routing tab log spam radvd: can't join ipv6-allrouters on <interface>



  • I keep getting log spam radvd: can't join ipv6-allrouters on cxgbe0.22
    Only solution seems to be stopping radvd and starting it again. Restarting does not work.
    But after a while the messages appears again.
    I don't know why, couldn't find a conclusive answer on google.

    926b914a-6c46-4918-9f62-721e8813c2cf-image.png

    my settings:
    16112b7f-4157-42c0-800b-3708a07202fb-image.png
    01626846-52f7-415a-9d1a-ae525d7e5eba-image.png
    22be4866-42c0-414b-844c-d50eaccef264-image.png
    7b957cd6-5684-4034-a242-7adcd2550fcb-image.png

    edit: note that i am able to visit ipv6 sites with no issues ipv6 breaks when this spam starts



  • I was able to trace back to beginning of the message. I looked at other logs like firewall, dhcp, general, dns but nothing significant happened at the time that this error started.

    Apr 10 10:42:37	radvd	15739	can't join ipv6-allrouters on cxgbe0.7
    Apr 10 10:42:32	radvd	15739	can't join ipv6-allrouters on cxgbe0.42
    Apr 10 10:42:31	radvd	15739	can't join ipv6-allrouters on cxgbe0.7
    Apr 10 10:42:23	radvd	15739	can't join ipv6-allrouters on cxgbe0.42
    Apr 10 01:13:14	radvd	15451	version 2.17 started
    Apr 10 01:06:08	radvd	8151	resuming normal operation
    Apr 10 01:06:08	radvd	8151	attempting to reread config file
    Apr 10 01:06:06	radvd	8151	resuming normal operation
    Apr 10 01:06:06	radvd	8151	attempting to reread config file
    Apr 10 01:04:53	radvd	8151	resuming normal operation
    Apr 10 01:04:53	radvd	8151	attempting to reread config file
    Apr 10 01:04:50	radvd	8151	resuming normal operation
    Apr 10 01:04:50	radvd	8151	attempting to reread config file
    Apr 10 01:02:51	radvd	8151	resuming normal operation
    Apr 10 01:02:51	radvd	8151	attempting to reread config file
    Apr 10 01:02:49	radvd	8151	resuming normal operation
    Apr 10 01:02:49	radvd	8151	attempting to reread config file
    Apr 10 01:00:39	radvd	8088	version 2.17 started
    Apr 10 00:55:32	radvd	22375	resuming normal operation
    Apr 10 00:55:32	radvd	22375	attempting to reread config file
    Apr 10 00:55:30	radvd	22375	resuming normal operation
    Apr 10 00:55:30	radvd	22375	attempting to reread config file
    Apr 10 00:54:19	radvd	22375	resuming normal operation
    Apr 10 00:54:19	radvd	22375	attempting to reread config file
    Apr 10 00:52:25	radvd	22206	version 2.17 started
    Apr 9 17:26:42	radvd	38270	resuming normal operation
    Apr 9 17:26:42	radvd	38270	attempting to reread config file
    Apr 9 17:26:40	radvd	38270	resuming normal operation
    Apr 9 17:26:40	radvd	38270	attempting to reread config file
    Apr 9 17:02:23	radvd	38270	resuming normal operation
    Apr 9 17:02:23	radvd	38270	attempting to reread config file
    Apr 9 17:02:21	radvd	38270	resuming normal operation
    Apr 9 17:02:21	radvd	38270	attempting to reread config file
    Apr 9 17:00:57	radvd	38142	version 2.17 started
    Apr 9 16:52:06	radvd	59526	resuming normal operation
    Apr 9 16:52:06	radvd	59526	attempting to reread config file
    Apr 9 16:52:02	radvd	59526	resuming normal operation
    Apr 9 16:52:02	radvd	59526	attempting to reread config file
    Apr 9 16:51:59	radvd	59526	resuming normal operation
    Apr 9 16:51:59	radvd	59526	attempting to reread config file
    Apr 9 16:51:56	radvd	59526	resuming normal operation
    Apr 9 16:51:56	radvd	59526	attempting to reread config file
    Apr 9 16:50:54	radvd	59267	invalid all-zeros prefix in /var/etc/radvd.conf, line 51
    Apr 9 16:50:54	radvd	59267	invalid all-zeros prefix in /var/etc/radvd.conf, line 37
    Apr 9 16:50:54	radvd	59267	invalid all-zeros prefix in /var/etc/radvd.conf, line 23
    Apr 9 16:50:54	radvd	59267	invalid all-zeros prefix in /var/etc/radvd.conf, line 9
    Apr 9 16:50:54	radvd	59267	version 2.17 started
    Apr 9 15:25:00	radvd	38338	resuming normal operation
    Apr 9 15:25:00	radvd	38338	attempting to reread config file
    Apr 9 15:06:11	radvd	38338	resuming normal operation
    Apr 9 15:06:11	radvd	38338	attempting to reread config file
    Apr 9 12:00:40	radvd	38338	resuming normal operation
    Apr 9 12:00:40	radvd	38338	attempting to reread config file
    Apr 9 12:00:38	radvd	38338	resuming normal operation
    Apr 9 12:00:38	radvd	38338	attempting to reread config file
    Apr 9 11:58:14	radvd	38120	version 2.17 started
    


  • bump



  • edit: I have disabled ipv6 for now



  • Just to note, when this spam starts, ipv6 stops working.



  • I can confirm this must be an issue.

    I have exactly the same situation, after around 6-7 hours of running, radvd spams the logs with "can't join ipv6-allrouters". IPv6 stops works as all clients release their IPv6 address since no more router advertisements are ever sent.
    As an interim fix, I stop and start radvd via cron every 6 hours, a click of restart does not fix it. It has to be a stop, sleep 1 second, and a start. Is there some memory leak in radvd? Or binding and not releasing to the interface? There is radvd 2.18 available now with numerous fixes.

    This issue does unfortunately break IPv6 when using un-managed router advertisements.

    I have turned up logging levels, but it does not add anything more related to radvd.



  • @Irata thank you, good to know that I am not the only one.
    Can someone from pfsense team let us know if a bug report needs to be opened?



  • @strangegopher thanks, glad I found your post as I also thought it was just me!

    It happens on all interfaces at the same time, so I took a packet capture on a quiet subnet - but nothing of interest again, can just see when the router advertisements stop.

    Sadly my restart workaround using cron has side effects, any active IPv6 connections are dropped each time it runs. And one Android device even disconnects from WiFi each time radvd is restarted, it treats a loss of IPv6 as a network problem, bizarre but true.

    Be interested to hear if anyone using radvd and IPv6 does not have this issue? Check Routing logs in System Logs.

    I'm at a loss what else to do. It's interesting that a periodic restart masks the problem, does that point to radvd being the problem?



  • I dont have this issue. Im on 2.5 with uptime 3 days.



  • @maverick_slo do u have ipv6?



  • Since 2013 😁



  • I've been having this problem as well and I also noticed that it's about 6 hours before IPv6 suddenly breaks. This should definitely be fixed before 2.5.0 is released.



  • I am having the same issue. IPv6 would stop working within 24 hours. I backup the config.xml from the 2.5 and restored the file to a 2.4.4_p2 and it does not have the problem.



  • I'm not sure how to find what's common.

    I'm running pfsense within VMware ESXi and have this issue. Are others with this issue the same?

    I can also confirm the latest 2.4 branch does not have this issue with the same config.



  • I am running it on baremetal dell box but I am using chelsio 10G sfp+ nic.


Log in to reply