Am 01/01/13 21:42, schrieb Manfred Antar: > At 11:48 AM 1/1/2013, Gleb Smirnoff wrote: >> On Tue, Jan 01, 2013 at 02:39:58AM -0800, Manfred Antar wrote: >> M> >On Thu, Dec 27, 2012 at 09:05:26AM -0800, Manfred Antar wrote: >> M> >M> For the past few days the loopback interface 127.0.0.1 is broken on current. >> M> >M> The last good kernel that works for me is r244662: Mon Dec 24 06:43:07 PST 2012 >> M> >M> Here are some of the errors from current today: >> M> > >> M> >Can you please track this down to the specific revision that made the >> M> >regression? The timeframe is samll, so binary search probably won't >> M> >take long time. >> M> Hi >> M> I'm not sure how to do this :( >> M> I just switched from CVS to SVN 2 weeks ago. >> M> I still have a current CVS tree on my machine. >> M> I guess i can try to check out a /usr/src/sys from the 25 on and see what changes were made. >> M> I'm just learning SVN so not sure how to do it with that. >> >> With SVN you don't need to care about dates, you just go through the >> sequential list of revisions. So you just seek between r244662 and the >> most recent revision tht didn't work for you. On first step you >> divide the bad revision (let it be 244768) and good one r244662: >> >> (244662 + 244768)/2 = 244715 >> >> Ok, it is 244715. Let's take it and test: >> >> svn up -r 244715 >> >> If it is fine, you check the one in the middle between 244715and >> 244768. Otherwise you chek in between 244715 and 244662. And so on... >> >> -- >> Totus tuus, Glebius. > > OK > I tracked it down to revision 244678 > 244677 works > the files involved are: > (src)5012}svn up -r 244678 > Updating '.': > U sys/netinet/in.c > U sys/netinet6/in6.c > Updated to revision 244678 > > It seems like the ip6 for localhost is configured but ip4 isn't: > Setting hostname: pozo.com. > ifa_del_loopback_route: deletion failed: 3 > ifconfig: ioctl (SIOCAIFADDR): File exists > bge0: link state changed to DOWN > ifa_del_loopback_route: deletion failed: 3 > > lo0: flags=8049<UP,LOOPBACK,RUNNING,MULTICAST> metric 0 mtu 16384 > options=600003<RXCSUM,TXCSUM,RXCSUM_IPV6,TXCSUM_IPV6> > inet6 ::1 prefixlen 128 > inet6 fe80::1%lo0 prefixlen 64 scopeid 0x2 > nd6 options=21<PERFORMNUD,AUTO_LINKLOCAL> > > From 244677: > Setting hostname: pozo.com. > > Starting Network: lo0 bge0. > lo0: flags=8049<UP,LOOPBACK,RUNNING,MULTICAST> metric 0 mtu 16384 > options=600003<RXCSUM,TXCSUM,RXCSUM_IPV6,TXCSUM_IPV6> > inet 127.0.0.1 netmask 0xff000000 > inet6 ::1 prefixlen 128 > inet6 fe80::1%lo0 prefixlen 64 scopeid 0x2 > nd6 options=21<PERFORMNUD,AUTO_LINKLOCAL> > > Manfred I have a similar issue here. It came out after a lot of problems with the amd automounter which rendered a server completeley useless and the very interesting issue is, that the loopback interface, lo0, has IPv6 configured, but it doesn't have IPv4 configured anymore. After a recompilation of the kernel on the 31.12.2012 and a reboot, the box is not responsive anymore (it a remote box I reach not before 07th Jan). NAGIOS is still reporting ICMP ping is good, but every other service is down. On another server I do not face this problem - or have faced that problem and I use the very same OS revisions on most of the CURRENT boxes. But all machines started suffering from some amd automounter issue before the problem got serious. This is strange ... oh
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:40:33 UTC