Re: context.xml created by Ant not discovered by Tomcat
Matt Galvin <[EMAIL PROTECTED]> wrote: Yes, there is, I usually put them in: $CATALINA_HOME/conf/Catalina/localhost/myapp.xml Could you post the contents of your context file? Maybe something needs to be tweak possibly. Here is context.xml... factoryorg.apache.commons.dbcp.BasicDataSourceFactory maxActive100 maxIdle30 maxWait1 driverClassNamecom.mysql.jdbc.Driver urljdbc:mysql://localhost/xxDb usernamexx passwordxx As you can see, I'm using a MySQL database. Also, apparently I DON'T have this problem when using Tomcat 5.5.9 (at first, it appeared that I did have the problem with 5.5.9, but after a restart of Tomcat and a reboot of the machine, it worked as desired, but I don't think I should've needed to reboot) - and I notice that the default Balancer app seems to use the same thing. So perhaps this is a 5.5.x-only thing? Thanks! -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
context.xml created by Ant not discovered by Tomcat
I am attempting to get some JDBC parameters passed from Ant to a Tomcat .xml file. The idea is for an Ant target to create a file called context.xml and place it in /webapps//META-INF/ - and THERE IS NO .war FILE INVOLVED. Upon startup, Tomcat will discover the file in that /META-INF directory and copy/move it to /conf///, where the JDBC parameters in it will then be used for the application. However, this doesn't work. I have tried both Tomcat 5.0.28 and 5.5.9. I have read over the docs, and from them, it appears that what I want to do is correct in theory. I read thru the archives of this list, and I found a message seemingly asking the same things I am - as in, does context.xml have to be in a .war and where should it be placed for a web application; the response to this was: "The META-INF/context.xml file is only consulted if inside a .war. Otherwise, put the config.xml file in the webapps directory itself (if autodeploy is on, which is the default), or whatever directory you designate as the Host's appBase(tomcat 4), or the /conf/[engine name]/[host name] directory (tomcat 5)." I have moved context.xml to all the various places described above and restarted, all with no success. Same with renaming it - it doesn't appear to matter whether the file is called context.xml or .xml. So, IS there a way to do what I'm trying to do? Thanks! -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Session lost when switching from https to http in Tomcat 5.
"Anhony" <[EMAIL PROTECTED]> wrote: I am using Tomcat 5.0.28 Users log into my application from https://xxx.com/login.jsp. When submitted, I check for a valid userID/Password, create a session with getSession(), and then save the userID/Password in a session variable. The validated user is then returned to my main tools page. If the user then selects a link to a non secure page, http://..., I get a return of null when performing getSession( false) when trying to check that the user is valid. I am using Tomcat 5.0.16. An https session will have a different JSESSIONID than a http session, so Tomcat will think going from one to another is stating a new session. The JESSIONID is stored as a secure cookie when you start from https; get this cookie and re-add it to the response as a non-secure cookie (setSecure(false)), and then redirect to the http page. -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
s
I inherited an application recently that uses Tomcat 5.0.16. I know that previous to Tomcat 5, elements were in Tomcat's server.xml, but with Tomcat 5 it is recommended that elements be in their own .xml files under /conf/Catalina//. In this application, however, there are elements in BOTH places - when this is the case, does one overwrite the other, do they "cancel" one another out, what? Thanks! -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
question
I was recently given a new application to maitain. It uses Tomcat 5.0.16 and an Oracle database. I have used various versions of Tomcat to work with applications on two other kinds of databases, one of which uses DBCP, so I am somewhat familiar with the needed parameters and went to the Tomcat JNDI HOW-TO and the Tomcat Configuration Reference to refresh my memory on just what this was doing, but I have to admit I don't completely understand this - what I have included is the .xml file for the "new" application. What I don't understand is that first set of statements - it appears to be saying to look at the "database" of users in the tomcat-users.xml file in /conf - but why? That file hasn't been altered to contain anything but the defaults, and the person that gave me the "new" app was the one that set up Tomcat for it, so I assume its set up as needed! Also, why is there no factory in the 2nd in this file? I thought one was needed. The person that gave me this said DBCP was implemented, but I don't see how this would do it. factory org.apache.catalina.users.MemoryUserDatabaseFactory pathname conf/tomcat-users.xml maxWait 5000 maxActive 100 password url jdbc:oracle:thin:@ driverClassName oracle.jdbc.driver.OracleDriver maxIdle 5 username Could someone please explain to me just how this is using DBCP? Thanks!! -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
