Re: Mod_jk working for failover
Thank you Rainer, Now Apache is working with disable mode. Maybe later it changed again. :) 2011/6/12 Rainer Jung rainer.j...@kippdata.de On 12.06.2011 09:30, Manuel Fernández Panzuela wrote: Is possible to combine both attributes distance - activation ? Distance looks like a very interesting attribute. Yes, of course you can combine. As long as you don't cnfuse yourself because it gets too complicated. Distace (e.g. 0 and 1) will lead to normally everything being sent to the close node, but when you switch its activation from active to disabled, new sessions will be created on the node with higher distance. Switching to stopped will result in every request being handled by the node with higher distance. Regards, Rainer 2011/6/11 Rainer Jung rainer.j...@kippdata.de On 08.06.2011 12:12, Felix Schumacher wrote: Am Mittwoch, den 08.06.2011, 09:47 +0200 schrieb Manuel Fernández Panzuela: Yes Felix, you are right i am working with 2 instances. I want primary node receives all requests and on failure send requests to the secondary node. Is it possible with your recomendation ? You may also want to read http://tomcat.apache.org/connectors-doc/generic_howto/loadbalancers.html if http://tomcat.apache.org/connectors-doc/reference/workers.html was not helpful. worker.43m2g1.activation=d degraded ??? is a condition ? then all requests will be redirected to 43m2g1 disabled The attribute distance provides an alternative approach. Regards, Rainer - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
Re: Mod_jk working for failover
Is possible to combine both attributes distance - activation ? Distance looks like a very interesting attribute. Regards 2011/6/11 Rainer Jung rainer.j...@kippdata.de On 08.06.2011 12:12, Felix Schumacher wrote: Am Mittwoch, den 08.06.2011, 09:47 +0200 schrieb Manuel Fernández Panzuela: Yes Felix, you are right i am working with 2 instances. I want primary node receives all requests and on failure send requests to the secondary node. Is it possible with your recomendation ? You may also want to read http://tomcat.apache.org/connectors-doc/generic_howto/loadbalancers.html if http://tomcat.apache.org/connectors-doc/reference/workers.html was not helpful. worker.43m2g1.activation=d degraded ??? is a condition ? then all requests will be redirected to 43m2g1 disabled The attribute distance provides an alternative approach. Regards, Rainer - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
Re: Mod_jk working for failover
On 12.06.2011 09:30, Manuel Fernández Panzuela wrote: Is possible to combine both attributes distance - activation ? Distance looks like a very interesting attribute. Yes, of course you can combine. As long as you don't cnfuse yourself because it gets too complicated. Distace (e.g. 0 and 1) will lead to normally everything being sent to the close node, but when you switch its activation from active to disabled, new sessions will be created on the node with higher distance. Switching to stopped will result in every request being handled by the node with higher distance. Regards, Rainer 2011/6/11 Rainer Jung rainer.j...@kippdata.de On 08.06.2011 12:12, Felix Schumacher wrote: Am Mittwoch, den 08.06.2011, 09:47 +0200 schrieb Manuel Fernández Panzuela: Yes Felix, you are right i am working with 2 instances. I want primary node receives all requests and on failure send requests to the secondary node. Is it possible with your recomendation ? You may also want to read http://tomcat.apache.org/connectors-doc/generic_howto/loadbalancers.html if http://tomcat.apache.org/connectors-doc/reference/workers.html was not helpful. worker.43m2g1.activation=d degraded ??? is a condition ? then all requests will be redirected to 43m2g1 disabled The attribute distance provides an alternative approach. Regards, Rainer - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
Re: Mod_jk working for failover
