory leak which frequently made our production system crash), downgraded to 5.0.18 and everything went back to normal. Just yesterday a colleague of mine came to the conclusion that it is not too unlikely that it is the problem described in the message http://www.mail-archive.com/[EMAIL PROTECTED]/msg53035.html It that is so, you can simply disable the JMX registration of requests to get rid of the problem. We will try that probably later today but of course you can give it a shot yourself. There is not much to lose. I was able to figure out how to reproduce this problem on my test machine (using multiple concurrent requests). I added to my jk2.properties: request.registerRequests=false This got rid of some of the messages in catalina.out but left these: Mar 19, 2004 1:47:51 PM org.apache.jk.common.HandlerRequest invoke INFO: Unknown message 0 Mar 19, 2004 1:47:52 PM org.apache.jk.common.HandlerRequest invoke INFO: Unknown message 0 ... and the memory leak did not go away. The message that I'm not getting anymore are: Mar 19, 2004 1:25:34 PM org.apache.jk.common.HandlerRequest decodeRequest WARNING: Error registering request Mar 19, 2004 1:25:34 PM org.apache.jk.common.HandlerRequest decodeRequest WARNING: Error registering request - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Horrible memory leak in tomcat 5.0.19 (JMX bug?)
Ralph Einfeldt wrote: That may be a bit to simple, as this is just performing one request after the other. You may have to force concurrent request: Why are those message generated by a thread starting up or something? If so I already have a jsp with a built in delay I can use to make sure there are a bunch of requests at the same time - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: Horrible memory leak in tomcat 5.0.19 (JMX bug?)
That may be a bit to simple, as this is just performing one request after the other. You may have to force concurrent request: i=1 while [ $i -lt 2 ] ; do j=1 while [ $j -lt 10 ] ; do # start lynx in background lynx -source $URL > /dev/null & j=`expr $j + 1` done wait # until all background tasks returned i=`expr $i + 1` done You may also have a look at ab or jMeter. They allow for a better test than home grown test like yours. (My enhancement might not help much, as the result of the script depends to a degree on the time that lynx needs to start.) > -Original Message- > From: Joseph Shraibman [mailto:[EMAIL PROTECTED] > Sent: Thursday, March 18, 2004 8:32 PM > To: Tomcat Users List > Subject: Re: Horrible memory leak in tomcat 5.0.19 (JMX bug?) > > > My test to try to reproduce this bug was: > i=1 > while [ $i -lt 2 ] ; do > lynx -source $URL > /dev/null > i=`expr $i + 1` > done > > ... where I changed $URL to point to different servlets or jsps. > - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Horrible memory leak in tomcat 5.0.19 (JMX bug?)
So how do I reproduce these messages: Mar 18, 2004 2:33:31 PM org.apache.jk.common.HandlerRequest invoke INFO: Unknown message 0 Mar 18, 2004 2:46:46 PM org.apache.jk.common.HandlerRequest decodeRequest WARNING: Error registering request ... which presumably are related to the memory leak? Shapira, Yoav wrote: Hi, What is this JMX thing anyway? Is it supposed to produce a report for the jk status page (that I could never get to work anyway?) It's a lot more than that: many tomcat components are now manageable via JMX (any JMX console will do). There's also a nice report in the Manager webapp (http://localhost:8080/manager/status is the default URL), not related to JK. Yoav Shapira This e-mail, including any attachments, is a confidential business communication, and may contain information that is confidential, proprietary and/or privileged. This e-mail is intended only for the individual(s) to whom it is addressed, and may not be saved, copied, printed, disclosed or used by anyone else. If you are not the(an) intended recipient, please immediately delete this e-mail from your computer system and notify the sender. Thank you. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: Horrible memory leak in tomcat 5.0.19 (JMX bug?)
Hi, >What is this JMX thing anyway? Is it supposed to produce a report for >the jk status page (that I could never get to work anyway?) It's a lot more than that: many tomcat components are now manageable via JMX (any JMX console will do). There's also a nice report in the Manager webapp (http://localhost:8080/manager/status is the default URL), not related to JK. Yoav Shapira This e-mail, including any attachments, is a confidential business communication, and may contain information that is confidential, proprietary and/or privileged. This e-mail is intended only for the individual(s) to whom it is addressed, and may not be saved, copied, printed, disclosed or used by anyone else. If you are not the(an) intended recipient, please immediately delete this e-mail from your computer system and notify the sender. Thank you. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Horrible memory leak in tomcat 5.0.19 (JMX bug?)