What does this mean? (Tomcat manager)
In using the Tomcat manager app with 5.0.16, we got some odd results when our backend database started acting funny and then crashed; I'm just curious as to whether the results we saw on the Tomcat manager indicate that there's something I should be looking at. Specifically, what I saw was that the "Current thread count" and "Current thread busy" on our http connector both specified NEGATIVE values soon after the database started having trouble, and on our https connector, the value listed for both items was "0". There were no errors in the catalina.out that appeared to be related to this - naturally, our database log had plenty of errors. But what I want to assure myself is that Tomcat was operating as intended - or does the display of either of those values have a particular significance to Tomcat? Thanks! -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: SocketException
"Peter Crowther" wrote: SEVERE: Remote Host /xxx.xxx.xxx.xxx SocketException: Invalid argument Hmm. An EINVAL from the socket layer. Are you running UNIX or Windows? Depending on which, I'd get hold of a system call tracer (Unix) or TDImon (Windows), monitor until you got the error, and then look through the logfile to see which system call was giving you the error back. That might give you some insight into what your system is doing when this occurs... I'm on a Solaris system. So I take it you mean something like truss or truss itself? I think I tried that once and gave up for some reason - I'll have to go back to my notes to see why, I could've just run out of time. Do you know why this error might occur? I think the thing I'm really looking for is to see if there's something I need to do on the server end, or if it's a client problem. And I never see the error on my test server, just on the production one that has a greater load - the error doesn't SEEM to hurt anything, but then again catalina.out classifies it as SEVERE, so maybe there's something to it that I'm missing! Thanks for your help! -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
SocketException
As I have mentioned in the past, I am still getting this error in my catalina.out: Nov 12, 2004 5:03:19 AM org.apache.tomcat.util.net.TcpWorkerThread runIt SEVERE: Remote Host /xxx.xxx.xxx.xxx SocketException: Invalid argument This is the complete text of the error - no stack trace, nothing more.I am using Tomcat 5.0.16 and this is the error I get; I got a similar error using 4.1.24. The error isn't constant, and I don't know how to reproduce it. I don't think it has anything to do with the number of open files or memory; I have looked into the possibility that the error might be due to either of those (after reading some comments I found after searching on Google and the archives of this list) and I think I've eliminated those possibilities. Does anyone have any idea what might cause this error? Thanks! -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Active connections in db pool
"Trond G. Ziarkowski" <[EMAIL PROTECTED]> wrote: Is there a way of finding out how many commections which are currently active when using a db connection pool? I got this error message in my logs: "Cannot get a connection, pool exhausted". When I increased the number of mxActive the error message of course disappeared. I tripled the number of connections just to make sure that it will not happen again for some while, but I would like to check the value from time to time so I can increase it yet again if that would be necessary or decrease it if I really needed just oldMax + 10. I had the same question - what I ended up doing was searching thru the docs about DBCP as well as reading the Wiki for DBCP. It was a few months back, but there I found some discussion about doing this and a .jsp that the developer used for testing - I'd post it for you, but I had to make a lot of alterations to it to get it to provide all the info I needed, and I'm afraid I didn't keep track of what I was adding and subtracting! Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
JAAS and Tomcat