On 08.06.2011 12:12, Felix Schumacher wrote: Am Mittwoch, den 08.06.2011, 09:47 +0200 schrieb Manuel Fernández Panzuela: Yes Felix, you are right i am working with 2 instances. I want primary node receives all requests and on failure send requests to the secondary node. Is it possible with your recomendation ? You may also want to read http://tomcat.apache.org/connectors-doc/generic_howto/loadbalancers.html if http://tomcat.apache.org/connectors-doc/reference/workers.html was not helpful. worker.43m2g1.activation=d degraded ??? is a condition ? then all requests will be redirected to 43m2g1 disabled The attribute distance provides an alternative approach. Regards, Rainer - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
Re: Mod_jk working for failover
my setup: # Define list of workers that will be used for mapping requests worker.list=loadbalancer43,status # Definicion de plantilla worker.template43.port=8009 worker.template43.type=ajp13 worker.template43.lbfactor=1 worker.template43.ping_timeout=1000 worker.template43.socket_timeout=10 worker.template43.ping_mode=A worker.template43.connection_pool_timeout=600 # NODO 43 m1g1 worker.43m1g1.reference=worker.template43 worker.43m1g1.host=xx.xx.xx.xx # NODO 43 m2g1 worker.43m2g1.reference=worker.template43 worker.43m2g1.host=xx.xx.xx.xx # Balanceador correspodiente a todos los grupos worker.loadbalancer43.type=lb worker.loadbalancer43.balance_workers=43m1g1,43m2g1 worker.loadbalancer43.sticky_session=1 Now is balancing between 4 instances, but I need to work against 43m1g1 node. If 43m1g1 node fails, then petitions will be redirects to 43m2g1 node. Thank you 2011/6/7 Christopher Schultz ch...@christopherschultz.net -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Manuel. On 6/7/2011 12:38 PM, Manuel Fernández Panzuela wrote: I want to work with mod_jk in failover mode. I do not want to balancing petitions until the node 1 is unavailable. If principal node (node 1) is down, then petitions will be redirected to node 2. How I can do it? What have you tried so far? - -chris -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk3uaLUACgkQ9CaO5/Lv0PD6TACcC6239gZILU+6OGpT/xsh9F0I 6osAmgLvY4GYifndETBYP3z5JbznRYal =Jp0Z -END PGP SIGNATURE- - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
Re: Mod_jk working for failover
Am Mittwoch, den 08.06.2011, 08:41 +0200 schrieb Manuel Fernández Panzuela: my setup: # Define list of workers that will be used for mapping requests worker.list=loadbalancer43,status # Definicion de plantilla worker.template43.port=8009 worker.template43.type=ajp13 worker.template43.lbfactor=1 worker.template43.ping_timeout=1000 worker.template43.socket_timeout=10 worker.template43.ping_mode=A worker.template43.connection_pool_timeout=600 # NODO 43 m1g1 worker.43m1g1.reference=worker.template43 worker.43m1g1.host=xx.xx.xx.xx # NODO 43 m2g1 worker.43m2g1.reference=worker.template43 worker.43m2g1.host=xx.xx.xx.xx # Balanceador correspodiente a todos los grupos worker.loadbalancer43.type=lb worker.loadbalancer43.balance_workers=43m1g1,43m2g1 worker.loadbalancer43.sticky_session=1 Now is balancing between 4 instances, but I need to work against 43m1g1 node. I only see two instances. If 43m1g1 node fails, then petitions will be redirects to 43m2g1 node. Look at http://tomcat.apache.org/connectors-doc/reference/workers.html and search for activation and redirect properties. You will have to add something like worker.43m2g1.activation=d worker.43m1g1.redirect=43m2g1 to your worker.properties. Felix Thank you 2011/6/7 Christopher Schultz ch...@christopherschultz.net -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Manuel. On 6/7/2011 12:38 PM, Manuel Fernández Panzuela wrote: I want to work with mod_jk in failover mode. I do not want to balancing petitions until the node 1 is unavailable. If principal node (node 1) is down, then petitions will be redirected to node 2. How I can do it? What have you tried so far? - -chris -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk3uaLUACgkQ9CaO5/Lv0PD6TACcC6239gZILU+6OGpT/xsh9F0I 6osAmgLvY4GYifndETBYP3z5JbznRYal =Jp0Z -END PGP SIGNATURE- - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
Re: Mod_jk working for failover
