[B.A.T.M.A.N.] Node freezes but still alive

[B.A.T.M.A.N.] Node freezes but still alive

[Derick Roman De Leon Leal]

First of all, thanks in advance for this fantastic mesh protocol. I am
kind of new at this so any advice is welcome.

I have a little setup with two nodes and one gateway, from time to
time one of the nodes just stops transmitting data.

I see the node on the originators list but the TQ never changes unlike
when its working OK that the TQ changes a bit during the day. The
thing is that the node is there, since the last-seen time is updating
every few seconds just like when is working normal. I have tried
changing my batman configuration with no positive results. Also, I
have changed the nodes and this is still happening.

This happens and resolves itself in a couple of hours, but it does not
happen in other setups I have with more nodes.

I tried to search the internet for this issue with zero luck. The
three devices are OM2P-HS v3 running Openwrt r49395 with batman-adv
2016.2 using BATMAN IV.

Have you guys have encountered this problem before?
Is my configuration wrong in any way?
Do you have any recommendations for me?

Below is the configuration I am using, hope this helps. Please let me
know if there is any more information I can make available to you to
help me solve this problem.

Thanks a lot for taking the time to read all this. I appreciate any
comments and feedback you can give.

Kind Regards,
Derick



Nodes Configuration: -----------------------------------------------------

batman-adv +++++++++++++++++++++

config 'mesh' 'bat0'
option 'interfaces' 'mesh'
option 'aggregated_ogms'
option 'ap_isolation' '1'
option 'bonding'
option 'fragmentation' '1'
option 'gw_bandwidth'
option 'gw_mode' 'client'
option 'gw_sel_class'
option 'log_level'
option 'orig_interval'
option 'vis_mode'
option 'bridge_loop_avoidance' '1'
option 'distributed_arp_table'
option 'multicast_mode'
option 'network_coding'
option 'hop_penalty'
option 'isolation_mark'

wireless +++++++++++++++++++++++++++

config wifi-device  radio0
option type     mac80211
option channel  6
option hwmode 11g
option path 'platform/ar934x_wmac'
option htmode HT20
option country MX

config wifi-iface wlan0
option ifname   wlan0
option device   radio0
option network  lan
option mode     sta
option encryption none
option disabled 0

config wifi-iface mesh0
option device radio0
option ifname   adhoc0
option network  mesh
option mode     adhoc
option ssid MY-SSID111
option bssid    CA:FE:CA:FE:CA:FE

network +++++++++++++++++++++++++

config interface 'loopback'
option ifname 'lo'
option proto 'static'
option ipaddr '127.0.0.1'
option netmask '255.0.0.0'

config globals 'globals'
option ula_prefix 'fdaa:8902:bcc0::/48'

config interface 'lan'
    option ifname 'eth0 bat0'
    option force_link '1'
    option type 'bridge'
    option proto 'dhcp'
    list dns 8.8.8.8
    list dns 8.8.4.4
    option peerdns 0

config interface 'wan'
    option ifname 'eth1'
    option proto 'dhcp'

config interface 'wan6'
    option ifname 'eth1'
    option proto 'dhcpv6'

config interface 'mesh'
    option proto 'batadv'
    option mtu '1528'
    option mesh 'bat0'


Gateway configuration
-------------------------------------------------------------
everything is the same as in the nodes configuration but batman-adv file.

config 'mesh' 'bat0'
option 'interfaces' 'mesh'
option 'aggregated_ogms'
option 'ap_isolation' '1'
option 'bonding'
option 'fragmentation' '1'
option 'gw_bandwidth'
option 'gw_mode' 'server'
option 'gw_sel_class'
option 'log_level'
option 'orig_interval'
option 'vis_mode'
option 'bridge_loop_avoidance' '1'
option 'distributed_arp_table'
option 'multicast_mode'
option 'network_coding'
option 'hop_penalty'
option 'isolation_mark'

Re: [B.A.T.M.A.N.] Node freezes but still alive

[Linus Lüssing]

Hi Derick,

On Thu, Feb 23, 2017 at 05:37:55PM -0600, Derick Roman De Leon Leal wrote:
> I have a little setup with two nodes and one gateway, from time to
> time one of the nodes just stops transmitting data.