Question - I have a webapp that does its own authentication. I use Tomcat as my servlet container. If I want to use JAAS for that application, with it doing authentication (and not Tomcat), I don't need to set Tomcat up for JAAS, do I? -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: SocketException
Hollerman Geralyn M wrote: I am getting this error in catalina.out: Sep 1, 2004 12:41:43 PM org.apache.tomcat.util.net.TcpWorkerThread runIt SEVERE: Remote Host /xxx.xxx.xxx.xxx SocketException: Invalid argument (this is the entire message) I am running Tomcat 5.0.16, going against a MySQL 4.0.17 database; JDK 1.4.2_05 is used on this Solaris 5.8 system. Apache (2.0.50) is used in a sort of uncommon way: all it does is redirect requests (using mod_rewrite), as there are at least 2 domain names that can be used to get to the same place - Tomcat is ultimately all that is used. Since the application I'm running is for students and the fall semester has recently begun, the load has increased significantly in the past few weeks. I have a development system, and I never see this error on there; I have only seen it on our production system, in the past few weeks, under the higher load. I *suspect* that the error is caused by some user action, as in pressing the "Back" button or repeatedly clicking somewhere - my suspicion is based on the fact that several of us in my building have tried to produce this error but are unable to, but then again, we know what to expect, what to do, and what not to do when it comes to this application! Some more info on this - I was looking around my logs files, and found that I also saw this message in Tomcat 4.1.24, but there was a different, longer error message: java.net.SocketException: Invalid argument at java.net.PlainSocketImpl.socketSetOption(Native Method) at java.net.PlainSocketImpl.setOption(PlainSocketImpl.java:240) at java.net.Socket.setTcpNoDelay(Socket.java:771) at org.apache.tomcat.util.net.PoolTcpEndpoint.setSocketOptions(PoolTcpEndpoint.java:468) at org.apache.tomcat.util.net.TcpWorkerThread.runIt(PoolTcpEndpoint.java:564) at org.apache.tomcat.util.threads.ThreadPool$ControlRunnable.run(ThreadPool.java:619) at java.lang.Thread.run(Thread.java:534) although, it appears that at the time this error message was generated, Tomcat was throwing an OutofMemory error - I'm not sure if one had anything to do with the other, but judging from the date on the log, that was an error we were getting often at that time... Hope this helps! -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
SocketException
I am getting this error in catalina.out: Sep 1, 2004 12:41:43 PM org.apache.tomcat.util.net.TcpWorkerThread runIt SEVERE: Remote Host /xxx.xxx.xxx.xxx SocketException: Invalid argument (this is the entire message) I am running Tomcat 5.0.16, going against a MySQL 4.0.17 database; JDK 1.4.2_05 is used on this Solaris 5.8 system. Apache (2.0.50) is used in a sort of uncommon way: all it does is redirect requests (using mod_rewrite), as there are at least 2 domain names that can be used to get to the same place - Tomcat is ultimately all that is used. Since the application I'm running is for students and the fall semester has recently begun, the load has increased significantly in the past few weeks. I have a development system, and I never see this error on there; I have only seen it on our production system, in the past few weeks, under the higher load. I *suspect* that the error is caused by some user action, as in pressing the "Back" button or repeatedly clicking somewhere - my suspicion is based on the fact that several of us in my building have tried to produce this error but are unable to, but then again, we know what to expect, what to do, and what not to do when it comes to this application! Has anyone seen this? Any ideas? Thanks! -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Tomcat Status page
Another question about the Tomcat status page - Concerning the list of requests for a particular connector - Is there anything in particular pointed to if one (or more) requests remains until a restart of the server? I have seen times in the "Time" column (which I understand indicates the elasped time in processing the request) that are in the hundreds of millions of ms. However, there ARE requests that come and go as expected. Do the long-lasting ones indicate an error, and if so, what would tell me if the error is in the code or some user or network problem? Thanks! -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Tomcat status
I have been looking for some information on the Tomcat status page in Tomcat 5 - can anyone tell me if information about how to read that page (past the obvious) is on-line somewhere? I must've missed it. The Tomcat docs talk about how it's a JMX proxy servlet, and a lot of the page is self-explanatory, but I don't know what some of the entries mean... Thanks! -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