Great, thanks for the info! But I already downgraded my production machine to 4.1.18 to avoid this problem. I wish I knew how to reproduce this on my development box so I can know for sure if this solves the problem. Otherwise I can't justify trying this in productions. My test to try to reproduce this bug was: i=1 while [ $i -lt 2 ] ; do lynx -source $URL > /dev/null i=`expr $i + 1` done ... where I changed $URL to point to different servlets or jsps. What is this JMX thing anyway? Is it supposed to produce a report for the jk status page (that I could never get to work anyway?) Robert Krüger wrote: Hi, we had the same problem (enormous memory leak which frequently made our production system crash), downgraded to 5.0.18 and everything went back to normal. Just yesterday a colleague of mine came to the conclusion that it is not too unlikely that it is the problem described in the message http://www.mail-archive.com/[EMAIL PROTECTED]/msg53035.html It that is so, you can simply disable the JMX registration of requests to get rid of the problem. We will try that probably later today but of course you can give it a shot yourself. There is not much to lose. Regards, Robert - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Horrible memory leak in tomcat 5.0.19
Hi, we had the same problem (enormous memory leak which frequently made our production system crash), downgraded to 5.0.18 and everything went back to normal. Just yesterday a colleague of mine came to the conclusion that it is not too unlikely that it is the problem described in the message http://www.mail-archive.com/[EMAIL PROTECTED]/msg53035.html It that is so, you can simply disable the JMX registration of requests to get rid of the problem. We will try that probably later today but of course you can give it a shot yourself. There is not much to lose. Regards, Robert Joseph Shraibman wrote: Shapira, Yoav wrote: Hi, Does the memory leak occur when you run tomcat standalone? I can't run the production server in standalone, and I can't seem to hit my test server hard enough to reproduce the problem clearly. I'll keep trying. Can you point out its source e.g. by running with a profiler? I don't have a profiler :( Was the load during these 50 minutes exceptionally heavy? On the heavy side, but not really stressing the server except for this memroy leak. Tomcat 4.1.18 handled loads like this all the time. BTW How can I view the jk2 status? I tried both of these: In httpd.conf: JkUriSet worker status in workers2.properties: [uri:/status/*] worker=status:status In my error_log I see: [Wed Mar 17 14:26:42 2004] [notice] Apache/2.0.48 (Unix) mod_ssl/2.0.48 OpenSSL/0.9.7a mod_jk2/2.0.2 configured -- resuming normal operations [Wed Mar 17 14:26:42 2004] [error] uriEnv.init() map to invalid worker /jkstatus/* status [Wed Mar 17 14:26:42 2004] [error] uriEnv.init() map to invalid worker /status/* status:status [Wed Mar 17 14:26:42 2004] [error] mod_jk child init 1 -2 [Wed Mar 17 14:26:43 2004] [error] jk2_init() Can't find child 2506 in scoreboard ... and going to the url just produces a 500 error. Also The faq at http://jakarta.apache.org/tomcat/faq/connectors.html says that mod_jk is the preferred connector and that jk2 is not production quality. My understanding was that jk was in maintence mode and jk2 is considered the stable connector. Is the faq right or is it just out of date? - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] -- Robert Krüger Signal7 GmbH Brüder Knauss Str. 79 64285 Darmstadt Germany - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Horrible memory leak in tomcat 5.0.19
Shapira, Yoav wrote: Hi, Does the memory leak occur when you run tomcat standalone? I can't run the production server in standalone, and I can't seem to hit my test server hard enough to reproduce the problem clearly. I'll keep trying. Can you point out its source e.g. by running with a profiler? I don't have a profiler :( Was the load during these 50 minutes exceptionally heavy? On the heavy side, but not really stressing the server except for this memroy leak. Tomcat 4.1.18 handled loads like this all the time. BTW How can I view the jk2 status? I tried both of these: In httpd.conf: JkUriSet worker status in workers2.properties: [uri:/status/*] worker=status:status In my error_log I see: [Wed Mar 17 14:26:42 2004] [notice] Apache/2.0.48 (Unix) mod_ssl/2.0.48 OpenSSL/0.9.7a mod_jk2/2.0.2 configured -- resuming normal operations [Wed Mar 17 14:26:42 2004] [error] uriEnv.init() map to invalid worker /jkstatus/* status [Wed Mar 17 14:26:42 2004] [error] uriEnv.init() map to invalid worker /status/* status:status [Wed Mar 17 14:26:42 2004] [error] mod_jk child init 1 -2 [Wed Mar 17 14:26:43 2004] [error] jk2_init() Can't find child 2506 in scoreboard ... and going to the url just produces a 500 error. Also The faq at http://jakarta.apache.org/tomcat/faq/connectors.html says that mod_jk is the preferred connector and that jk2 is not production quality. My understanding was that jk was in maintence mode and jk2 is considered the stable connector. Is the faq right or is it just out of date? - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: Horrible memory leak in tomcat 5.0.19
