From: <dep...@al...> - 2010-11-18 10:44:45
|
Hi, We are having a problem starting SmartFrog under some accounts. Here is the output that we get when running sfDaemon from Windows command line: ================================================================================== D:\tmp\Externals\Java\smartfrog\bin>sfDaemon SmartFrog 3.18.002 (2010-09-20 04:13:24 EDT) (C) Copyright 1998-2010 Hewlett-Packard Development Company, LP 2010/11/18 12:13:14:859 EET [WARN ][main] SFCORE_LOG - SmartFrog security is NOT active 2010/11/18 12:13:14:866 EET [DEBUG][main] SFCORE_LOG - Security Manager is java. lang.SecurityManager@1a73d3c<mailto:lang.SecurityManager@1a73d3c> 2010/11/18 12:13:14:868 EET [DEBUG][main] SFCORE_LOG - Using java security polic y: =D:\tmp\Externals\Java\smartfrog\private\sf.no.security.policy 2010/11/18 12:13:14:878 EET [WARN ][main] SFCORE_LOG - Possible problem with cla sspath: 3 occurrences for smartfrog- D:\tmp\Externals\Java\smartfrog\lib\smartfrog-src-3.18.002.jar D:\tmp\Externals\Java\smartfrog\lib\smartfrog-jdocs-3.18.002.jar D:\tmp\Externals\Java\smartfrog\lib\smartfrog-3.18.002.jar 2010/11/18 12:13:14:895 EET [DEBUG][main] SFCORE_LOG - Network test localhost: h ostname 'BY0013D4B30637.alliedtesting.com', ip '192.168.198.99', [Successful], 1ms 2010/11/18 12:13:14:902 EET [DEBUG][main] SFCORE_LOG - Command Line args: [] SmartFrogDeploymentException: unnamed component. SmartFrogRuntimeException:: Sma rtFrog daemon could not start because another instance is running, cause: java.r mi.server.ExportException: Port already in use: 3800; nested exception is: java.net.BindException: Address already in use: JVM_Bind, SmartFrog 3.18 .002 (2010-09-20 04:13:24 EDT), cause: SmartFrogRuntimeException:: SmartFrog dae mon could not start because another instance is running, cause: java.rmi.serv er.ExportException: Port already in use: 3800; nested exception is: java.net.BindException: Address already in use: JVM_Bind, SmartFrog 3 .18.002 (2010-09-20 04:13:24 EDT) 2010/11/18 12:13:16:263 EET [DEBUG][main] SFCORE_LOG - SmartFrog System.Exit cod e: 1, v3.18.002 D:\tmp\Externals\Java\smartfrog\bin> ================================================================================== We get the same exception when launching SmartFrog using the test harness. We have checked the port and found that it is not in use at any point and that no SmartFrog daemon instances are running in the background. The most remarkable thing is that in multi-user environments the problem occurs only with certain accounts. When SmartFrog is launched on the same machine from the same location by a different user, we don't see this exception. What is also puzzling is that the problem emerged suddenly and for no apparent reason: we are not aware of any changes to the system that could have occasioned this error. This happens with SmartFrog v. 3.18.002 running on at least three different machines under Windows 7 and Windows Server 2008 R2 Standard with Java SE v. 1.6 runtime environment. The problem account has administrative privileges. Any help would be appreciated. Dmitry |
From: Steve L. <ste...@hp...> - 2010-12-07 10:32:41
|
On 18/11/10 10:44, dep...@al... wrote: > > > Hi, > > We are having a problem starting SmartFrog under some accounts. Here is > the output that we get when running sfDaemon from Windows command line: > > ================================================================================== > > > D:\tmp\Externals\Java\smartfrog\bin>sfDaemon > > SmartFrog 3.18.002 (2010-09-20 04:13:24 EDT) > > (C) Copyright 1998-2010 Hewlett-Packard Development Company, LP > > 2010/11/18 12:13:14:859 EET [WARN ][main] SFCORE_LOG - SmartFrog > security is NOT > > active > > 2010/11/18 12:13:14:866 EET [DEBUG][main] SFCORE_LOG - Security Manager > is java. > > lang.SecurityManager@1a73d3c <mailto:lang.SecurityManager@1a73d3c> > > 2010/11/18 12:13:14:868 EET [DEBUG][main] SFCORE_LOG - Using java > security polic > > y: =D:\tmp\Externals\Java\smartfrog\private\sf.no.security.policy > > 2010/11/18 12:13:14:878 EET [WARN ][main] SFCORE_LOG - Possible problem > with cla > > sspath: > > 3 occurrences for smartfrog- > > D:\tmp\Externals\Java\smartfrog\lib\smartfrog-src-3.18.002.jar > > D:\tmp\Externals\Java\smartfrog\lib\smartfrog-jdocs-3.18.002.jar > > D:\tmp\Externals\Java\smartfrog\lib\smartfrog-3.18.002.jar > > 2010/11/18 12:13:14:895 EET [DEBUG][main] SFCORE_LOG - Network test > localhost: h > > ostname 'BY0013D4B30637.alliedtesting.com', ip '192.168.198.99', > [Successful], > > 1ms > > 2010/11/18 12:13:14:902 EET [DEBUG][main] SFCORE_LOG - Command Line > args: [] > > SmartFrogDeploymentException: unnamed component. > SmartFrogRuntimeException:: Sma > > rtFrog daemon could not start because another instance is running, > cause: java.r > > mi.server.ExportException: Port already in use: 3800; nested exception is: > > java.net.BindException: Address already in use: JVM_Bind, SmartFrog 3.18 > > .002 (2010-09-20 04:13:24 EDT), cause: SmartFrogRuntimeException:: > SmartFrog dae > > mon could not start because another instance is running, cause: > java.rmi.serv > > er.ExportException: Port already in use: 3800; nested exception is: > > java.net.BindException: Address already in use: JVM_Bind, SmartFrog 3 > Usually means something is on the port. jps -v lists the Java processes, netstat -a -p will tie the port in use to the process to assign blame. If there's some problem with SF not shutting down, open a JIRA issue and we'll see what we can do to diagnose the problem, it may be some child component isn't shutting down, and the process compound may need some improvements in detecting and reacting to this |
From: <dep...@al...> - 2010-12-07 10:48:55
|
> On 18/11/10 10:44, dep...@al... wrote: > > > > > > Hi, > > > > We are having a problem starting SmartFrog under some accounts. Here > > is the output that we get when running sfDaemon from Windows > command line: > > > > > Usually means something is on the port. > > > jps -v lists the Java processes, netstat -a -p will tie the port in use to the > process to assign blame. > > If there's some problem with SF not shutting down, open a JIRA issue and > we'll see what we can do to diagnose the problem, it may be some child > component isn't shutting down, and the process compound may need some > improvements in detecting and reacting to this We checked the port and checked for running SmartFrog processes, but found nothing. The problem manifests itself with one particular Windows Server 2008 account. We even had two users log in at the same time and take turns trying to launch SmartFrog. One always succeeded, while the other always failed. Dmitry |
From: <dep...@al...> - 2010-12-07 13:17:25
|
> On 07/12/10 10:48, dep...@al... wrote: > > > We checked the port and checked for running SmartFrog processes, but > found nothing. The problem manifests itself with one particular Windows > Server 2008 account. We even had two users log in at the same time and take > turns trying to launch SmartFrog. One always succeeded, while the other > always failed. > > well, that looks like the problem: whoever logs in first and starts the process > wins. No, that wasn't the case. One user could start SF repeatedly, at any time, while the other (the one with the problem account) could not start it at any time. There is a way to start SF on a different port, it can be set in default.ini > > org.smartfrog.sfcore.processcompound.sfRootLocatorPort=3800 > > if the other user gets a different port, there's no conflict on what is a system > wide resource. We tried changing the port number, to no avail. I have googled this issue, but didn't find anything specific to this situation. However, it seems that this "port in use" error can in some cases be misleading, indicating some other problem, such as a firewall issue. |
From: Steve L. <ste...@hp...> - 2010-12-07 17:02:26
|
On 07/12/10 13:17, dep...@al... wrote: >> On 07/12/10 10:48, dep...@al... wrote: >> >> > We checked the port and checked for running SmartFrog processes, but >> found nothing. The problem manifests itself with one particular Windows >> Server 2008 account. We even had two users log in at the same time and take >> turns trying to launch SmartFrog. One always succeeded, while the other >> always failed. >> >> well, that looks like the problem: whoever logs in first and starts the process >> wins. > > No, that wasn't the case. One user could start SF repeatedly, at any time, while the other (the one with the problem account) could not start it at any time. > > There is a way to start SF on a different port, it can be set in default.ini >> >> org.smartfrog.sfcore.processcompound.sfRootLocatorPort=3800 >> >> if the other user gets a different port, there's no conflict on what is a system >> wide resource. > > We tried changing the port number, to no avail. > > I have googled this issue, but didn't find anything specific to this situation. However, it seems that this "port in use" error can in some cases be misleading, indicating some other problem, such as a firewall issue. > That's interesting. "Connection Refused" I associate with firewalls, but Port in Use usually means something is there. If you get it, try telnet-ing to the port and seeing what's there, and try netstat -a -p to see what's on the port -that's my usual first step in tracking down port use problems. -steve |