JFluid
I saw the JFluid tool from Sun mentioned on this list - just out of curiousity, has anyone tried it? Any comments, tips, advice on using it? Thanks! -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Tomcat start-up
I started up my Tomcat 5.0.16 server the other afternoon at about 4pm. A few hours later, I happened to look at the log, and found these entries: May 18, 2004 6:42:47 PM org.apache.struts.util.PropertyMessageResources INFO: Initializing, config='org.apache.struts.taglib.html.LocalStrings', returnNull=true May 18, 2004 6:42:47 PM org.apache.struts.util.PropertyMessageResources INFO: Initializing, config='org.apache.struts.util.LocalStrings', returnNull=true May 18, 2004 6:42:47 PM org.apache.struts.util.PropertyMessageResources INFO: Initializing, config='org.apache.struts.taglib.bean.LocalStrings', returnNull=true May 18, 2004 6:42:47 PM org.apache.struts.util.PropertyMessageResources INFO: Initializing, config='org.apache.struts.taglib.html.LocalStrings', returnNull=true They appear to be perfectly normal initialization messages - but I'm not using Struts. What caused them to appear a few hours after I had started the server, rather than at startup with all the other initialization messages? Does this indicate a problem, or can I safely disregard these? Thanks! -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: maxThreads
At 03:03 AM 5/11/2004 +, Bill Barker wrote: By default, Tomcat will start maxThreads/2 threads incase it needs lots of threads in a hurry. Almost all of them will be blocked, so there generally isn't that much OS overhead in creating them. However, if you set the 'maxSpareThreads' attribute on the connector, then you can keep Tomcat down to a relatively small number of threads unless it actually has to handle a large number of requests. Ok, so if my maxThreads is 1000, at startup there will be 500 threads created? I thought it just created the number in minSpareThreads and built on that, no? In the situation we're using Tomcat, there will be a large number of requests - for about the next week, as students get their grades - and then things will calm down somewhat. Our system is forever exceeding maxThreads, and all increasing that attribute does is make the system exceed it faster. That doesn't seem correct. Lynn. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
maxThreads
Evidently, I'm misunderstanding something about the maxThreads attribute on the HTTP ; I saw from the docs that in Tomcat 5.0.19, this is the maximum number of request processing threads to be created by this connector. So, what that said to me was that if I wanted to allow lots of requests, I ought to make that number larger than the default, which the docs say is 200. I also assumed that this attribute, maxThreads, is similar to the "maxProcessors" attribute that was used in, for instance, Tomcat 4.0.4. When we were using Tomcat 4.0.4, we often got the message "No processor available, rejecting this connection", and the way I was told to correct that error was to increase maxProcessors, and this worked fine. But to correct another problem, we were told to upgrade to 4.1.24, and then to 5.0.16. However, I have noticed that with maxThreads, if I increase the value, the number of threads I see on a Solaris 8 system grows quickly, like it's trying to meet whatever limit I set; it's as if I make that value larger, thinking that that increase is what Tomcat needs, the number of threads gets to that limit quicker. This can't be right. What am I missing? Thanks! -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: secureCookie parameter for Tomcat 5.0.19?
Tim Funk <[EMAIL PROTECTED]> wrote: This was discussed a few times in the past and there might be a bugzilla report. If you submit a patch - it might have a chance at being committed - but the past discussions tried to discourage the proposed functionality. A workaround might be to "RESET" (that is, resend) the cookie without the SECURE flag turned on. -Tim [EMAIL PROTECTED] wrote: Hello configuration gurus, I recently upgraded to Tomcat 5.0.19, and I wonder if there is a solution to this problem yet. The problem is that in the Tomcat 4 (and apparently Tomcat 5) series, tomcat forces it so that if a session starts under https, then the session cookie is FORCED to be secure. I understand that this is to protect people from having hijacked sessions as a general case. However, there are some of us that manage this on our own (sending our own secure cookie that we use to make sure https pages aren't hijacked), and we would like our session cookie to be available to http AND https, as we move back