After some time after starting I get this message:
"general protection ip:7f9cb07a88e0 sp:7fffc69beff0 error:0 in radiusd[7f9cb078a000+46000]"
and my radiusd service stops working. I have no major load on the server, but it stops and I don't know why is this message popping.
I'm using Centos 6.5 with latest updates and used Mfawa Installer script. I did some radiusd and mysql tweaks: Basicly increasing the number of "max_servers =" and "num_sql_socks ="
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Every 10 minutes when "/var/www/html/cake2/rd_cake/Console/cake -app /var/www/html/cake2/rd_cake Freeradius >> /dev/null 2>&1" is executed by cron I get this error in radius.log:
Wed Sep 24 19:20:05 2014 : Info: Loaded virtual server dynamic_client_server
Wed Sep 24 19:20:05 2014 : Info: Loaded virtual server inner-tunnel
Wed Sep 24 19:20:05 2014 : Info: Ready to process requests.
Wed Sep 24 19:20:05 2014 : Info: Loaded virtual server <default>
Wed Sep 24 19:20:05 2014 : Info: Loaded virtual server dynamic_client_server
Wed Sep 24 19:20:05 2014 : Info: Loaded virtual server inner-tunnel
Wed Sep 24 19:20:05 2014 : Error: Failed binding to authentication address * port 1812: Address already in use
Wed Sep 24 19:20:05 2014 : Error: /etc/raddb/radiusd.conf[273]: Error binding to port for 0.0.0.0 port 1812
It's like this line:
exec("sudo /var/www/html/cake2/rd_cake/Setup/Scripts/radmin_wrapper.pl start freeradius");
in the file:
/var/www/html/cake2/rd_cake/Console/Command/FreeradiusShell.php
is executed twice, because radius server is stopped, then starred -- everything goes well and then It's stared again but the listening port is already occupied.
P.S. When I restart radius from command line with the command: service radiusd restart, I don't see the error message about the port binding!
Last edit: nikonaum 2014-09-24
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I've went through all the forum and could not find person(post) with my problem. Just to get me right.
1. I get the GENERAL PROTECTION error message every 10 minutes -- the period of time in which "/var/www/html/cake2/rd_cake/Console/cake -app /var/www/html/cake2/rd_cake Freeradius >> /dev/null 2>&1" is executed by cron.d/rd.
2. The radiusd service does not stop after the error message but in some future (random (at least I could not see any pattern)) time.
3. I cannot see any load problems with my server -- enough processing and memory resources available, load is 0.5.
4. I did some radiusd and mysqld optimizations as advised here: http://www.slideshare.net/dashohoxha/pppoe-with-mikrotik-and-radius-16009022
5. Something constantly makes the file /var/www/html/cake2/rd_cake/tmp/cache/models/myapp_cake_model_default_rd_checks and
/var/www/html/cake2/rd_cake/tmp/cache/persistent/myapp_cake_core_method_cache
I'm not sure for the correct names of the files but one file in each of these directory suddenly becomes owned by root.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
This time I'm not having this error message. Now I see fine log:
Wed Sep 24 22:40:04 2014 : Info: rlm_sql (sql): Connected new DB handle, #32
Wed Sep 24 22:40:04 2014 : Info: rlm_sql (sql): Attempting to connect rlm_sql_mysql #33
Wed Sep 24 22:40:04 2014 : Info: rlm_sql_mysql: Starting connect to MySQL server for #33
Wed Sep 24 22:40:04 2014 : Info: rlm_sql (sql): Connected new DB handle, #33
Wed Sep 24 22:40:04 2014 : Info: Loaded virtual server <default>
Wed Sep 24 22:40:04 2014 : Info: Loaded virtual server dynamic_client_server
Wed Sep 24 22:40:04 2014 : Info: Loaded virtual server inner-tunnel
Wed Sep 24 22:40:04 2014 : Info: Ready to process requests.
So it must be some kind of inconsistency in the cron script that is stoping and then starting radiusd service.
I'm now running radiusd for 3 hours clean with no sudden stops. I'll get you posted.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I met the same problem and I think I have found the root cause:
radiusd is being started too quickly before it has exit by '/usr/share/nginx/html/cake2/rd_cake/Console/Command/FreeradiusShell.php'.
I solved the problem by adding a 'sleep 2' in '/usr/share/nginx/html/cake2/rd_cake/Setup/Scripts/radmin_wrapper.pl':
if($arg1 eq 'stop'){
system("/etc/init.d/radiusd stop;sleep 2");
}
Last edit: Shane 2014-12-04
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
After some time after starting I get this message:
"general protection ip:7f9cb07a88e0 sp:7fffc69beff0 error:0 in radiusd[7f9cb078a000+46000]"
and my radiusd service stops working. I have no major load on the server, but it stops and I don't know why is this message popping.