From: Guijarro, J. <jul...@hp...> - 2010-12-07 17:16:23
|
>> well, that looks like the problem: whoever logs in first and starts >> the process wins. Do you mean that you are login in into two accounts in the same machine and starting a daemon in each of the accounts at the same time on the same port? One machine, 1 port. You cannot have two daemon running in the same port in the same machine. It does not matter if they are in different accounts. If you need a daemon per account then you could have the second daemon joining the first one, but only one will be the root daemon. Is this what you are trying to do, one daemon per account? Thanks and best regards, Julio Guijarro -----Original Message----- From: Steve Loughran [mailto:ste...@hp...] Sent: 07 December 2010 17:02 To: dep...@al... Cc: sma...@li...; pga...@al... Subject: Re: [Smartfrog-users] "Port already in use" exception when starting SmartFrog On 07/12/10 13:17, dep...@al... wrote: >> On 07/12/10 10:48, dep...@al... wrote: >> >> > We checked the port and checked for running SmartFrog processes, >> but found nothing. The problem manifests itself with one particular >> Windows Server 2008 account. We even had two users log in at the same >> time and take turns trying to launch SmartFrog. One always succeeded, >> while the other always failed. >> >> well, that looks like the problem: whoever logs in first and starts >> the process wins. > > No, that wasn't the case. One user could start SF repeatedly, at any time, while the other (the one with the problem account) could not start it at any time. > > There is a way to start SF on a different port, it can be set in > default.ini >> >> org.smartfrog.sfcore.processcompound.sfRootLocatorPort=3800 >> >> if the other user gets a different port, there's no conflict on what >> is a system wide resource. > > We tried changing the port number, to no avail. > > I have googled this issue, but didn't find anything specific to this situation. However, it seems that this "port in use" error can in some cases be misleading, indicating some other problem, such as a firewall issue. > That's interesting. "Connection Refused" I associate with firewalls, but Port in Use usually means something is there. If you get it, try telnet-ing to the port and seeing what's there, and try netstat -a -p to see what's on the port -that's my usual first step in tracking down port use problems. -steve ------------------------------------------------------------------------------ What happens now with your Lotus Notes apps - do you make another costly upgrade, or settle for being marooned without product support? Time to move off Lotus Notes and onto the cloud with Force.com, apps are easier to build, use, and manage than apps on traditional platforms. Sign up for the Lotus Notes Migration Kit to learn more. http://p.sf.net/sfu/salesforce-d2d _______________________________________________ Smartfrog-users mailing list Sma...@li... https://lists.sourceforge.net/lists/listinfo/smartfrog-users |
From: <dep...@al...> - 2010-12-08 11:15:00
|
> >> well, that looks like the problem: whoever logs in first and starts > >> the process wins. > > Do you mean that you are login in into two accounts in the same machine and > starting a daemon in each of the accounts at the same time on the same > port? > > One machine, 1 port. You cannot have two daemon running in the same port > in the same machine. It does not matter if they are in different accounts. No, I mean that one user would successfully start _and stop_ the daemon. The other could not do that at any point, getting the "port already in use error." Ironically, the problem account was that of our system administrator. I am sure that he knew how to ascertain that the port was not, in fact, in use. > If you need a daemon per account then you could have the second daemon > joining the first one, but only one will be the root daemon. Is this what you > are trying to do, one daemon per account? No, we were just trying to get the daemon to run under the sysadmin's account. Regards, Dmitry |