Hi, Does the memory leak occur when you run tomcat standalone? Can you point out its source e.g. by running with a profiler? Was the load during these 50 minutes exceptionally heavy? Yoav Shapira Millennium Research Informatics >-Original Message- >From: Joseph Shraibman [mailto:[EMAIL PROTECTED] >Sent: Wednesday, March 17, 2004 1:55 PM >To: Tomcat Users List >Subject: Horrible memory leak in tomcat 5.0.19 > >I recently upgraded my production server from jk 1.2.5/tomcat 4.1.18 to >jk 2.0.2/tomcat 5.0.19. I'm having a horrible memory leak now and need >to restart tomcat frequently when it runs out of memory. I always had >set the max memory to 512m, now after running around 50 minutes memory >usage is already 178,143,080 bytes. Does anyone have any idea what >could be the problem? The only messages in catalina.out are: > >Mar 17, 2004 1:52:51 PM org.apache.jk.common.HandlerRequest invoke >INFO: Unknown message 0 >Mar 17, 2004 1:52:52 PM org.apache.jk.common.HandlerRequest invoke >INFO: Unknown message 0 >Mar 17, 2004 1:52:53 PM org.apache.jk.common.HandlerRequest invoke >INFO: Unknown message 0 >Mar 17, 2004 1:52:57 PM org.apache.jk.common.HandlerRequest invoke >INFO: Unknown message 0 >Mar 17, 2004 1:52:59 PM org.apache.jk.common.HandlerRequest invoke >INFO: Unknown message 0 >Mar 17, 2004 1:53:02 PM org.apache.jk.common.HandlerRequest invoke >INFO: Unknown message 0 >Mar 17, 2004 1:53:03 PM org.apache.jk.common.HandlerRequest invoke >INFO: Unknown message 0 >Mar 17, 2004 1:53:06 PM org.apache.jk.common.HandlerRequest invoke >INFO: Unknown message 0 >Mar 17, 2004 1:53:11 PM org.apache.jk.common.HandlerRequest decodeRequest >WARNING: Error registering request >Mar 17, 2004 1:53:35 PM org.apache.jk.common.HandlerRequest decodeRequest >WARNING: Error registering request >Mar 17, 2004 1:53:37 PM org.apache.jk.common.HandlerRequest decodeRequest >WARNING: Error registering request >Mar 17, 2004 1:53:42 PM org.apache.jk.common.HandlerRequest decodeRequest >WARNING: Error registering request > > >PS The faq at http://jakarta.apache.org/tomcat/faq/connectors.html says >that mod_jk is the preferred connector and that jk2 is not production >quality. My understanding was that jk was in maintence mode and jk2 is >considered the stable connector. Is the faq right or is it just out of >date? > >- >To unsubscribe, e-mail: [EMAIL PROTECTED] >For additional commands, e-mail: [EMAIL PROTECTED] This e-mail, including any attachments, is a confidential business communication, and may contain information that is confidential, proprietary and/or privileged. This e-mail is intended only for the individual(s) to whom it is addressed, and may not be saved, copied, printed, disclosed or used by anyone else. If you are not the(an) intended recipient, please immediately delete this e-mail from your computer system and notify the sender. Thank you. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Horrible memory leak in tomcat 5.0.19
I recently upgraded my production server from jk 1.2.5/tomcat 4.1.18 to jk 2.0.2/tomcat 5.0.19. I'm having a horrible memory leak now and need to restart tomcat frequently when it runs out of memory. I always had set the max memory to 512m, now after running around 50 minutes memory usage is already 178,143,080 bytes. Does anyone have any idea what could be the problem? The only messages in catalina.out are: Mar 17, 2004 1:52:51 PM org.apache.jk.common.HandlerRequest invoke INFO: Unknown message 0 Mar 17, 2004 1:52:52 PM org.apache.jk.common.HandlerRequest invoke INFO: Unknown message 0 Mar 17, 2004 1:52:53 PM org.apache.jk.common.HandlerRequest invoke INFO: Unknown message 0 Mar 17, 2004 1:52:57 PM org.apache.jk.common.HandlerRequest invoke INFO: Unknown message 0 Mar 17, 2004 1:52:59 PM org.apache.jk.common.HandlerRequest invoke INFO: Unknown message 0 Mar 17, 2004 1:53:02 PM org.apache.jk.common.HandlerRequest invoke INFO: Unknown message 0 Mar 17, 2004 1:53:03 PM org.apache.jk.common.HandlerRequest invoke INFO: Unknown message 0 Mar 17, 2004 1:53:06 PM org.apache.jk.common.HandlerRequest invoke INFO: Unknown message 0 Mar 17, 2004 1:53:11 PM org.apache.jk.common.HandlerRequest decodeRequest WARNING: Error registering request Mar 17, 2004 1:53:35 PM org.apache.jk.common.HandlerRequest decodeRequest WARNING: Error registering request Mar 17, 2004 1:53:37 PM org.apache.jk.common.HandlerRequest decodeRequest WARNING: Error registering request Mar 17, 2004 1:53:42 PM org.apache.jk.common.HandlerRequest decodeRequest WARNING: Error registering request PS The faq at http://jakarta.apache.org/tomcat/faq/connectors.html says that mod_jk is the preferred connector and that jk2 is not production quality. My understanding was that jk was in maintence mode and jk2 is considered the stable connector. Is the faq right or is it just out of date? - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: memory leak in tomcat 5.0.16 ?
