Re: Corrupted IPv6 routing table
On 1/7/2015 6:14 PM, Przemyslaw Frasunek wrote: Dear all, We are running FreeBSD 9.2-RELEASE-p3 on few PPPoE access servers, each servicing about 1000 customers. Each server exchanges customers' /32 (for IPv4) and /64 (for IPv6) routes using OSPF and BIRD. Few times in a month, we are experiencing routing table corruption, which causes spurious IPv6 routes appearing with prefixlen 0: I saw this a few yrs ago when I tried mpd with ipv6 after about a month. There are some threads (i think on freebsd-net) about it and some of the areas that needed to be fixed. I had to give up on v6 and mpd as a result. I seem to recall some commits to RELENG_10 that might had addresses some of the bugs I saw, but thats just speculation. Have you tried RELENG_10 as an LNS ? ---Mike -- --- Mike Tancsa, tel +1 519 651 3400 Sentex Communications, m...@sentex.net Providing Internet services since 1994 www.sentex.net Cambridge, Ontario Canada http://www.tancsa.com/ ___ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
Re: Corrupted IPv6 routing table
-BEGIN PGP SIGNED MESSAGE- Hash: SHA256 I use mpd5 v4 and v6 in freebsd 10.0 and 10.1, I had no problem just like your for now. On 01/08/2015 09:54 AM, Mike Tancsa wrote: > On 1/7/2015 6:14 PM, Przemyslaw Frasunek wrote: >> Dear all, >> >> We are running FreeBSD 9.2-RELEASE-p3 on few PPPoE access >> servers, each servicing about 1000 customers. Each server >> exchanges customers' /32 (for IPv4) and /64 (for IPv6) routes >> using OSPF and BIRD. >> >> Few times in a month, we are experiencing routing table >> corruption, which causes spurious IPv6 routes appearing with >> prefixlen 0: > > I saw this a few yrs ago when I tried mpd with ipv6 after about a > month. There are some threads (i think on freebsd-net) about it and > some of the areas that needed to be fixed. I had to give up on v6 > and mpd as a result. I seem to recall some commits to RELENG_10 > that might had addresses some of the bugs I saw, but thats just > speculation. Have you tried RELENG_10 as an LNS ? > > ---Mike > > - -- = Tiago Felipe Gonçalves. Gerente de Infraestrutura de TI. +55 19 981811543 PGP Fingerprint - A2:82:BD:48:EE:8D:C4:99:C2:4E:81:D4:C4:7B:1C:2E:C7:F3:04:C9 = -BEGIN PGP SIGNATURE- Version: GnuPG v2 iQIcBAEBCAAGBQJUrnMuAAoJEMR7HC7H8wTJMXcP/3SGWpZOayZSoFr0eatFu+m7 LskNoUkszQLqy/7L+sGdomPQHi3Cudb+1+PyxKtpWwPcbQJXFyGrB4vxkyzqV5LZ 8te1/Zw0JBD7y9CKE3NswXpbZvggojTyxvr9GVna2rTIa8LQJnxbiVZFnzmcSlq4 fDv/NlPO58qqhI7cXa9JXpbb6wGM/IF5R/f1Hg1LOkf6nEfxalAeDV4R6g37bKgL OS/gcf+33VCQWirnNlCNgJ/rIV9YB5n+5BTj+ZsWpg1W/ZpN00hbttOnqQhpLLPV eLS/vBgjmoxwHWpdjonEpLYVKOuZK3bADeg1u0PuyVT4qCi9t7rzhMMRKlKxWvnv vbIGIt7BtnlnlqbvFvDbHT0GNt0LqcrW3lmTnSV0qbWdbT5oOWVQIQQOBZ5MmDYG cqew8rK0i5dSokYaA4eaQzQDqndVWbwQaiXXm+rAmoG+HsF1602Ez98ilkxjY0DH fCvFGJHhWJgjSKgl+XEMeLEXD/FfWG52n9UmOEEoRsLm2nGnvmtmln61pyjLz/dt InUBS3zu935FXEuveaFZEmbEjdmqEndQOwfF8Xkt8fdvZ8X+wii0Uq6/aq8o3Eib s1iG/LFPHuE2UlMSKj/HqIu5Il3s68xe1zYr2e3fbzwsx/7FMJL9Mrf93IP4UQmP EC2pHoxaqHqDmUYGHC7D =bkKO -END PGP SIGNATURE- ___ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
Re: Corrupted IPv6 routing table
On 08.01.2015 02:14, Przemyslaw Frasunek wrote: > Dear all, > > We are running FreeBSD 9.2-RELEASE-p3 on few PPPoE access servers, each > servicing about 1000 customers. Each server exchanges customers' /32 (for > IPv4) > and /64 (for IPv6) routes using OSPF and BIRD. > > Few times in a month, we are experiencing routing table corruption, which > causes > spurious IPv6 routes appearing with prefixlen 0: > > # netstat -nr -f inet6 | grep "/\0" > 2a02:2928:6:989e::/0 fe80::21e:67ff:fe02:e82b%vlan0 UG1 > vlan0 > 2a02:2928:6:989f::/0 fe80::21e:67ff:fe02:e82b%vlan0 UG1 > vlan0 > > It is impossible to delete such routes and they override default route, making > IPv6 networking nonfunctional. They definitely does not originate from OSPF, > as > there is an export filter which denies prefixlen < 32. > > Has anyone seen this before? Fixed in 9 by r257389 (So you should try either stable or 10.x). > ___ > freebsd-net@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-net > To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org" > ___ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
How to get the IP and TCP fields in mbuf