Yes Felix, you are right i am working with 2 instances. I want primary node receives all requests and on failure send requests to the secondary node. Is it possible with your recomendation ? worker.43m2g1.activation=d degraded ??? is a condition ? then all requests will be redirected to 43m2g1 ?? worker.43m1g1.redirect=43m2g1 Do I need more changes in my setup? Regards 2011/6/8 Felix Schumacher felix.schumac...@internetallee.de Am Mittwoch, den 08.06.2011, 08:41 +0200 schrieb Manuel Fernández Panzuela: my setup: # Define list of workers that will be used for mapping requests worker.list=loadbalancer43,status # Definicion de plantilla worker.template43.port=8009 worker.template43.type=ajp13 worker.template43.lbfactor=1 worker.template43.ping_timeout=1000 worker.template43.socket_timeout=10 worker.template43.ping_mode=A worker.template43.connection_pool_timeout=600 # NODO 43 m1g1 worker.43m1g1.reference=worker.template43 worker.43m1g1.host=xx.xx.xx.xx # NODO 43 m2g1 worker.43m2g1.reference=worker.template43 worker.43m2g1.host=xx.xx.xx.xx # Balanceador correspodiente a todos los grupos worker.loadbalancer43.type=lb worker.loadbalancer43.balance_workers=43m1g1,43m2g1 worker.loadbalancer43.sticky_session=1 Now is balancing between 4 instances, but I need to work against 43m1g1 node. I only see two instances. If 43m1g1 node fails, then petitions will be redirects to 43m2g1 node. Look at http://tomcat.apache.org/connectors-doc/reference/workers.html and search for activation and redirect properties. You will have to add something like worker.43m2g1.activation=d worker.43m1g1.redirect=43m2g1 to your worker.properties. Felix Thank you 2011/6/7 Christopher Schultz ch...@christopherschultz.net -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Manuel. On 6/7/2011 12:38 PM, Manuel Fernández Panzuela wrote: I want to work with mod_jk in failover mode. I do not want to balancing petitions until the node 1 is unavailable. If principal node (node 1) is down, then petitions will be redirected to node 2. How I can do it? What have you tried so far? - -chris -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk3uaLUACgkQ9CaO5/Lv0PD6TACcC6239gZILU+6OGpT/xsh9F0I 6osAmgLvY4GYifndETBYP3z5JbznRYal =Jp0Z -END PGP SIGNATURE- - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
Re: Mod_jk working for failover
Am Mittwoch, den 08.06.2011, 09:47 +0200 schrieb Manuel Fernández Panzuela: Yes Felix, you are right i am working with 2 instances. I want primary node receives all requests and on failure send requests to the secondary node. Is it possible with your recomendation ? You may also want to read http://tomcat.apache.org/connectors-doc/generic_howto/loadbalancers.html if http://tomcat.apache.org/connectors-doc/reference/workers.html was not helpful. worker.43m2g1.activation=d degraded ??? is a condition ? then all requests will be redirected to 43m2g1 disabled ?? worker.43m1g1.redirect=43m2g1 Do I need more changes in my setup? I don't think so. Felix Regards 2011/6/8 Felix Schumacher felix.schumac...@internetallee.de Am Mittwoch, den 08.06.2011, 08:41 +0200 schrieb Manuel Fernández Panzuela: my setup: # Define list of workers that will be used for mapping requests worker.list=loadbalancer43,status # Definicion de plantilla worker.template43.port=8009 worker.template43.type=ajp13 worker.template43.lbfactor=1 worker.template43.ping_timeout=1000 worker.template43.socket_timeout=10 worker.template43.ping_mode=A worker.template43.connection_pool_timeout=600 # NODO 43 m1g1 worker.43m1g1.reference=worker.template43 worker.43m1g1.host=xx.xx.xx.xx # NODO 43 m2g1 worker.43m2g1.reference=worker.template43 worker.43m2g1.host=xx.xx.xx.xx # Balanceador correspodiente a todos los grupos worker.loadbalancer43.type=lb worker.loadbalancer43.balance_workers=43m1g1,43m2g1 worker.loadbalancer43.sticky_session=1 Now is balancing between 4 instances, but I need to work against 43m1g1 node. I only see two instances. If 43m1g1 node fails, then petitions will be redirects to 43m2g1 node. Look at http://tomcat.apache.org/connectors-doc/reference/workers.html and search for activation and redirect properties. You will have to add something like worker.43m2g1.activation=d worker.43m1g1.redirect=43m2g1 to your worker.properties. Felix Thank you 2011/6/7 Christopher Schultz ch...