and forth through those pages depending on the sensitivity of the content. If the session is created under http, then it is available to http and https. The problem only happens when the session is created under https. And, unfortunately, my app has quite a few places where the session NEEDS to be created under https, but needs to be available to http and https pages. This is EXACTLY what I just figured out was happening to me under 5.0.16; in my app, it looked like some session information wasn't getting passed. My solution was to get the secure cookie, make in "un-secure", and put it back so when my app switched to http after https authentication, that cookie would be found. But then, I'm fortunate in that there's only one place in my code where this happens, so I had a minimal amount of code to change. It would be nice if an https-generated cookie was available to both http and https on a configurable basis. -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Apache security certificate vs. Tomcat security certificate
I currently have Tomcat 5.0.16 running using the SSL connector and a self-signed certificate - I followed the directions in the Tomcat SSL HOW-TO in how to create the certificate and set up Tomcat for SSL. This is running with no problems in my development environment. I have been asked to put SSL on our production Tomcat. It is also v 5.0.16. My sysadmin gave me two files he uses for Apache, running with SSL, on another system; he said he thought I would need them. They are called server.crt and server.key - he said he was giving me the certificate and the key for it. How do I intergrate this with Tomcat? How does this correspond to what I've got for Tomcat - I presume that the .crt file is like the .cer file I exported from cacerts.jks (my keystore), but what does the .key file correspond to? How does that fit in? Thanks! -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
SocketException and NullPointerException
Folks, I have 2 errors in my catalina.out that I'm trying to chase down, and for both I've about come up with zero in terms of figuring out what's causing them and/or how to "fix" them. I'd appreciate ANY ideas/suggestions/comments! The first is of the form Mar 22, 2004 8:38:32 AM org.apache.tomcat.util.net.TcpWorkerThread runIt SEVERE: Remote Host /xxx.xxx.xxx.xxx SocketException: Invalid argument This one shows up somewhat sporadically - some days there will be several, other days, none. The second looks like StandardWrapperValve[uPortal]: Servlet.service() for servlet uPortal threw exception java.lang.RuntimeException: java.lang.NullPointerException java.lang.RuntimeException: java.lang.NullPointerException at org.apache.xalan.transformer.TransformerImpl.run(TransformerImpl.java:3364) at org.apache.xalan.transformer.TransformerHandlerImpl.endDocument(TransformerHandlerImpl.java:427) at org.jasig.portal.utils.SAX2FilterImpl.endDocument(SAX2FilterImpl.java:684) at org.jasig.portal.utils.SAX2BufferImpl.outputBuffer(SAX2BufferImpl.java:210) at org.jasig.portal.UserInstance.renderState(UserInstance.java:538) at org.jasig.portal.UserInstance.writeContent(UserInstance.java:180) at org.jasig.portal.PortalSessionManager.doGet(PortalSessionManager.java:246) at javax.servlet.http.HttpServlet.service(HttpServlet.java:743) [...] (I can give you the full trace, but it's long and ugly - I think you get the idea here) Now, I know this one has something to do with the XML parser and probably the stylesheet used - something that's passed is null for some reason, and as far as the program is concerned, it shouldn't be. What could cause this? I get this error from our production system - I'm running the same version of Tomcat (5.0.16) on a development system as well, and the error never shows up there. I have the same versions of XML parsers on both boxes - the .class files in them that are used in the stack trace (like TransformerImpl) are the same on both systems (yes, I checked.). I'd really like to get this fixed - the server appears to start-up fine, with no errors listed. Thanks for any feedback!! -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: cookies, Safari, and Tomcat