Thanks, I will upgrade to 5.0.18 and see if it works ! /Torstein -Original Message- From: Allistair Crossley [mailto:[EMAIL PROTECTED] Sent: 22. januar 2004 11:44 To: Tomcat Users List Subject: RE: memory leak in tomcat 5.0.16 ? We had exactly the same problem. 2 users on 5.0.16 after 20 minutes the RAM consumed was 158MB and then it crashed. Upgraded to 5.0.18 yesterday and RAM is a steady 30MB. I dont care what anyone says, 5.0.16 had a problem! -Original Message- From: Francois JEANMOUGIN [mailto:[EMAIL PROTECTED] Sent: 22 January 2004 10:19 To: Tomcat Users List Subject: RE: memory leak in tomcat 5.0.16 ? > -Original Message- > From: Dale, Matt [mailto:[EMAIL PROTECTED] > Sent: Thursday, January 22, 2004 11:05 AM > To: Tomcat Users List > Subject: RE: memory leak in tomcat 5.0.16 ? > > > There is a known memory leak in 5.0.16, I'd upgrade to 5.0.18 and see if > this fixes your problem. Note that the download page on Jakarta.apache.org is not updated with this new release. You need to figure the good URL by yourself (not so hard). François. -- --- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] -- - QAS Ltd. Developers of QuickAddress Software http://www.qas.com";>www.qas.com Registered in England: No 2582055 Registered in Australia: No 082 851 474 -- - -- --- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: memory leak in tomcat 5.0.16 ?
On Thu, January 22, 2004 1at 0:25 am, Torstein Nilsen wrote: > I have now upgraded to the latest tomcat release 5.0.18 but I'm > afraid this didn't solve the problem - the tomcat-process is still > growing. > > I have monitored the ressources used very closely with 5.0.18 and it > shows a slow grow in mem-usage a couple of hours and then suddenly in > a matter of ca. 10 secs. it goes from 15% to 40% of total memory and > stays there. Tomcat access log-files doesn't show extrordinary > activity during the "bloat". Sounds like you're going to need to get a profiler to figure out where the memory is being used. -Dave - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: memory leak in tomcat 5.0.16 ?
I have now upgraded to the latest tomcat release 5.0.18 but I'm afraid this didn't solve the problem - the tomcat-process is still growing. I have monitored the ressources used very closely with 5.0.18 and it shows a slow grow in mem-usage a couple of hours and then suddenly in a matter of ca. 10 secs. it goes from 15% to 40% of total memory and stays there. Tomcat access log-files doesn't show extrordinary activity during the "bloat". Thanx anyway. I will try the max / min threads and see if this helps. Torstein -Original Message- From: Filip Hanik [mailto:[EMAIL PROTECTED] Sent: 22. januar 2004 18:58 To: Tomcat Users List Subject: Re: memory leak in tomcat 5.0.16 ? try setting maxSpareThreads==minSpareThreads==maxThreads in your connector, Filip - Original Message - From: "Ralph Einfeldt" <[EMAIL PROTECTED]> To: "Tomcat Users List" <[EMAIL PROTECTED]> Sent: Thursday, January 22, 2004 7:18 AM Subject: RE: memory leak in tomcat 5.0.16 ? > The heap size has nothing to do with the memory size that > is seen by the system. > > You have to look at least at the total memory. > (That is used + free memory) > > To that you have to add > - thread stacks (At least some vm's don't allocate them on the heap) > - static memory (Like the jvm itself, static strings, classes, jars, ...) > - some os memory that is used by the vm to manage it self > - ... > > I wouldn't expect that the diff between total memory and system memory > is more than a few megs. (Far less than 30MB) > > > -Original Message- > > From: Allistair Crossley [mailto:[EMAIL PROTECTED] > > Sent: Thursday, January 22, 2004 4:06 PM > > To: Tomcat Users List > > Subject: RE: memory leak in tomcat 5.0.16 ? > > > > What puzzles me is the Windows task manager process memory as > > this never ever matches anywhere near the JProfiler reported > > memory. I know there may be some system overheads but the > > 30MB heap that JProfiler reveals is actually 90MB in Windows > > task manager. > > > > -- --- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] -- --- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: memory leak in tomcat 5.0.16 ?
