[Xorp-users] OpenBSD IGMP/PIM Issue

Pavlin Radoslavov pavlin at ICSI.Berkeley.EDU
Wed Oct 8 09:48:42 PDT 2008


Stegen Smith <stegen at owns.com> wrote:

> Hello All,
> 
> I've got encountered an issue while trying to run xorp on an OpenBSD  4.3 i386
> GENERIC system using IGMP and PIM.
> I've got net.inet.ip.forwarding and net.inet.ip.mforwarding enabled,  and I'll
> paste my config below.

Stegen,

Do you have the following in your /etc/rc.conf.local file (mentioned
in the XORP User Manual, the PIM-SM section):

# Enable multicast routing (see netstart(8) for details).
multicast_host=NO                       
multicast_router=YES

The sendmsg() "No route to host" error seems an indication that the
outgoing multicast control packets are dropped by the IPv4
forwarding table check. Typically this happens if the forwarding
table is missing a default route or a 224.0.0.0/4 route (you need
either one).
The above /etc/rc.conf.local configuration fixes that.

The xorp_fib2mrib crash seems an orthogonal issue that must be
investigated and fixed, hence please send us the coredump trace.

Pavlin

> interfaces {
>     restore-original-config-on-shutdown: false
>     interface vlan40 {
>         description: "data interface"
>         disable: false
>         default-system-config
>     }
> }
> 
> fea {
>     unicast-forwarding4 {
>         disable: false
>     }
>     click {
>         disable: true
>     }
> }
> 
> plumbing {
>     mfea4 {
>         disable: false
>         interface vlan40 {
>             vif vlan40 {
>                 disable: false
>             }
>         }
>         interface register_vif {
>             vif register_vif {
>                 disable: false
>             }
>         }
>         traceoptions {
>             flag all {
>                 disable: false
>             }
>         }
>     }
> }
> 
> protocols {
>     igmp {
>         disable: false
>         interface vlan40 {
>             vif vlan40 {
>                 disable: false
>                 version: 2
>                 enable-ip-router-alert-option-check: false
>                 query-interval: 30
>                 query-last-member-interval: 1
>                 query-response-interval: 10
>                 robust-count: 2
>             }
>         }
>         traceoptions {
>             flag all {
>                 disable: false
>             }
>         }
>     }
>     pimsm4 {
>         disable: false
>         interface vlan40 {
>             vif vlan40 {
>                 disable: false
>                 dr-priority: 1
>                 hello-period: 30
>                 hello-triggered-delay: 5
>             }
>         }
>         interface register_vif {
>             vif register_vif {
>                 disable: false
>             }
>         }
>         bootstrap {
>             disable: false
>             cand-bsr {
>                 scope-zone 224.0.0.0/4 {
>                     is-scope-zone: false
>                     cand-bsr-by-vif-name: "vlan40"
>                     cand-bsr-by-vif-addr: 10.3.48.2
>                     bsr-priority: 1
>                     hash-mask-len: 30
>                 }
>             }
>         }
>         switch-to-spt-threshold {
>             /* approx. 1K bytes/s (10Kbps) threshold */
>             disable: false
>             interval: 100
>             bytes: 102400
>         }
> 
>         traceoptions {
>             flag all {
>                 disable: false
>             }
>         }
> 
>     }
> }
> 
> Note that I've only got it enabled on vlan40 because I'm not seeing  any igmp
> queries coming from the system at all, and I just wanted a  trimmed down
> configuration until I can see the queries.  Now here's an  interesting log
> entry that I've seen.
> 
> [ 2008/10/07 23:38:13 WARNING xorp_fea XrlFeaTarget ] Handling method  for
> raw_packet4/0.1/send failed: XrlCmdError 102 Command failed  sendmsg(proto 2
> size 32 from 10.3.48.2 to 224.0.0.1 on interface  vlan40 vif vlan40) failed:
> No route to host
> 
> I saw something similar in an older post, but there wasn't any follow  up to
> it:
> http://mailman.icsi.berkeley.edu/pipermail/xorp-users/2004-December/000332.html
> 
> I'm sure that other folks have gotten this to work on other openbsd  systems,
> but I'm not sure what I'm missing.  I'm using an intel em0  interface for
> vlan40, and MULTICAST is enabled on the interface.  I  must just be missing
> something silly!
> 
> I'm also not sure if I should have fib2mrib enabled or not, but when I  do
> enable it, the xorp_fib2mrib module dies.
> [ 2008/10/07 23:43:57  ERROR xorp_rtrmgr:14018 RTRMGR +747  module_manager.cc
> done_cb ] Command "/usr/local/xorp/fib2mrib/ xorp_fib2mrib": terminated with
> signal 6; aborted with a core dump.
> [ 2008/10/07 23:43:57  INFO xorp_rtrmgr:14018 RTRMGR +291  module_manager.cc
> module_exited ] Module coredumped: fib2mrib
> [ 2008/10/07 23:43:57 INFO xorp_rib RIB ] Received death event for  protocol
> fib2mrib shutting down -------
> OriginTable: fib2mrib
> IGP
> next table = Redist:fib2mrib
> 
> ... xorp churns and churns and ultimately ends up here ...
> 
> [ 2008/10/07 23:46:24 TRACE xorp_fea MFEA ] RX kernel signal:  message_type =
> 1 vif_index = 0 src = 10.3.49.57 dst = 239.1.1.1
> [ 2008/10/07 23:46:24 TRACE xorp_fea MFEA ] RX kernel signal:  message_type =
> 1 vif_index = 0 src = 10.3.49.33 dst = 239.1.1.1
> [ 2008/10/07 23:46:24  WARNING xorp_rtrmgr:14018 XrlFinderTarget  +406
> ../xrl/targets/finder_base.cc handle_finder_0_2_resolve_xrl ]  Handling method
> for finder/0.2/resolve_xrl failed: XrlCmdError 102  Command failed Target
> "fib2mrib" does not exist or is not enabled.
> [ 2008/10/07 23:46:24  ERROR xorp_rtrmgr:14018 RTRMGR +1998 task.cc  task_fail
> ] Shutting down fatally wounded process (fib2mrib)
> [ 2008/10/07 23:46:24  ERROR xorp_rtrmgr:14018 RTRMGR +681
> master_conf_tree.cc commit_pass2_done ] Commit failed: Can't validate  config
> ready of process fib2mrib
> [ 2008/10/07 23:46:24  ERROR xorp_rtrmgr:14018 RTRMGR +251
> master_conf_tree.cc config_done ] Configuration failed: Can't validate  config
> ready of process fib2mrib
> xorp_rtrmgr in free(): error: chunk is already free
> [1]  + 14018 abort (core dumped)  ./bin/xorp_rtrmgr
> root134p0at-ig0.sv/usr/local/xorp
> 
> So I've got it disabled for now, but I figured I'd let this issue be  known
> too.
> 
> If anyone could provide me with some insight as to what I'm missing  I'd
> definitely be appreciative as I'd love to get this working. =)   Also please
> let me know if I should provide any more information.
> 
> thanks for the help
> stegen_______________________________________________
> Xorp-users mailing list
> Xorp-users at xorp.org
> http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/xorp-users



More information about the Xorp-users mailing list