I'm using Centos 6.5 with latest updates and used Mfawa Installer script. I did some radiusd and mysql tweaks: Basicly increasing the number of "max_servers =" and "num_sql_socks ="
Every 10 minutes when "/var/www/html/cake2/rd_cake/Console/cake -app /var/www/html/cake2/rd_cake Freeradius >> /dev/null 2>&1" is executed by cron I get this error in radius.log:
Wed Sep 24 19:20:05 2014 : Info: Loaded virtual server dynamic_client_server
Wed Sep 24 19:20:05 2014 : Info: Loaded virtual server inner-tunnel
Wed Sep 24 19:20:05 2014 : Info: Ready to process requests.
Wed Sep 24 19:20:05 2014 : Info: Loaded virtual server <default>
Wed Sep 24 19:20:05 2014 : Info: Loaded virtual server dynamic_client_server
Wed Sep 24 19:20:05 2014 : Info: Loaded virtual server inner-tunnel
Wed Sep 24 19:20:05 2014 : Error: Failed binding to authentication address * port 1812: Address already in use
Wed Sep 24 19:20:05 2014 : Error: /etc/raddb/radiusd.conf[273]: Error binding to port for 0.0.0.0 port 1812
It's like this line:
exec("sudo /var/www/html/cake2/rd_cake/Setup/Scripts/radmin_wrapper.pl start freeradius");
in the file:
/var/www/html/cake2/rd_cake/Console/Command/FreeradiusShell.php
is executed twice, because radius server is stopped, then starred -- everything goes well and then It's stared again but the listening port is already occupied.
P.S. When I restart radius from command line with the command: service radiusd restart, I don't see the error message about the port binding!
Last edit: nikonaum 2014-09-24
There was someone that had similar problems on Ubuntu and was not sure how
he/she got around that.
I've went through all the forum and could not find person(post) with my problem. Just to get me right.
1. I get the GENERAL PROTECTION error message every 10 minutes -- the period of time in which "/var/www/html/cake2/rd_cake/Console/cake -app /var/www/html/cake2/rd_cake Freeradius >> /dev/null 2>&1" is executed by cron.d/rd.
2. The radiusd service does not stop after the error message but in some future (random (at least I could not see any pattern)) time.
3. I cannot see any load problems with my server -- enough processing and memory resources available, load is 0.5.
4. I did some radiusd and mysqld optimizations as advised here: http://www.slideshare.net/dashohoxha/pppoe-with-mikrotik-and-radius-16009022
5. Something constantly makes the file /var/www/html/cake2/rd_cake/tmp/cache/models/myapp_cake_model_default_rd_checks and
/var/www/html/cake2/rd_cake/tmp/cache/persistent/myapp_cake_core_method_cache
I'm not sure for the correct names of the files but one file in each of these directory suddenly becomes owned by root.
This is the issue I was referring to:
https://github.com/muffycompo/radiusdesk-installer/issues/1
I disabled AppAmmor in Ubuntu to ensure that the files don't get changed as
you mentioned.
I'm on CentOS 6.5 with NO AppArmmor installed.
where do I find the default mysql password and whatnot?.... do you know why
I wouldn't get a login prompt?
On Wed, Sep 24, 2014 at 10:35 AM, Mfawa Alfred Onen <muffycompo@users.sf.net
This time I'm not having this error message. Now I see fine log:
Wed Sep 24 22:40:04 2014 : Info: rlm_sql (sql): Connected new DB handle, #32
Wed Sep 24 22:40:04 2014 : Info: rlm_sql (sql): Attempting to connect rlm_sql_mysql #33
Wed Sep 24 22:40:04 2014 : Info: rlm_sql_mysql: Starting connect to MySQL server for #33
Wed Sep 24 22:40:04 2014 : Info: rlm_sql (sql): Connected new DB handle, #33
Wed Sep 24 22:40:04 2014 : Info: Loaded virtual server <default>
Wed Sep 24 22:40:04 2014 : Info: Loaded virtual server dynamic_client_server
Wed Sep 24 22:40:04 2014 : Info: Loaded virtual server inner-tunnel
Wed Sep 24 22:40:04 2014 : Info: Ready to process requests.
So it must be some kind of inconsistency in the cron script that is stoping and then starting radiusd service.
I'm now running radiusd for 3 hours clean with no sudden stops. I'll get you posted.
I met the same problem and I think I have found the root cause:
radiusd is being started too quickly before it has exit by '/usr/share/nginx/html/cake2/rd_cake/Console/Command/FreeradiusShell.php'.
I solved the problem by adding a 'sleep 2' in '/usr/share/nginx/html/cake2/rd_cake/Setup/Scripts/radmin_wrapper.pl':
if($arg1 eq 'stop'){
system("/etc/init.d/radiusd stop;sleep 2");
}
Last edit: Shane 2014-12-04