try setting maxSpareThreads==minSpareThreads==maxThreads in your connector, Filip - Original Message - From: "Ralph Einfeldt" <[EMAIL PROTECTED]> To: "Tomcat Users List" <[EMAIL PROTECTED]> Sent: Thursday, January 22, 2004 7:18 AM Subject: RE: memory leak in tomcat 5.0.16 ? > The heap size has nothing to do with the memory size that > is seen by the system. > > You have to look at least at the total memory. > (That is used + free memory) > > To that you have to add > - thread stacks (At least some vm's don't allocate them on the heap) > - static memory (Like the jvm itself, static strings, classes, jars, ...) > - some os memory that is used by the vm to manage it self > - ... > > I wouldn't expect that the diff between total memory and system memory > is more than a few megs. (Far less than 30MB) > > > -Original Message- > > From: Allistair Crossley [mailto:[EMAIL PROTECTED] > > Sent: Thursday, January 22, 2004 4:06 PM > > To: Tomcat Users List > > Subject: RE: memory leak in tomcat 5.0.16 ? > > > > What puzzles me is the Windows task manager process memory as > > this never ever matches anywhere near the JProfiler reported > > memory. I know there may be some system overheads but the > > 30MB heap that JProfiler reveals is actually 90MB in Windows > > task manager. > > > > - > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: memory leak in tomcat 5.0.16 ?
The heap size has nothing to do with the memory size that is seen by the system. You have to look at least at the total memory. (That is used + free memory) To that you have to add - thread stacks (At least some vm's don't allocate them on the heap) - static memory (Like the jvm itself, static strings, classes, jars, ...) - some os memory that is used by the vm to manage it self - ... I wouldn't expect that the diff between total memory and system memory is more than a few megs. (Far less than 30MB) > -Original Message- > From: Allistair Crossley [mailto:[EMAIL PROTECTED] > Sent: Thursday, January 22, 2004 4:06 PM > To: Tomcat Users List > Subject: RE: memory leak in tomcat 5.0.16 ? > > What puzzles me is the Windows task manager process memory as > this never ever matches anywhere near the JProfiler reported > memory. I know there may be some system overheads but the > 30MB heap that JProfiler reveals is actually 90MB in Windows > task manager. > - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: memory leak in tomcat 5.0.16 ?
1.4.1_03 on W2K. I have no idea what the consumption of memory should be like but previous posts where I mentioned this about of RAM have had replies saying it was a lot. Depends on what you app does I guess. Ours has a real large XML nav tree in app scope that is used a lot on each request, we have a backend CMS to get search results for thousands of docs and all their meta data gets stored in properties files and we also do lots with the SQL server. At the moment all that is down to 30MB as shown in JProfiler's heap used view. What puzzles me is the Windows task manager process memory as this never ever matches anywhere near the JProfiler reported memory. I know there may be some system overheads but the 30MB heap that JProfiler reveals is actually 90MB in Windows task manager. Go figure? -Original Message- From: Derek Mahar [mailto:[EMAIL PROTECTED] Sent: 22 January 2004 14:58 To: Tomcat Users List Cc: [EMAIL PROTECTED] Subject: RE: memory leak in tomcat 5.0.16 ? Out of curiosity, which JVM do you run? I run Tomcat 5.0.18, JVM 1.4.2_03 for Linux on Red Hat, and two instances of JSPWiki serving no more than 200 users. This combination consumes a steady 121MB. Is this normal or excessive? Derek -Original Message- From: Allistair Crossley [mailto:[EMAIL PROTECTED] Sent: January 22, 2004 5:44 AM To: Tomcat Users List Subject: RE: memory leak in tomcat 5.0.16 ? We had exactly the same problem. 2 users on 5.0.16 after 20 minutes the RAM consumed was 158MB and then it crashed. Upgraded to 5.0.18 yesterday and RAM is a steady 30MB. I dont care what anyone says, 5.0.16 had a problem! -Original Message- From: Francois JEANMOUGIN [mailto:[EMAIL PROTECTED] Sent: 22 January 2004 10:19 To: Tomcat Users List Subject: RE: memory leak in tomcat 5.0.16 ? > -Original Message- > From: Dale, Matt [mailto:[EMAIL PROTECTED] > Sent: Thursday, January 22, 2004 11:05 AM > To: Tomcat Users List > Subject: RE: memory leak in tomcat 5.0.16 ? > > > There is a known memory leak in 5.0.16, I'd upgrade to 5.0.18 and see > if this fixes your problem. Note that the download page on Jakarta.apache.org is not updated with this new release. You need to figure the good URL by yourself (not so hard). François. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] --- QAS Ltd. Developers of QuickAddress Software http://www.qas.com";>www.qas.com Registered in England: No 2582055 Registered in Australia: No 082 851 474 --- - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: memory leak in tomcat 5.0.16 ?