Aadi Deshpande wrote: Hollerman Geralyn M wrote: I'm trying to figure out some behavior I'm seeing only when I use Safari (v1.25? - downloaded from the Apple site last week) and Tomcat. This involves cookies. I am using Tomcat 5.0.16. I have written a servlet that sends a cookie back to the server for use later on; I can see this cookie when using Mozilla, Netscape, or IE when I look at a list of cookies. I also see the JESSIONID cookie that Tomcat uses in this list. However, when I use Safari, I see the JESSIONID cookie, but not the cookie I created. I'm not sure what the reason is for this; if I didn't do something in my code, I'd like to find out what it is, and for this I'd imagine I'd have to look thru the Tomcat source and see what's done when the JSESSIONID cookie is created. Unfortunately, I can't seem to find this. Has anyone seen this behavior before? How should I deal with it? Or can someone point me to the program where the JESSSIONID cookie for Tomcat is created? Thanks! What is the cookie version of the cookie that you created? We've noticed some odd behaviour when using cookies and that magically went away when we started using version 0 cookies. I have tried creating both version 0 and version 1 cookies; it doesn't appear to matter - the behavior is the same. Also, I have been told that "the bug that affected Safari and cookies has been repaired prior to (the version I am using) of Safari"; it's entirely possible that this is a different cookie problem with Safari, tho! -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
cookies, Safari, and Tomcat
I'm trying to figure out some behavior I'm seeing only when I use Safari (v1.25 - downloaded from the Apple site last week) and Tomcat. This involves cookies. I am using Tomcat 5.0.16. I have written a servlet that sends a cookie back to the server for use later on; I can see this cookie when using Mozilla, Netscape, or IE when I look at a list of cookies. I also see the JESSIONID cookie that Tomcat uses in this list. However, when I use Safari, I see the JESSIONID cookie, but not the cookie I created. I'm not sure what the reason is for this; if I didn't do something in my code, I'd like to find out what it is, and for this I'd imagine I'd have to look thru the Tomcat source and see what's done when the JSESSIONID cookie is created. Unfortunately, I can't seem to find this. Has anyone seen this behavior before? How should I deal with it? Or can someone point me to the program where the JESSSIONID cookie for Tomcat is created? Thanks! -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
SEVERE: in startup of TC 5.0.16
I am receiving a SEVERE: error in my startup of Tomcat 5.0.16; also, in my catalina_log I get the message: StandardContext[/balancer]Exception starting filter BalancerFilter java.lang.NoClassDefFoundError: org/apache/commons/digester/Digester at org.apache.webapp.balancer.RulesParser.createDigester(RulesParser.java:107) [...] None of this seems to effect the startup or running of TC 5.0.16 or any of my webapps that I can see, but I'd like to know if this indicates there's something wrong that I need to look into. I'm testing Tomcat 5.0.16 using JDK 1.4.2_b28 on a Solaris 8 system. Here is a snippet of my catalina.out: --- Jan 15, 2004 2:30:14 PM org.apache.catalina.core.StandardHost getDeployer INFO: Create Host deployer for direct deployment ( non-jmx ) Jan 15, 2004 2:30:14 PM org.apache.catalina.core.StandardHostDeployer install INFO: Installing web application at context path /jsp-examples from URL file:/home/gmh2441/uPortal/jakarta-tomcat-5.0.16/webapps/jsp-examples Jan 15, 2004 2:30:19 PM org.apache.catalina.core.StandardHostDeployer install INFO: Installing web application at context path /tomcat-docs from URL file:/home/gmh2441/uPortal/jakarta-tomcat-5.0.16/webapps/tomcat-docs Jan 15, 2004 2:30:21 PM org.apache.catalina.core.StandardHostDeployer install INFO: Installing web application at context path /servlets-examples from URL file:/home/gmh2441/uPortal/jakarta-tomcat-5.0.16/webapps/servlets-examples Jan 15, 2004 2:30:25 PM org.apache.catalina.core.StandardHostDeployer install INFO: Installing web application at context path /balancer from URL file:/home/gmh2441/uPortal/jakarta-tomcat-5.0.16/webapps/balancer Jan 15, 2004 2:30:27 PM org.apache.catalina.core.StandardContext start SEVERE: Error filterStart Jan 15, 2004 2:30:27 PM org.apache.catalina.core.StandardContext start SEVERE: Context startup failed due to previous errors Jan 15, 2004 2:30:29 PM org.apache.coyote.http11.Http11Protocol start INFO: Starting Coyote HTTP/1.1 on port 8080 Jan 15, 2004 2:30:32 PM org.apache.coyote.http11.Http11Protocol start INFO: Starting Coyote HTTP/1.1 on port 8443 Jan 15, 2004 2:30:32 PM org.apache.catalina.startup.Catalina start INFO: Server startup in 41919 ms --- My apps startup just fine, and there aren't any errors related to this in their logs. In looking thru the archives, I saw mention of this being related to commons-digester.jar, but nothing about what to do to "fix" this problem. Any suggestions? Thanks! -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
REPOST: /devices/pseudo/mm@0:null