@christopherschultz.net -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Manuel. On 6/7/2011 12:38 PM, Manuel Fernández Panzuela wrote: I want to work with mod_jk in failover mode. I do not want to balancing petitions until the node 1 is unavailable. If principal node (node 1) is down, then petitions will be redirected to node 2. How I can do it? What have you tried so far? - -chris -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk3uaLUACgkQ9CaO5/Lv0PD6TACcC6239gZILU+6OGpT/xsh9F0I 6osAmgLvY4GYifndETBYP3z5JbznRYal =Jp0Z -END PGP SIGNATURE- - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
Re: Mod_jk working for failover
Woowww Now, mod_jk is working perfectly Thank you so much FELIX !!! Best regards 2011/6/8 Felix Schumacher felix.schumac...@internetallee.de Am Mittwoch, den 08.06.2011, 09:47 +0200 schrieb Manuel Fernández Panzuela: Yes Felix, you are right i am working with 2 instances. I want primary node receives all requests and on failure send requests to the secondary node. Is it possible with your recomendation ? You may also want to read http://tomcat.apache.org/connectors-doc/generic_howto/loadbalancers.html if http://tomcat.apache.org/connectors-doc/reference/workers.html was not helpful. worker.43m2g1.activation=d degraded ??? is a condition ? then all requests will be redirected to 43m2g1 disabled ?? worker.43m1g1.redirect=43m2g1 Do I need more changes in my setup? I don't think so. Felix Regards 2011/6/8 Felix Schumacher felix.schumac...@internetallee.de Am Mittwoch, den 08.06.2011, 08:41 +0200 schrieb Manuel Fernández Panzuela: my setup: # Define list of workers that will be used for mapping requests worker.list=loadbalancer43,status # Definicion de plantilla worker.template43.port=8009 worker.template43.type=ajp13 worker.template43.lbfactor=1 worker.template43.ping_timeout=1000 worker.template43.socket_timeout=10 worker.template43.ping_mode=A worker.template43.connection_pool_timeout=600 # NODO 43 m1g1 worker.43m1g1.reference=worker.template43 worker.43m1g1.host=xx.xx.xx.xx # NODO 43 m2g1 worker.43m2g1.reference=worker.template43 worker.43m2g1.host=xx.xx.xx.xx # Balanceador correspodiente a todos los grupos worker.loadbalancer43.type=lb worker.loadbalancer43.balance_workers=43m1g1,43m2g1 worker.loadbalancer43.sticky_session=1 Now is balancing between 4 instances, but I need to work against 43m1g1 node. I only see two instances. If 43m1g1 node fails, then petitions will be redirects to 43m2g1 node. Look at http://tomcat.apache.org/connectors-doc/reference/workers.html and search for activation and redirect properties. You will have to add something like worker.43m2g1.activation=d worker.43m1g1.redirect=43m2g1 to your worker.properties. Felix Thank you 2011/6/7 Christopher Schultz ch...@christopherschultz.net -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Manuel. On 6/7/2011 12:38 PM, Manuel Fernández Panzuela wrote: I want to work with mod_jk in failover mode. I do not want to balancing petitions until the node 1 is unavailable. If principal node (node 1) is down, then petitions will be redirected to node 2. How I can do it? What have you tried so far? - -chris -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk3uaLUACgkQ9CaO5/Lv0PD6TACcC6239gZILU+6OGpT/xsh9F0I 6osAmgLvY4GYifndETBYP3z5JbznRYal =Jp0Z -END PGP SIGNATURE- - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
Re: Mod_jk working for failover
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Manuel. On 6/7/2011 12:38 PM, Manuel Fernández Panzuela wrote: I want to work with mod_jk in failover mode. I do not want to balancing petitions until the node 1 is unavailable. If principal node (node 1) is down, then petitions will be redirected to node 2. How I can do it? What have you tried so far? - -chris -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk3uaLUACgkQ9CaO5/Lv0PD6TACcC6239gZILU+6OGpT/xsh9F0I 6osAmgLvY4GYifndETBYP3z5JbznRYal =Jp0Z -END PGP SIGNATURE- - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