Out of curiosity, which JVM do you run? I run Tomcat 5.0.18, JVM 1.4.2_03 for Linux on Red Hat, and two instances of JSPWiki serving no more than 200 users. This combination consumes a steady 121MB. Is this normal or excessive? Derek -Original Message- From: Allistair Crossley [mailto:[EMAIL PROTECTED] Sent: January 22, 2004 5:44 AM To: Tomcat Users List Subject: RE: memory leak in tomcat 5.0.16 ? We had exactly the same problem. 2 users on 5.0.16 after 20 minutes the RAM consumed was 158MB and then it crashed. Upgraded to 5.0.18 yesterday and RAM is a steady 30MB. I dont care what anyone says, 5.0.16 had a problem! -Original Message- From: Francois JEANMOUGIN [mailto:[EMAIL PROTECTED] Sent: 22 January 2004 10:19 To: Tomcat Users List Subject: RE: memory leak in tomcat 5.0.16 ? > -Original Message- > From: Dale, Matt [mailto:[EMAIL PROTECTED] > Sent: Thursday, January 22, 2004 11:05 AM > To: Tomcat Users List > Subject: RE: memory leak in tomcat 5.0.16 ? > > > There is a known memory leak in 5.0.16, I'd upgrade to 5.0.18 and see > if this fixes your problem. Note that the download page on Jakarta.apache.org is not updated with this new release. You need to figure the good URL by yourself (not so hard). François. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] --- QAS Ltd. Developers of QuickAddress Software http://www.qas.com";>www.qas.com Registered in England: No 2582055 Registered in Australia: No 082 851 474 --- - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: memory leak in tomcat 5.0.16 ?
Francois JEANMOUGIN wrote: There is a known memory leak in 5.0.16, I'd upgrade to 5.0.18 and see if this fixes your problem. Note that the download page on Jakarta.apache.org is not updated with this new release. You need to figure the good URL by yourself (not so hard). Before announcing something, one has to wait for: - voting to complete - mirrors to replicate the build -- x Rémy Maucherat Senior Developer & Consultant JBoss Group (Europe) SàRL x - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: memory leak in tomcat 5.0.16 ?
We had exactly the same problem. 2 users on 5.0.16 after 20 minutes the RAM consumed was 158MB and then it crashed. Upgraded to 5.0.18 yesterday and RAM is a steady 30MB. I dont care what anyone says, 5.0.16 had a problem! -Original Message- From: Francois JEANMOUGIN [mailto:[EMAIL PROTECTED] Sent: 22 January 2004 10:19 To: Tomcat Users List Subject: RE: memory leak in tomcat 5.0.16 ? > -Original Message- > From: Dale, Matt [mailto:[EMAIL PROTECTED] > Sent: Thursday, January 22, 2004 11:05 AM > To: Tomcat Users List > Subject: RE: memory leak in tomcat 5.0.16 ? > > > There is a known memory leak in 5.0.16, I'd upgrade to 5.0.18 and see if > this fixes your problem. Note that the download page on Jakarta.apache.org is not updated with this new release. You need to figure the good URL by yourself (not so hard). François. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] --- QAS Ltd. Developers of QuickAddress Software http://www.qas.com";>www.qas.com Registered in England: No 2582055 Registered in Australia: No 082 851 474 --- - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: memory leak in tomcat 5.0.16 ?