While using Tomcat, I have often encountered a problem (a java.net.SocketException) with "Too many open files". I understand that to combat this, I need to up the limit on file descriptors in my Solaris 8 system (ulimit command). The system went from 256 -> 1024 descriptors, but I still get the error, and in listing the open files, I will see "/devices/pseudo/[EMAIL PROTECTED]:null" as being used by my Tomcat process - over and over and over. I take it the Tomcat process is writing to "null"? Is it supposed to do that - or so much of it? Or does this indicate a problem somewhere? Thanks! -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: /devices/pseudo/mm@0:null
Ralph Einfeldt wrote: Although my solaris days are long gone I would guess that this aren't 'real files' but memory mapped files. Are you shure that there are several handles on the same file, or is the same handle just reuses several times ? I agree, I think they're memory mapped files, but I don't see why there are all those entries. I did an "fuser -u /devices/pseudo/[EMAIL PROTECTED]:null" to see the processes associated with those "files", and got several PIDs, including my Tomcat process. "lsof" shows all the open files - and there are hundreds for Tomcat, and they all appear to be the same thing. Why are there so many, used over and over? -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
: Re: /devices/pseudo/mm@0:null
Antonio Fiol Bonnín wrote: Not that I have used Tomcat on Solaris a lot, but I have never seen that file. A general advice is to check that you close everything you open. That's just it - I never open /devices/pseudo/[EMAIL PROTECTED]:null! Something obviously IS opening it - but how do I find out what and close it? -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
/devices/pseudo/mm@0:null
While using Tomcat, I have often encountered a problem (a java.net.SocketException) with "Too many open files". I understand that to combat this, I need to up the limit on file descriptors in my Solaris 8 system (ulimit command). The system went from 256 -> 1024 descriptors, but I still get the error, and in listing the open files, I will see "/devices/pseudo/[EMAIL PROTECTED]:null" as being used by my Tomcat process - over and over and over. I take it the Tomcat process is writing to "null"? Is it supposed to do that - or so much of it? Or does this indicate a problem somewhere? Thanks! -- Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Tomcat problems every morning.
On Sun, 01 Jun 2003 18:14:05 -0700, Hemendra Rana wrote > Are you using mysql database on the backend too. Someone here > pointed out that by default, mysql driver times out after every 8 > hours. I still needs to verify whether that is the cause of my > problem though. Will let you know of my finds. Why, yes, I am. And I thought perhaps MySQL was the problem - but I was told by some people using Oracle and PostgreSQL databases that they'd seen the problem as well. I tried putting in some code to set the driver time-out value, but evidently I didn't do it correctly, as the problem persisted. I don't see the error with versions of Tomcat previous to 4.1.x, tho. Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Tomcat problems every morning.
On Sun, 01 Jun 2003 10:31:15 -0700, Hemendra Rana wrote > I am running Tomcat 4.1.24 server on Linux and every morning I get > the following exception. Things work perfectly fine throughout the > day after I restart the Tomcat server. Any ideas? Could this be a > logger issue, because my server logs rotate every night. ... > 2003-06-01 08:32:31 ApplicationDispatcher[] Servlet.service() for > servlet jsp threw exception > org.apache.jasper.JasperException: Communication link failure: > java.io.IOException I had this same error using 4.1.12 - it seemed like it would happen after about every 8 hours - a default timeout, perhaps? I was never able to find the reason for it - the advice I got was to re-start the server every morning, and I really don't like doing that... Any ideas? Thanks! Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Thanks to John and Geralyn
On Sun, 1 Jun 2003 14:56:19 +0800, joe wrote > coyote. What should be the proper way to build mod-jk2? or can i > just use the binary version available on apache website which is > meant for apache 2.0.42, wherelse i'm using > 2.0.46? Joe, I used both a version of mod_jk2 that my sysadmin built (he is root on the system - and he's built Apache connectors for other things) and a binary (I think it was said it was for Apache 2.0.43? There were only 3 of them there for Solaris8 - I believe this was on the standard Apache-Tomcat connectors site) that I downloaded; both worked for me running Apache 2.0.45. As far as which was fast or more efficient, that's to be determined later! Lynn Hollerman. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
