OK, the configuration is fine for such context. Please can you provide verbose output from monit (using -v option) for the critical time frame where the cs_host becomes unmonitored?
Regards, Martin On May 12, 2010, at 11:34 AM, Nick Upson wrote: > sorry I don't understand, you say "replace the exec action with > restart" but the exec action already includes the restart of pppd > (which is also run from the same monit config file). It seems that > once the check host has failed once it becomes unmonitored but I need > it to be checked every period unless pppd is down (when there is no > point) > > On 11 May 2010 23:14, Martin Pala <[email protected]> wrote: >> Hi, >> i think you can replace the exec action with restart and t should work - >> even though the cs_host has no start/stop program, the restart action will >> be propagated to pppd via dependency. If it won't help, please can you >> provide output of 'monit -vI'? >> Regards, >> Martin >> >> On May 7, 2010, at 10:40 AM, Nick Upson wrote: >> >> I have the following in my monit configuration, which works fine >> except that once the ping has failed and restarted pppd, the entry >> goes to unmonitored but I need it to try again and continue doing so. >> >> currently monit-4.10.1 on fedora 8 >> >> >> >> check host cs_host with address 10.4.13.254 >> every 5 cycles >> if failed icmp type echo count 10 with timeout 10 seconds >> then exec "/bin/bash -c '/opt/unb/bin/monit.sh restart pppd'" >> depends on pppd >> >> >> -- >> To unsubscribe: >> http://lists.nongnu.org/mailman/listinfo/monit-general >> > > -- > To unsubscribe: > http://lists.nongnu.org/mailman/listinfo/monit-general -- To unsubscribe: http://lists.nongnu.org/mailman/listinfo/monit-general