> -Original Message- > From: Dale, Matt [mailto:[EMAIL PROTECTED] > Sent: Thursday, January 22, 2004 11:05 AM > To: Tomcat Users List > Subject: RE: memory leak in tomcat 5.0.16 ? > > > There is a known memory leak in 5.0.16, I'd upgrade to 5.0.18 and see if > this fixes your problem. Note that the download page on Jakarta.apache.org is not updated with this new release. You need to figure the good URL by yourself (not so hard). François. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: memory leak in tomcat 5.0.16 ?
Dale, Matt wrote: There is a known memory leak in 5.0.16, I'd upgrade to 5.0.18 and see if this fixes your problem. You should read his report. 1) I don't see how he would be affected, since he seems to have a rather small server; you need large variations in traffic to get the leak (and the workaround is easy enough anyway) 2) He's using AJP, not HTTP ;) I am developing an intranet for a housing community using apache/tomcat and mysql. It was put in production to 200 users a few weeks ago and I noticed that the java-proces (tomcat) was "growing" from 8% memory usage when started to over 50% (seen with top / ps aux). When the mem-usage reach a certain level the java-proces uses most of the CPU- ressources as well - verbose:GC showed that this is the GC trying to keep up. This usually happens in a matter of 2-5 hours depending on the Xmx-settings - the number of active users / load seems to speed the process up (not confirmed). Restarting tomcat solves the problem for a while - but I would prefer a better permament solution. System settings: CPU: Pentium 1600 mhz RAM: 512 MB OS: Linux (Redhat) Java: j2re1.4.2_03 / jikes TOMCAT: 5.0.16 Connector: org.apache.coyote.tomcat5.CoyoteConnector (AJP/1.3) CATALINA_OPTS: -Xmx200m -Xms200m -Dbuild.compiler.emacs=true (have tried 117 different settings) At first I used tomcat 4 and j2sdk1.4.1_03 and the "normal" javac-compiler. I changed to jakarta-tomcat-5.0.16 / j2re1.4.2_03 and jikes - this seems to have made the problem even worse. Before the update tomcat could go for more than 12 hours whitout restarting - now I have to restart every few hours. The application is quite DB-intensive: Every 30 secs. a java-thread queries a mysql-table with updated network-traffic data (used for traffic-shaping). In the same loop I SAX-parse a little XML-string using the JDOM-API (I have read about the StringBuffer-problem but this is not the cause since I'm now using j2re1.4.2_03 - right ?). I'm using mysql-connector-3.0.9 as JDBC-driver and protomatter-1.1.8 to pool DB-connections. In the same loop I connect to a TCP- socket on the local server. I have done some profiling with HPJmeter and the -Xrunhprof argument with different settings. I'm a newbie in profiling but these observations might be useful: -Using HPJmeter' "guess memory leaks" the top-4 suggested candidates are: java.util.vector, org.apache.commons.modeler.Registry, org.apache.jk.core.MsgContext and org.apache.coyote.Request -"Residual objects" shows the following top-4-classes (bytes): int[] (10 MB) char[] (4 MB) java.lang.String (2 MB) byte[] (2 MB) Any help would be much appreciated... Regards Torstein Nilsen -- x Rémy Maucherat Senior Developer & Consultant JBoss Group (Europe) SàRL x - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: memory leak in tomcat 5.0.16 ?