Hi, I am wondering if there is existing macros in the FreeBSD kernel to extract to get to the point of the IP and TCP head in a given mbuf. Thanks, Wei ___ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
Re: How to get the IP and TCP fields in mbuf
Hi Wei, Are you looking for something like this? + struct ip *ip = mtod(m, struct ip *); m is a struct mbuf *. You cann look into my knock-patch for FreeBSD [0] for further examples. Best, Julian [0] https://gnunet.org/knock On 08.01.2015 14:18, Wei Hu wrote: > Hi, > > I am wondering if there is existing macros in the FreeBSD kernel to extract > to get to the point of the IP and TCP head in a given mbuf. > > Thanks, > Wei > > > ___ > freebsd-net@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-net > To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org" > -- Julian Kirsch Chair for IT-Security (I20) Department of Computer Science, Technische Universität München Boltzmannstrasse 3, D-85748 Garching bei München Phone: +49-89-289-18587 Room: 01.08.53 ___ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
Re: Corrupted IPv6 routing table
>> Has anyone seen this before? > Fixed in 9 by r257389 (So you should try either stable or 10.x). Thanks a lot! I'll schedule an upgrade. ___ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
Re: CARP Problem/Bug? on 10.1-RELEASE
Le Mon, 05 Jan 2015 07:57:33 +0100, Sascha a écrit : Hello, > you made my day! It's working again. But I continue my research. > > Rebooting takes the Port-Channel on the switch down. When the machine > comes up the port-channel on the switch needs some seconds until it's > up. Any spanning tree on the port switch ? You can try to disable it Regards, ___ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
Re: ntpd bind() failure: Can't assign requested address
Hi,
> On Wed, 7 Jan 2015 16:56:58 -0800
> hiren panchasara said:
hiren> I see this message on july 2014 time-frame -CURRENT. I tried to
hiren> disable IPv6 in rc.conf via ipv6_activate_all_interfaces="NO" without
hiren> success.
hiren> Has this been fixed? Or any workarounds?
Perhaps, your interface has an IPv6 link-local address which is
tentative state.
Please try the attached patch. This patch ignores a tentative
address.
Sincerely,
Index: contrib/ntp/ntpd/ntp_io.c
===
--- contrib/ntp/ntpd/ntp_io.c (revision 276829)
+++ contrib/ntp/ntpd/ntp_io.c (working copy)
@@ -1176,12 +1176,13 @@
#ifdef INCLUDE_IPV6_SUPPORT
static isc_boolean_t
-is_anycast(struct sockaddr *sa, char *name)
+is_not_bindable(struct sockaddr *sa, char *name)
{
-#if defined(SIOCGIFAFLAG_IN6) && defined(IN6_IFF_ANYCAST)
+#if defined(SIOCGIFAFLAG_IN6) && \
+ (defined(IN6_IFF_ANYCAST) || defined(IN6_IFF_NOTREADY))
struct in6_ifreq ifr6;
int fd;
- u_int32_t flags6;
+ u_int32_t flags6, exclude = 0;
if (sa->sa_family != AF_INET6)
return ISC_FALSE;
@@ -1197,9 +1198,15 @@
}
close(fd);
flags6 = ifr6.ifr_ifru.ifru_flags6;
- if ((flags6 & IN6_IFF_ANYCAST) != 0)
+#if defined(IN6_IFF_ANYCAST)
+ exclude |= IN6_IFF_ANYCAST;
+#endif
+#if defined(IN6_IFF_NOTREADY)
+ exclude |= IN6_IFF_NOTREADY;
+#endif
+ if ((flags6 & exclude) != 0)
return ISC_TRUE;
-#endif /* !SIOCGIFAFLAG_IN6 || !IN6_IFF_ANYCAST */
+#endif /* !SIOCGIFAFLAG_IN6 || !(IN6_IFF_ANYCAST && IN6_IFF_NOTREADY) */
return ISC_FALSE;
}
#endif /* !INCLUDE_IPV6_SUPPORT */
@@ -1344,7 +1351,7 @@
continue;
#ifdef INCLUDE_IPV6_SUPPORT
- if (is_anycast((struct sockaddr *)&interface.sin, isc_if.name))
+ if (is_not_bindable((struct sockaddr *)&interface.sin, isc_if.name))
continue;
#endif /* !INCLUDE_IPV6_SUPPORT */
--
Hajimu UMEMOTO
u...@mahoroba.org ume@{,jp.}FreeBSD.org
http://www.mahoroba.org/~ume/
___
freebsd-net@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-net
To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
Re: ntpd bind() failure: Can't assign requested address
On Thu, Jan 8, 2015 at 9:28 AM, Hajimu UMEMOTO wrote: > Hi, > >> On Wed, 7 Jan 2015 16:56:58 -0800 >> hiren panchasara said: > > hiren> I see this message on july 2014 time-frame -CURRENT. I tried to > hiren> disable IPv6 in rc.conf via ipv6_activate_all_interfaces="NO" without > hiren> success. > > hiren> Has this been fixed? Or any workarounds? > > Perhaps, your interface has an IPv6 link-local address which is > tentative state. > Please try the attached patch. This patch ignores a tentative > address. Indeed its in tentative state and I figured it was causing the issue. Getting it out of tentative state (by enabling IPv6 by 'inet6 -ifdisabled' "fixed" the problem.) But your patch seems like a right way to do it. I'll try it out and report back. Btw, do you mind opening a phabricator review request so we can get a closure on this? cheers, Hiren ___ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
