Re: Radius dies frequently
Sergey Poznyakoff <[email protected]>
| Newsgroups | gmane.comp.gnu.radius.general |
|---|---|
| Organization | Farlep-Internet |
| Message-ID | <[email protected]> |
Dan Beldiman <[email protected]> wrote: > I am using GNU Radius version 1.2 on Linux with MySQL, and it has been > working fine for a long time. Try upgrading to 1.3. > The radius server was not completely unresponsive, it did reply all request > with DENY messages (which prevented a failover) > > Jan 18 06:49:28 Main.info: Reloading configuration now > Jan 18 06:49:28 Main.info: Terminating the subprocesses > Jan 18 06:49:28 Main.notice: child 23102 exited with status 0 > Jan 18 06:49:28 Main.notice: child 23098 exited with status 0 > Jan 18 06:49:28 Main.notice: child 23094 exited with status 0 > (this line for about 12 child processes) This means somebody used `radctl reload' to force the server to reload its configuration files. Along with the fact that the server begun sending denies to all queries afterward, it seems to indicate that the new configuration was erroneous and prevented users from authenticating. > Jan 18 06:29:47 Auth.notice: Killing unresponsive ACCT child 20528 This means that accounting process did not respond within `time-to-live' seconds. This in turn usually indicates that some part of the accounting procedure was too slow (e.g. updating the SQL database, running external program with Acct-Ext-Program, etc.). Also, version 1.2 had some problems that under some circumstances could led to the similar behavior. These were fixed in version 1.3. Regards, Sergey