There is a known memory leak in 5.0.16, I'd upgrade to 5.0.18 and see if this fixes your problem. Ta Matt -Original Message- From: Torstein Nilsen [mailto:[EMAIL PROTECTED] Sent: 22 January 2004 10:01 To: [EMAIL PROTECTED] Subject: memory leak in tomcat 5.0.16 ? I am developing an intranet for a housing community using apache/tomcat and mysql. It was put in production to 200 users a few weeks ago and I noticed that the java-proces (tomcat) was "growing" from 8% memory usage when started to over 50% (seen with top / ps aux). When the mem-usage reach a certain level the java-proces uses most of the CPU- ressources as well - verbose:GC showed that this is the GC trying to keep up. This usually happens in a matter of 2-5 hours depending on the Xmx-settings - the number of active users / load seems to speed the process up (not confirmed). Restarting tomcat solves the problem for a while - but I would prefer a better permament solution. System settings: CPU: Pentium 1600 mhz RAM: 512 MB OS: Linux (Redhat) Java: j2re1.4.2_03 / jikes TOMCAT: 5.0.16 Connector: org.apache.coyote.tomcat5.CoyoteConnector (AJP/1.3) CATALINA_OPTS: -Xmx200m -Xms200m -Dbuild.compiler.emacs=true (have tried 117 different settings) At first I used tomcat 4 and j2sdk1.4.1_03 and the "normal" javac-compiler. I changed to jakarta-tomcat-5.0.16 / j2re1.4.2_03 and jikes - this seems to have made the problem even worse. Before the update tomcat could go for more than 12 hours whitout restarting - now I have to restart every few hours. The application is quite DB-intensive: Every 30 secs. a java-thread queries a mysql-table with updated network-traffic data (used for traffic-shaping). In the same loop I SAX-parse a little XML-string using the JDOM-API (I have read about the StringBuffer-problem but this is not the cause since I'm now using j2re1.4.2_03 - right ?). I'm using mysql-connector-3.0.9 as JDBC-driver and protomatter-1.1.8 to pool DB-connections. In the same loop I connect to a TCP- socket on the local server. I have done some profiling with HPJmeter and the -Xrunhprof argument with different settings. I'm a newbie in profiling but these observations might be useful: -Using HPJmeter' "guess memory leaks" the top-4 suggested candidates are: java.util.vector, org.apache.commons.modeler.Registry, org.apache.jk.core.MsgContext and org.apache.coyote.Request -"Residual objects" shows the following top-4-classes (bytes): int[] (10 MB) char[] (4 MB) java.lang.String (2 MB) byte[] (2 MB) Any help would be much appreciated... Regards Torstein Nilsen - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] Any opinions expressed in this E-mail may be those of the individual and not necessarily the company. This E-mail and any files transmitted with it are confidential and solely for the use of the intended recipient. If you are not the intended recipient or the person responsible for delivering to the intended recipient, be advised that you have received this E-mail in error and that any use or copying is strictly prohibited. If you have received this E-mail in error please notify the beCogent postmaster at [EMAIL PROTECTED] Unless expressly stated, opinions in this email are those of the individual sender and not beCogent Ltd. You must take full responsibility for virus checking this email and any attachments. Please note that the content of this email or any of its attachments may contain data that falls within the scope of the Data Protection Acts and that you must ensure that any handling or processing of such data by you is fully compliant with the terms and provisions of the Data Protection Act 1984 and 1998. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
memory leak in tomcat 5.0.16 ?
I am developing an intranet for a housing community using apache/tomcat and mysql. It was put in production to 200 users a few weeks ago and I noticed that the java-proces (tomcat) was "growing" from 8% memory usage when started to over 50% (seen with top / ps aux). When the mem-usage reach a certain level the java-proces uses most of the CPU- ressources as well - verbose:GC showed that this is the GC trying to keep up. This usually happens in a matter of 2-5 hours depending on the Xmx-settings - the number of active users / load seems to speed the process up (not confirmed). Restarting tomcat solves the problem for a while - but I would prefer a better permament solution. System settings: CPU: Pentium 1600 mhz RAM: 512 MB OS: Linux (Redhat) Java: j2re1.4.2_03 / jikes TOMCAT: 5.0.16 Connector: org.apache.coyote.tomcat5.CoyoteConnector (AJP/1.3) CATALINA_OPTS: -Xmx200m -Xms200m -Dbuild.compiler.emacs=true (have tried 117 different settings) At first I used tomcat 4 and j2sdk1.4.1_03 and the "normal" javac-compiler. I changed to jakarta-tomcat-5.0.16 / j2re1.4.2_03 and jikes - this seems to have made the problem even worse. Before the update tomcat could go for more than 12 hours whitout restarting - now I have to restart every few hours. The application is quite DB-intensive: Every 30 secs. a java-thread queries a mysql-table with updated network-traffic data (used for traffic-shaping). In the same loop I SAX-parse a little XML-string using the JDOM-API (I have read about the StringBuffer-problem but this is not the cause since I'm now using j2re1.4.2_03 - right ?). I'm using mysql-connector-3.0.9 as JDBC-driver and protomatter-1.1.8 to pool DB-connections. In the same loop I connect to a TCP- socket on the local server. I have done some profiling with HPJmeter and the -Xrunhprof argument with different settings. I'm a newbie in profiling but these observations might be useful: -Using HPJmeter' "guess memory leaks" the top-4 suggested candidates are: java.util.vector, org.apache.commons.modeler.Registry, org.apache.jk.core.MsgContext and org.apache.coyote.Request -"Residual objects" shows the following top-4-classes (bytes): int[] (10 MB) char[] (4 MB) java.lang.String (2 MB) byte[] (2 MB) Any help would be much appreciated... Regards Torstein Nilsen - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]