Re: ntpd bind() failure: Can't assign requested address
On Thu, Jan 8, 2015 at 9:48 AM, hiren panchasara wrote: > On Thu, Jan 8, 2015 at 9:28 AM, Hajimu UMEMOTO wrote: >> Hi, >> >>> On Wed, 7 Jan 2015 16:56:58 -0800 >>> hiren panchasara said: >> >> hiren> I see this message on july 2014 time-frame -CURRENT. I tried to >> hiren> disable IPv6 in rc.conf via ipv6_activate_all_interfaces="NO" without >> hiren> success. >> >> hiren> Has this been fixed? Or any workarounds? >> >> Perhaps, your interface has an IPv6 link-local address which is >> tentative state. >> Please try the attached patch. This patch ignores a tentative >> address. > > Indeed its in tentative state and I figured it was causing the issue. > Getting it out of tentative state (by enabling IPv6 by 'inet6 > -ifdisabled' "fixed" the problem.) > But your patch seems like a right way to do it. I'll try it out and report > back. > Attached patch is fixing the problem for me. cheers, Hiren > Btw, do you mind opening a phabricator review request so we can get a > closure on this? ___ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
Re: CARP Problem/Bug? on 10.1-RELEASE
Spanning Tree is disabled on the ports. Best Regards On 08/01/2015 16:45, Sascha wrote: Le Mon, 05 Jan 2015 07:57:33 +0100, Sascha a écrit : Hello, you made my day! It's working again. But I continue my research. Rebooting takes the Port-Channel on the switch down. When the machine comes up the port-channel on the switch needs some seconds until it's up. Any spanning tree on the port switch ? You can try to disable it Regards, ___ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
[Bug 196501] [em] Intel 82573 nic built on my pdsbm-ln2 1U server and only one port will work.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=196501 --- Comment #5 from Mark --- Here is my pciconf -lv none1@pci0:2:0:0: class=0x02 card=0x109a15d9 chip=0x109a8086 rev=0x00 hdr=0x00 class = network subclass = ethernet em0@pci0:3:0:0: class=0x02 card=0x109a15d9 chip=0x109a8086 rev=0x00 hdr=0x00 class = network subclass = ethernet Please note when this happens I can not install PCI card.. Either a wifi one or Intel 2 port pci.. which is slightly older chips. They work fine if you turn off both built on ports then you can use PCI and both older Intel work but I need more then 2 ports so I am stuck -- You are receiving this mail because: You are the assignee for the bug. ___ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
[Bug 196501] [em] Intel 82573 nic built on my pdsbm-ln2 1U server and only one port will work.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=196501 Mark changed: What|Removed |Added URL||http://pastebin.com/tW9Qe3g ||2 -- You are receiving this mail because: You are the assignee for the bug. ___ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
[Bug 196501] [em] Intel 82573 nic built on my pdsbm-ln2 1U server and only one port will work.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=196501 --- Comment #6 from Mark --- And not a hardware problem, Ubuntu works fine, Windows works fine, only thing that I seen this problem with is going from 8.1 to 10.1 .. Never had any problems and will not have problem if I go back to 8.1 or run some other system -- You are receiving this mail because: You are the assignee for the bug. ___ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
[Bug 196501] [em] Intel 82573 nic built on my pdsbm-ln2 1U server and only one port will work.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=196501 --- Comment #7 from Mark --- Note: I read your reply again. I noticed you do not have a dual port but 2 different chips running. That might by why you can run both -- You are receiving this mail because: You are the assignee for the bug. ___ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "freebsd-net-unsubscr...@freebsd.org"