Very odd indeed. Can you check whether transmitting IPv4 or
IPv6 packets on adhoc0 directly still works when this happens?

Cheers, Linus

Re: [B.A.T.M.A.N.] Node freezes but still alive

[Derick Roman De Leon Leal]

Thanks for the quick reply Linus,

No transmission on IPv4, haven't tried on IPv6 but I assume is going
to be the same. Luan Jinlu suggested that I should use a bssid that
starts with 02. Is this a requirement for the mesh to work?

Should I share a batman-adv log? which packet type should be more useful?

Thanks a lot for your kind help!!

Best Regards,
Derick





> ---------- Forwarded message ----------
> From: "Linus Lüssing" <linus.luessing@c0d3.blue>
> To: The list for a Better Approach To Mobile Ad-hoc Networking <b.a.t.m.a.n@lists.open-mesh.org>
> Cc:
> Bcc:
> Date: Fri, 24 Feb 2017 09:28:11 +0100
> Subject: Re: [B.A.T.M.A.N.] Node freezes but still alive
> Hi Derick,
>
> On Thu, Feb 23, 2017 at 05:37:55PM -0600, Derick Roman De Leon Leal wrote:
> > I have a little setup with two nodes and one gateway, from time to
> > time one of the nodes just stops transmitting data.
>
> Very odd indeed. Can you check whether transmitting IPv4 or
> IPv6 packets on adhoc0 directly still works when this happens?
>
> Cheers, Linus
>
>
> _______________________________________________
> B.A.T.M.A.N mailing list
> B.A.T.M.A.N@lists.open-mesh.org
> https://lists.open-mesh.org/mm/listinfo/b.a.t.m.a.n
>

Re: [B.A.T.M.A.N.] Node freezes but still alive

[Derick Roman De Leon Leal]

Thanks for the quick reply Linus,

No transmission on IPv4, haven't tried on IPv6 but I assume is going
to be the same. Someone suggested that I should use a bssid that
starts with 02 (eg. 02:12:34:56:78:9A). Is this a requirement for the
mesh to work?

Should I share a batman-adv log? which packet type should be more useful?

Thanks a lot!!

Kind Regards,
Derick





> ---------- Forwarded message ----------
> From: "Linus Lüssing" <linus.luessing@c0d3.blue>
> To: The list for a Better Approach To Mobile Ad-hoc Networking <b.a.t.m.a.n@lists.open-mesh.org>
> Cc:
> Bcc:
> Date: Fri, 24 Feb 2017 09:28:11 +0100
> Subject: Re: [B.A.T.M.A.N.] Node freezes but still alive
> Hi Derick,
>
> On Thu, Feb 23, 2017 at 05:37:55PM -0600, Derick Roman De Leon Leal wrote:
> > I have a little setup with two nodes and one gateway, from time to
> > time one of the nodes just stops transmitting data.
>
> Very odd indeed. Can you check whether transmitting IPv4 or
> IPv6 packets on adhoc0 directly still works when this happens?
>
> Cheers, Linus
>
>
> _______________________________________________
> B.A.T.M.A.N mailing list
> B.A.T.M.A.N@lists.open-mesh.org
> https://lists.open-mesh.org/mm/listinfo/b.a.t.m.a.n
>

Re: [B.A.T.M.A.N.] Node freezes but still alive

[Linus Lüssing]

On Fri, Feb 24, 2017 at 10:48:35AM -0600, Derick Roman De Leon Leal wrote:
> Thanks for the quick reply Linus,
> 
> No transmission on IPv4

If that's for the adhoc0 interface directly then this seems to be
a wireless driver issue and not a batman-adv one :(.

Hm, if I'm not mistaken then the OM2P-HS should have an Atheros AR9285
chipset. So an ath9k driver which many people use in adhoc mode
for meshing. 

> Someone suggested that I should use a bssid that
> starts with 02 (eg. 02:12:34:56:78:9A). Is this a requirement for the
> mesh to work?

Not a "requirement for the mesh to work", afaik. But when a MAC
address is self-generated, then it should have the second lowest bit set:

https://en.wikipedia.org/wiki/MAC_address#Address_details


Not sure, maybe you just hit a bad OpenWRT revision? Maybe give
the new LEDE release [*] or a newer OpenWRT version a try?

