centos7.4 ipv6 automatically disabled after a while.

Comments

• sirpumpkin Member

  November 2017

  Try adding

  net.ipv6.conf.default.accept_ra=0 net.ipv6.conf.default.autoconf=0 net.ipv6.conf.all.accept_ra=0 net.ipv6.conf.all.autoconf=0 net.ipv6.conf.eth0.accept_ra=0 net.ipv6.conf.eth0.autoconf=0

  to /etc/sysctl.d/50-IPv6.conf, reboot and see if it helps
• svmo Member

  November 2017

  @qtwrk said:
  Hi , this is most troubling

  I have plesk installed on centos 7.4 , I configured ipv6 before plesk , and tested by
  ping6 google.com confirm it's working

  and then I set my site to run on ipv6 , initially it works , but after a while , like an hour or so , ipv6 is not longer working , ping6 google.com failed
  reboot will fix the issue , but I can't just reboot it every now and then.

  I have 2 VPS with same installation , this one lost its ipv6 every once a while , but other VPS seems keep it working all the time.

  anyone has ever experienced such issue ?

  a help is appreciated .

  How is your system set up to get its ipv6 address ... ?

  Sounds like your system autoconfigures an ipv6 on start up by sending a router solicitation
  and getting a reply - however if the router is not sending out periodic router advertisements
  or they are blocked/lost your machine will loose its address ... until next reboot.

  You could configure the address you get as well as gateway and default route manually.
• Hxxx Member

  November 2017

  its hacked! start over.
• WSS Member

  November 2017

  when it's working, run 'ip -6 route' and paste the results here.
• qtwrk Member

  November 2017

  @WSS said:
  when it's working, run 'ip -6 route' and paste the results here.

  unreachable ::/96 dev lo metric 1024 error -113
  unreachable ::ffff:0.0.0.0/96 dev lo metric 1024 error -113
  unreachable 2002:a00::/24 dev lo metric 1024 error -113
  unreachable 2002:7f00::/24 dev lo metric 1024 error -113
  unreachable 2002:a9fe::/32 dev lo metric 1024 error -113
  unreachable 2002:ac10::/28 dev lo metric 1024 error -113
  unreachable 2002:c0a8::/32 dev lo metric 1024 error -113
  unreachable 2002:e000::/19 dev lo metric 1024 error -113
  2a03:4000:21:4::/64 dev eth0 proto kernel metric 256
  unreachable 3ffe:ffff::/32 dev lo metric 1024 error -113
  fe80::1 dev eth0 proto static metric 100
  fe80::/64 dev eth0 proto kernel metric 256
  default via fe80::1 dev eth0 proto static metric 100
  

  I have compared when it's not working , output the same.

  @svmo said:

  How is your system set up to get its ipv6 address ... ?

  I got it followed this guide here https://unix.stackexchange.com/questions/316730/how-to-setup-ipv6-on-centos

  @sirpumpkin said:
  Try adding
  to /etc/sysctl.d/50-IPv6.conf, reboot and see if it helps

  thanks , will see how it goes.
• WSS Member

  November 2017

  @qtwrk said:
  I have compared when it's not working , output the same.

  Yeah. It's not dropping your route because it's out of band, which @sirpumpkin was providing a workaround for.
• qtwrk Member

  November 2017

  @WSS said:

  @qtwrk said:
  I have compared when it's not working , output the same.

  Yeah. It's not dropping your route because it's out of band, which @sirpumpkin was providing a workaround for.

  thanks , I will see how it goes
• qtwrk Member

  November 2017

  @sirpumpkin said:
  Try adding
  to /etc/sysctl.d/50-IPv6.conf, reboot and see if it helps

  @WSS said:
  Yeah. It's not dropping your route because it's out of band, which @sirpumpkin was providing a workaround for.

  thanks , but unfortunately ipv6 still went off after a while...
• vanhels Member

  November 2017

  It's strange, I have a VPS with Centos-7 Minimal, working with IPV6, and it works very well without any additional configuration except to place it in the adapter, you should consider opening a ticket to your provider.
• qtwrk Member

  November 2017

  @vanhels said:
  It's strange, I have a VPS with Centos-7 Minimal, working with IPV6, and it works very well without any additional configuration except to place it in the adapter, you should consider opening a ticket to your provider.

  indeed it is strange , I have 2 VPS from same provider , with exact same conf and setting , one works all the time , another one lost ipv6 few minutes after reboot
• vanhels Member

  November 2017

  For the above, open a ticket with your provider and tell him that his 2 VPS are configured in the same way, and that one works in IPV6 and in another not, sure that it helps.
• qtwrk Member

  November 2017

  @vanhels said:
  For the above, open a ticket with your provider and tell him that his 2 VPS are configured in the same way, and that one works in IPV6 and in another not, sure that it helps.

  arrrgggg , this problem is killing me here.

  ticket sent , let's see how it goes ...

  thanks for advice
• qtwrk Member

  November 2017

  I worked out , but still didnt get it why and how.

  The provider is Netcup , if i set ipv6 during ISO installation , ipv6 works fine all the way .

  But if i install system first and then add ipv6 sftwards , ipv6 will become unusable after a while.

  I write it here in case someone who got same problme and luckily google'd this thread , hope this could be a help.
• WSS Member

  November 2017

  For NetCup, @sirpumpkin's suggestion generally works.
• qtwrk Member

  November 2017

  @WSS said:
  For NetCup, @sirpumpkin's suggestion generally works.

  well , I think I am the "de-generally" case ...
  but any way , it works now ...