Red Hat Bugzilla – Bug 214862
Time to live (TTL) on some multicast packets incorrect using ipv6 module
Last modified: 2007-12-10 11:07:06 EST
Description of problem:
We have a local daemon process that publishes multicast heartbeats on
18.104.22.168. The ttl is hard coded into the application(JAVA) and set at
12. If the IPV6 module is enabled the multicast packets TTL field as
discovered with ethereal is set to 1. If we turn off the IPV6 module using
alias net-pf-10 off in the modprobe.conf file the ttl is then set to 15 which
still is not correct but it allows the packet to get past the router.
Version-Release number of selected component (if applicable):
RHEL 4 Update 3, Fedora Core 3 - 5
Steps to Reproduce:
1. Start the daemon process
2. Use Ethereal to sniff outbound traffic on eth0 of the box the daemon is
3. Watch the outbound packet leave eth0 and expand the IP header
4. See TTL set to 1
5. Make change listed above to the modprobe.conf - reboot
6. repeat steps 1-3
7. See TTL set to 15
8. Go to distant end (other subnet) see mcast packets arrive as expected
Ethereal discovers TTL set to 1
Ethereal discovers TTL set to 12 (per the hard coding in the application)
Could find nothing in bugzilla. Not sure what module this belongs in.
Can you test on the lastest kernel update (4.5) and indicate if the problem
still occurs? Thanks!
ping. any update?
again, ping. Any update?
closing. no response