"ipv6disable=1" in F19 is ignored
i get tired every few months research how IPv6 to disable because it permanently changes, maybe it is F19/systemd maybe it is the F20 3.10.0-1.fc20.x86_64 because it is the only 3.10 build currently, however the kernel-line is clear
on pure ipv4 networks there is no need for "inet6 fe80::20c:29ff:fe30:82b9" period
eth0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500 inet 192.168.196.18 netmask 255.255.255.0 broadcast 192.168.196.255 inet6 fe80::20c:29ff:fe30:82b9 prefixlen 64 scopeid 0x20<link> ether 00:0c:29:30:82:b9 txqueuelen 1000 (Ethernet) RX packets 755 bytes 71786 (70.1 KiB) RX errors 0 dropped 0 overruns 0 frame 0 TX packets 594 bytes 162204 (158.4 KiB) TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0
On Sun, 2013-07-07 at 18:18 +0200, Reindl Harald wrote:
"ipv6disable=1" in F19 is ignored
i get tired every few months research how IPv6 to disable because it permanently changes, maybe it is F19/systemd maybe it is the F20 3.10.0-1.fc20.x86_64 because it is the only 3.10 build currently, however the kernel-line is clear
on pure ipv4 networks there is no need for "inet6 fe80::20c:29ff:fe30:82b9" period
eth0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500 inet 192.168.196.18 netmask 255.255.255.0 broadcast 192.168.196.255 inet6 fe80::20c:29ff:fe30:82b9 prefixlen 64 scopeid 0x20<link> ether 00:0c:29:30:82:b9 txqueuelen 1000 (Ethernet) RX packets 755 bytes 71786 (70.1 KiB) RX errors 0 dropped 0 overruns 0 frame 0 TX packets 594 bytes 162204 (158.4 KiB) TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0
Do you mind to disable IPV6 in ifcfg-eth0 (comment out IPV6_...)? You will probably lose your default (ipv4) route! You're not alone.
C. Sava
Am 08.07.2013 15:56, schrieb Cristian Sava:
On Sun, 2013-07-07 at 18:18 +0200, Reindl Harald wrote:
"ipv6disable=1" in F19 is ignored
i get tired every few months research how IPv6 to disable because it permanently changes, maybe it is F19/systemd maybe it is the F20 3.10.0-1.fc20.x86_64 because it is the only 3.10 build currently, however the kernel-line is clear
on pure ipv4 networks there is no need for "inet6 fe80::20c:29ff:fe30:82b9" period
eth0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500 inet 192.168.196.18 netmask 255.255.255.0 broadcast 192.168.196.255 inet6 fe80::20c:29ff:fe30:82b9 prefixlen 64 scopeid 0x20<link> ether 00:0c:29:30:82:b9 txqueuelen 1000 (Ethernet) RX packets 755 bytes 71786 (70.1 KiB) RX errors 0 dropped 0 overruns 0 frame 0 TX packets 594 bytes 162204 (158.4 KiB) TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0
Do you mind to disable IPV6 in ifcfg-eth0 (comment out IPV6_...)? You will probably lose your default (ipv4) route! You're not alone.
this has notthing to do with ifcfg-files there is "IPV6INIT=no" since many years on my machines
"ipv6disable=1" is for disable the *entire* ipv6 stack and a good admin disableds *anything* which is not in use for security reasons - i have ip4-rules on servers which are even block 127.0.0.1 for Samba while it is allowed from specific LAN IP's but not from php-scripts __________________
syslog message at boot: ipv6: Loaded, but administratively disabled, reboot required to enable
this does currently no longer work and every release the magic how to achive this is randomly changed from sysctl to kernel-params, different params or modprobe-aliases a few years before
On Mon, 2013-07-08 at 16:15 +0200, Reindl Harald wrote:
Am 08.07.2013 15:56, schrieb Cristian Sava:
On Sun, 2013-07-07 at 18:18 +0200, Reindl Harald wrote:
"ipv6disable=1" in F19 is ignored
i get tired every few months research how IPv6 to disable because it permanently changes, maybe it is F19/systemd maybe it is the F20 3.10.0-1.fc20.x86_64 because it is the only 3.10 build currently, however the kernel-line is clear
on pure ipv4 networks there is no need for "inet6 fe80::20c:29ff:fe30:82b9" period
eth0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500 inet 192.168.196.18 netmask 255.255.255.0 broadcast 192.168.196.255 inet6 fe80::20c:29ff:fe30:82b9 prefixlen 64 scopeid 0x20<link> ether 00:0c:29:30:82:b9 txqueuelen 1000 (Ethernet) RX packets 755 bytes 71786 (70.1 KiB) RX errors 0 dropped 0 overruns 0 frame 0 TX packets 594 bytes 162204 (158.4 KiB) TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0
Do you mind to disable IPV6 in ifcfg-eth0 (comment out IPV6_...)? You will probably lose your default (ipv4) route! You're not alone.
this has notthing to do with ifcfg-files there is "IPV6INIT=no" since many years on my machines
"ipv6disable=1" is for disable the *entire* ipv6 stack and a good admin disableds *anything* which is not in use for security reasons - i have ip4-rules on servers which are even block 127.0.0.1 for Samba while it is allowed from specific LAN IP's but not from php-scripts __________________
syslog message at boot: ipv6: Loaded, but administratively disabled, reboot required to enable
this does currently no longer work and every release the magic how to achive this is randomly changed from sysctl to kernel-params, different params or modprobe-aliases a few years before
I wanted to say that if IPV6=no (does not matter where or how) many things will go wrong. Some bugs are present in the networking and I was hit by some of them. Hope they will be fixed soon.
C. Sava
Am 08.07.2013 18:16, schrieb Cristian Sava:
On Mon, 2013-07-08 at 16:15 +0200, Reindl Harald wrote:
syslog message at boot: ipv6: Loaded, but administratively disabled, reboot required to enable
this does currently no longer work and every release the magic how to achive this is randomly changed from sysctl to kernel-params, different params or modprobe-aliases a few years before
I wanted to say that if IPV6=no (does not matter where or how) many things will go wrong. Some bugs are present in the networking and I was hit by some of them. Hope they will be fixed soon
i doubt that
having a ton production machines with F17/F18 with all kind of network services (dns, dhcp, httpd, afpd, smbd, ftp, smtp/pop3/imap/sieve) as well as routing/firewalls/NAT/WLAN-AP's and maybe any kind of network services except NFS since years for some hundret domains and there is no single network problem all the years