Regards, Linus

[*]: https://lede-project.org/releases/17.01/start

Re: [B.A.T.M.A.N.] Node freezes but still alive

[Antonio Quartulli]

[-- Attachment #1: Type: text/plain, Size: 1031 bytes --]

On Fri, Feb 24, 2017 at 06:56:58PM +0100, Linus Lüssing wrote:
> On Fri, Feb 24, 2017 at 10:48:35AM -0600, Derick Roman De Leon Leal wrote:
> > Thanks for the quick reply Linus,
> > 
> > No transmission on IPv4
> 
> If that's for the adhoc0 interface directly then this seems to be
> a wireless driver issue and not a batman-adv one :(.
> 
> Hm, if I'm not mistaken then the OM2P-HS should have an Atheros AR9285
> chipset. So an ath9k driver which many people use in adhoc mode
> for meshing. 
> 
> > Someone suggested that I should use a bssid that
> > starts with 02 (eg. 02:12:34:56:78:9A). Is this a requirement for the
> > mesh to work?
> 
> Not a "requirement for the mesh to work", afaik. But when a MAC
> address is self-generated, then it should have the second lowest bit set:
> 
> https://en.wikipedia.org/wiki/MAC_address#Address_details

This is true for the interface MAC address, but this should not apply for
the Cell-ID/BSSID. It can be a random ID.

Cheers,


-- 
Antonio Quartulli

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 801 bytes --]

Re: [B.A.T.M.A.N.] Node freezes but still alive

[Derick Roman De Leon Leal]

Thanks Linus and Antonio for your replies.

I guess is a driver issue, the thing is that I have been using this
exact same configuration in other 10 locations without a problem. It
sure is strange.

I don't know if this helps, but I saw this in the log.

Sat Feb 25 17:23:52 2017 kern.info kernel: [21168.652823] batman_adv:
bat0: IGMP Querier disappeared - multicast optimizations disabled
Sat Feb 25 17:24:02 2017 kern.info kernel: [21178.691583] batman_adv:
bat0: IGMP Querier appeared

A quick google search showed that this has to do with multicast, which
is disabled in my configuration. Should I enable it?

Also, I have noticed that the node with the issue is constantly
requesting who-has the gateway.

I ran this command from the gateway.
batctl td -x257 adhoc0

18:23:44.134458 BAT ae:86:74:47:8b:02: BCAST, orig ae:86:74:47:8b:02,
seq 1933, ARP, Request who-has 10.10.0.1 tell 10.10.0.142
(ac:86:74:47:8b:00), length 28
18:23:44.145748 BAT ae:86:74:47:8b:02: BCAST, orig ae:86:74:47:8b:02,
seq 1933, ARP, Request who-has 10.10.0.1 tell 10.10.0.142
(ac:86:74:47:8b:00), length 28
18:23:45.124692 BAT ae:86:74:47:8b:02: BCAST, orig ae:86:74:47:8b:02,
seq 1934, ARP, Request who-has 10.10.0.1 tell 10.10.0.142
(ac:86:74:47:8b:00), length 28
18:23:45.125725 BAT 60:e3:27:f2:71:1c > ae:86:74:47:8b:02: UCAST, ttvn
2, ttl 50, ARP, Reply 10.10.0.1 is-at 6a:64:42:e7:bf:cb, length 28
18:23:45.139625 BAT ae:86:74:47:8b:02: BCAST, orig ae:86:74:47:8b:02,
seq 1934, ARP, Request who-has 10.10.0.1 tell 10.10.0.142
(ac:86:74:47:8b:00), length 28
18:23:45.144519 BAT ae:86:74:47:8b:02: BCAST, orig ae:86:74:47:8b:02,
seq 1934, ARP, Request who-has 10.10.0.1 tell 10.10.0.142
(ac:86:74:47:8b:00), length 28

Does this means that there is IPv4 transmission?

Is there a way to make sure this is a driver problem?

In the meanwhile I will try a LEDE and a newer Openwrt.

Thanks again for all your help, any comment is welcome.


Best Regards,
Derick