Hi @Eugene, could you kindly check this ? Regards.
Hello Eugene, it happened again : route_monitor.log 03:01:03.057 PID 0 add/repl iface iface#225 ng192 admin DOWN oper DOWN mtu 1500 03:01:03.058 PID 0 add/repl iface iface#225 ng192 admin DOWN oper DOWN mtu 1492 03:01:03.078 PID 0 add/repl iface iface#225 ng192 admin UP oper UP mtu 1492 03:01:03.078 PID 0 add/repl addr 10.129.39.254/32 -> 10.129.36.24 iface ng192 03:01:03.078 PID 0 add/repl addr 10.129.39.254/32 -> 10.129.36.24 iface ng192 03:01:03.078 PID 0 add/repl iface iface#225 ng192 admin UP...
Hi Eugene, the issue just happened, here are the grabbed log files. Please let us know your thoughts. Regards.
Hi Eugene, thank you for replying and your support. The script is running, we are waitinf for the next route event. Regards.
Hi Grzesiek, It looks we have the same bug, I've updated to FreeBSD 14.3-RELEASE and the bug is still present for me. please check my post : https://sourceforge.net/p/mpd/discussion/44693/thread/2b3e5ea536/?limit=100
Hi Eugene, We are on FreeBSD 14.3-RELEASE and the bug is still present. Which options do we have without moving to STABLE versions ? Thanks in advance for your advice.
Hi @dadv, any update for us please ? Cheers?
This bug looks similar to the previous thread "IFACE: Adding IPv4 address to ng0 failed: File exists + iBGP or OSPF" but I am on freeBSD 14.2, it should be fixed.... I also posted in FreeBSF Forum : https://forums.freebsd.org/threads/cannot-delete-ip-route.97263/#post-694914
This bug looks similar to the previous thread "IFACE: Adding IPv4 address to ng0 failed: File exists + iBGP or OSPF" but I am on freeBSD 14.2, it should be fixed....
Hello, normally mpd should delete the connected route when the pppoe session is terminated, but in my case we have a "ghost" route still present and when a new pppoe connection is requesting 10.128.32.4 which is supposed to be available, we have an error from mpd : Mar 20 09:00:02 bng1 mpd[77832]: [BT1517-167] IPCP: state change Ack-Rcvd --> Opened Mar 20 09:00:02 bng1 mpd[77832]: [BT1517-167] IPCP: LayerUp Mar 20 09:00:02 bng1 mpd[77832]: [BT1517-167] 10.128.35.254 -> 10.128.32.4 Mar 20 09:00:02...
Hi all, I found this PDF from Hiroki SATO, which quotes MPD5 as PPPoE server with IPv6. I wonder if anybody has already successfully tested DHCPv6 with Prefix Delegation ? Regards.
Yes ! no more more crashes observed. We still have some strange PPPoE incoming request : Sep 8 05:27:06 bng1 mpd[23657]: Ignored incoming PPPoE connection request via ix0_1515: for service "" due to bad length 28172 > 40 Sep 8 05:32:03 bng1 mpd[23657]: Ignored incoming PPPoE connection request via ix0_1515: for service "" due to bad length 204 > 40 Sep 8 05:57:33 bng1 mpd[23657]: Ignored incoming PPPoE connection request via ix0_1515: for service "" due to bad length 24588 > 40 Sep 8 05:59:05 bng1...
Hi Eugene, here is the status : 1) done 2) done 3) waiting for future crash Fyi, we have 2 servers running MPD5 and the process crashes at the same time on both servers, we are suspecting a malformed frame. We will provide backtrace info asap. Stay tuned.
Hi Eugene, it happened again : Sep 1 20:01:21 bng2 mpd[1334]: caught fatal signal SEGV Sep 1 20:01:21 bng2 mpd[1334]: 0x28438e <???> at /usr/local/sbin/mpd5 Sep 1 20:01:21 bng2 mpd[1334]: 0x283ffa <???> at /usr/local/sbin/mpd5 Sep 1 20:01:21 bng2 mpd[1334]: 0x80077ab70 <_pthread_sigmask+0x530> at /lib/libthr.so.3 let us what you need to investigate, thank you in advance.
It works much better with FreeBSD 12.2, thanks ! I will post this issue on the right forum.
We use mpd as a pppoe server, it accepts pppoe requests from several vlan interfaces on 10GE NIC. We have around 400 ngX sessions. We upgraded from 5.8 to 5.9_1 on FreeBSD12.2 today, we monitor it to check if crash happens again, we will let you know.
I've tried this but got "ipfw pullup failed" messages on console and system freeze with sometimes kernel crashes on FreeBSD 13.0, any idea ?
Hi Eugene, I will test BSDrp1.99 which has MPD 5.9. Is there some debug option to enable on mpd ? log +all is enough ?
Hi all, got sometimes this crash : mpd.log : Aug 22 10:23:17 mpd[25132]: caught fatal signal SEGV Aug 22 10:23:17 mpd[25132]: 0x283b3e <???> at /usr/local/sbin/mpd5 Aug 22 10:23:17 mpd[25132]: 0x2837aa <???> at /usr/local/sbin/mpd5 Aug 22 10:23:17 mpd[25132]: 0x800766bb5 <_pthread_sigmask+0x535> at /lib/libthr.so.3 any fix or tuning ? I am using BSDrp 1.97 MPD 5.8 Many thanks in advance.