Xinetd error message repeating every 15 minutes - HELP
-
It's indeed the TFTP proxy. Unchecking all the interfaces just prevents binding I guess, but does not stop the service. I think that it was cmb who confirmed on some thread that this was "sort of" a bug
-
It is most likely from the filter sync. That message is from xinetd that is activating for one service on port 6969, which would be the TFTP proxy. Deselect all interfaces on System > Advanced, Firewall & NAT tab in the TFTP Proxy list, then save, and it should deactivate that service.
I'm not using TFTP proxy and this is happening quite frequently and disturbing traffic when it happens.
-
I'm not using TFTP proxy and this is happening quite frequently and disturbing traffic when it happens.
It's absolutely not disturbing traffic, it's nothing but log spam.
Whatever is triggering the filter reload is the source of the "disturbing traffic", but it's probably that you're having ISP issues.
-
@cmb:
I'm not using TFTP proxy and this is happening quite frequently and disturbing traffic when it happens.
It's absolutely not disturbing traffic, it's nothing but log spam.
Whatever is triggering the filter reload is the source of the "disturbing traffic", but it's probably that you're having ISP issues.
Across internal networks.
-
If you have a gateway showing as "down", and state killing on gateway failure enabled, then it would reset the states each time the filter reloads.
So fix the down gateway, disable monitoring for it, or unset the state killing on gateway failure option.
-
I have this same issue as well since upgrading from 2.2.6 to 2.3.1 (and now 2.3.2).
-
Gateways are up
-
Happens whether monitoring is disabled or enabled.
-
Happens whether state killing on gateway failure is disabled or enabled
-
-
I'm in the same boat. Not using TFTP, but this reload happens often and resets my IPSEC tunnel as well…
-
Anyone else have a cron entry; 0,15,30,45;;;;;root;/etc/rc.filter_configure.sync ? This appears to be causing it for me. Can this cron entry be removed without causing other issues? How often should the filter be resync'd?
-
That cron entry would be from having scheduled rules (or at least having schedules defined). If you are not using scheduled rules, you can removed it. The filter will sync when config changes are applied or from automated events like gateway failures/recovery.
-
So if you need the filters to resync, how do you stop the log spam from occurring? I've followed all of the advice provided.
-
At the moment, you don't. There isn't a way in xinetd to stop it from being so chatty.
-
What about stopping the underlying service that appears to be causing it to be chatty (i.e. the TFTP service)?
-
Although I also have no interfaces selected I find a service entry in /var/etc/xinetd.conf, see attached. So I rem'd it out.
-
Now I only get log entries "check_reload_status" "syncing firewall" every 15 minutes without readjusting service 6969-udp. I'll reduce the cron frequency as well.
-
Well that didn't last long. Some other file removed the #rem from xinetd.conf. Jimp, can you say how the service gets entered into the file so TFTP proxy can be rem'd out permanently?
-
FYI I just pushed a fix for this for all of the next release branches. If no interfaces are selected for TFTP proxy and reflection is not set to NAT+Proxy, the daemon will not attempt to run.
-
Yeah ;) Thanks JimP
-
FYI I just pushed a fix for this for all of the next release branches. If no interfaces are selected for TFTP proxy and reflection is not set to NAT+Proxy, the daemon will not attempt to run.
-
Will this fix be in the 2.4 release?
-
2.4, 2.3.3, and 2.3.2_2, it's in all the branches. Whichever one we put out next will have it.
