OK, and the results from 0.92? Good too? Testing the last working version doesn't help much. :-)
Best, Tom Hetmer CDN77 Operations [email protected] / +44 (0) 20 3514 2399 / www.cdn77.com ----- Původní zpráva ----- > Odesilatel: "Robert Woodcock" <[email protected]> > Příjemce: "Tom Hetmer" <[email protected]>, [email protected] > Datum: 11/23/17 17:03 > Předmět: Re: Bug#882331: mtr: source routing with --address does not work > > You've got something else going on. > > I did a brand new install of Debian stable, created in a KVM VM using > the following command: > virt-install -n test --os-type=Linux --os-variant=debiansqueeze > --ram=512 --vcpus=1 --disk path=test.img,bus=virtio,size=10 --graphics > none --location > http://ftp.us.debian.org/debian/dists/testing/main/installer-amd64/ > --network bridge:br0 --network bridge:br1 --console > pty,target_type=serial --extra-args 'console=ttyS0,115200n8 serial' > > I installed tcpdump and mtr (the 0.87 version from stable), and > configured the two network interfaces with static IPs, with a default > gateway on the first interface. > > root@test:~# mtr --version > mtr 0.87 > root@test:~# ip -4 addr > 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN > group default qlen 1 > inet 127.0.0.1/8 scope host lo > valid_lft forever preferred_lft forever > 2: ens2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast > state UP group default qlen 1000 > inet 192.168.9.29/24 brd 192.168.9.255 scope global ens2 > valid_lft forever preferred_lft forever > 3: ens3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast > state UP group default qlen 1000 > inet 192.168.10.29/24 brd 192.168.10.255 scope global ens3 > valid_lft forever preferred_lft forever > root@test:~# ip route > default via 192.168.9.1 dev ens2 onlink > 192.168.9.0/24 dev ens2 proto kernel scope link src 192.168.9.29 > 192.168.10.0/24 dev ens3 proto kernel scope link src 192.168.10.29 > > I ran mtr in one window, using -a to specify the IP for ens3. In other > windows, I ran tcpdump on ens2 and ens3: > > root@test:~# mtr -c 1 --report -a 192.168.10.29 1.2.3.4 > Start: Thu Nov 23 07:51:36 2017 > HOST: test Loss% Snt Last Avg Best Wrst StDev > > root@test:~# tcpdump -ni ens2 icmp > tcpdump: verbose output suppressed, use -v or -vv for full protocol decode > listening on ens2, link-type EN10MB (Ethernet), capture size 262144 bytes > 07:51:37.025564 IP 192.168.10.29 > 1.2.3.4: ICMP echo request, id 22275, > seq 59520, length 44 > 07:51:37.126989 IP 192.168.10.29 > 1.2.3.4: ICMP echo request, id 22275, > seq 59776, length 44 > 07:51:37.228430 IP 192.168.10.29 > 1.2.3.4: ICMP echo request, id 22275, > seq 60032, length 44 > 07:51:37.328767 IP 192.168.10.29 > 1.2.3.4: ICMP echo request, id 22275, > seq 60288, length 44 > 07:51:37.429117 IP 192.168.10.29 > 1.2.3.4: ICMP echo request, id 22275, > seq 60544, length 44 > 07:51:37.529475 IP 192.168.10.29 > 1.2.3.4: ICMP echo request, id 22275, > seq 60800, length 44 > 07:51:37.629855 IP 192.168.10.29 > 1.2.3.4: ICMP echo request, id 22275, > seq 61056, length 44 > > root@test:~# tcpdump -ni ens3 icmp > tcpdump: verbose output suppressed, use -v or -vv for full protocol decode > listening on ens3, link-type EN10MB (Ethernet), capture size 262144 bytes > > The packets egress via ens2 using the 192.168.10.29 IP. > > Bottom line: If you want to do this, you need to set up a separate > routing table in Linux and you need to configure rules to use that > routing table. This is not a regression in mtr.
