You are not logged in.

#1 2016-06-26 03:17:15

ondoho
Member
Registered: 2013-04-30
Posts: 692
Website

no more ipv4 on eth0 - ipv6 seems to be working fully

This has happened on 2 machines, both today.
it didn't start simultaneously; on the netbook it has been present for a while already, but i didn't investigate until now.
on my desktop it started today, presumably after an update & reboot.
kernel versions, netbook: 4.4.14-1-lts; desktop: 4.6.2-1-ARCH.

right now, both machines hang on the same router.
for this thread, i am concentrating onthe desktop.

but at the moment the symptoms are identical:

$ ping -c1 google.com
PING google.com(arn09s05-in-x0e.1e100.net (2a00:1450:400f:804::200e)) 56 data bytes
64 bytes from arn09s05-in-x0e.1e100.net (2a00:1450:400f:804::200e): icmp_seq=1 ttl=55 time=17.3 ms

--- google.com ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 17.334/17.334/17.334/0.000 ms

$ ping -c1 2a00:1450:400f:805::200e
PING 2a00:1450:400f:805::200e(2a00:1450:400f:805::200e) 56 data bytes
64 bytes from 2a00:1450:400f:805::200e: icmp_seq=1 ttl=55 time=16.4 ms

--- 2a00:1450:400f:805::200e ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 16.476/16.476/16.476/0.000 ms

$ ping -c1 archlinux.org
connect: Network is unreachable

$ ping -c1 8.8.8.8
connect: Network is unreachable

$ ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
    link/ether nn:nn:nn:nn:nn:nn brd ff:ff:ff:ff:ff:ff
    inet6 nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn/64 scope global mngtmpaddr noprefixroute dynamic 
       valid_lft 17990sec preferred_lft 11240sec
    inet6 nnnn:nnnn:nnnn:nnnn::4/128 scope global noprefixroute dynamic 
       valid_lft 2836sec preferred_lft 1036sec
    inet6 nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn/64 scope global mngtmpaddr noprefixroute dynamic 
       valid_lft 17990sec preferred_lft 11240sec
    inet6 nnnn::nnnn:nnnn:nnnn:nnnn:nnnn/64 scope link 
       valid_lft forever preferred_lft forever

$ systemctl status dhcpcd
● dhcpcd.service - dhcpcd on all interfaces
   Loaded: loaded (/usr/lib/systemd/system/dhcpcd.service; disabled; vendor preset: disabled)
   Active: active (running) since Sun 2016-06-26 05:45:45 EEST; 19min ago
  Process: 3810 ExecStart=/usr/bin/dhcpcd -q -b (code=exited, status=0/SUCCESS)
 Main PID: 3811 (dhcpcd)
    Tasks: 1 (limit: 512)
   CGroup: /system.slice/dhcpcd.service
           └─3811 /usr/bin/dhcpcd -q -b

Jun 26 05:45:45 arch systemd[1]: Started dhcpcd on all interfaces.
Jun 26 05:45:45 arch dhcpcd[3811]: DUID nn:nn:nn:nn:nn:nn:nn:nn:nn:nn:nn:nn:nn
Jun 26 05:45:45 arch dhcpcd[3811]: eth0: IAID 8a:c1:b3:ae
Jun 26 05:45:45 arch dhcpcd[3811]: eth0: soliciting an IPv6 router
Jun 26 05:45:45 arch dhcpcd[3811]: eth0: Router Advertisement from nnnn::nnnn:nnnn:nnnn:nnnn
Jun 26 05:45:45 arch dhcpcd[3811]: eth0: adding address nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn/64
Jun 26 05:45:45 arch dhcpcd[3811]: eth0: adding route to nnnn:nnnn:nnnn::/64
Jun 26 05:45:45 arch dhcpcd[3811]: eth0: adding default route via nnnn::nnnn:nnnn:nnnn:nnnn
Jun 26 05:45:46 arch dhcpcd[3811]: eth0: soliciting a DHCPv6 lease
Jun 26 05:45:46 arch dhcpcd[3811]: eth0: soliciting a DHCP lease

$ lspci -v
(snip)
02:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller (rev 15)
	Subsystem: Micro-Star International Co., Ltd. [MSI] Device 7982
	Flags: bus master, fast devsel, latency 0, IRQ 318
	I/O ports at d000 [size=256]
	Memory at df104000 (64-bit, non-prefetchable) [size=4K]
	Memory at df100000 (64-bit, non-prefetchable) [size=16K]
	Capabilities: [40] Power Management version 3
	Capabilities: [50] MSI: Enable+ Count=1/1 Maskable- 64bit+
	Capabilities: [70] Express Endpoint, MSI 01
	Capabilities: [b0] MSI-X: Enable- Count=4 Masked-
	Capabilities: [100] Advanced Error Reporting
	Capabilities: [140] Virtual Channel
	Capabilities: [160] Device Serial Number nn-nn-nn-nn-nn-nn-nn-nn
	Capabilities: [170] Latency Tolerance Reporting
	Capabilities: [178] L1 PM Substates
	Kernel driver in use: r8169
	Kernel modules: r8169

only ipv6, no ipv4.
i suspect that is the reason why the internet is sort of half gone.
as expected, i can open google.com in a browser, and connect to most sites by opening links from google search results.
i can connect to some sites directly, e.g. wikipedia, but not e.g. to archlinux.org or my own website. not even to my router.
i can connect to bbs.archlinux.org; i am writing this from my desktop.

right now, both machines hang on the same router.
the netbook won't even connect to my phone's wifi hotspot anymore. wifi-menu sees it, but i can't connect.
yet i was able to update the system with pacman.

my debian server also hangs on the same router, but is not affected. i can also visit it from the outside (from my phone, not from my desktop).

systemd "just connects" to the internet on my desktop, but on the netbook i have been using different software, but lately only netctl (mostly because of wifi-menu).

while troubleshooting, i went through this: https://bbs.archlinux.org/viewtopic.php?id=213363
because i have been having that symptom, too, and installed the file as per post #11. on both machines.

i went through the non-optional motions here: https://wiki.archlinux.org/index.php/sy … using_DHCP
added /etc/systemd/network/eth0.network as per wiki.
no errors, everything seems to take normally.
on both machines.

rebooted; no change.

any help greatly appreciated!

=============================================================

additional info:

right now i am doing everything in stereo, and the results are about 90% persistent across both devices (desktop and netbook).
for example, i added some ipv4 addresses to /etc/systemd/resolved.conf to use as nameservers (OpenNIC).
on my desktop, i couldn't ping google anymore, on my netbook i still could.
i commented these entries out again, on both devices, and restarted systemd-resolved.service, and the connectivity is there as described earlier.

apart from that, the journalctl output is virtually identical.

it really looks as if ipv4 support has been dropped completely from my system; i can't even find error messages about ipv4:

# journalctl -b|grep eth0
Jun 26 11:53:17 arch kernel: r8169 0000:02:00.0 eth0: RTL8168h/8111h at 0xffffc90000ebe000, nn:nn:nn:nn:nn:nn, XID 14100800 IRQ 318
Jun 26 11:53:17 arch kernel: r8169 0000:02:00.0 eth0: jumbo features [frames: 9200 bytes, tx checksumming: ko]
Jun 26 11:53:20 arch systemd[1]: Starting dhcpcd on eth0...
Jun 26 11:53:21 arch dhcpcd[457]: eth0: waiting for carrier
Jun 26 11:53:21 arch kernel: r8169 0000:02:00.0 eth0: link down
Jun 26 11:53:21 arch kernel: r8169 0000:02:00.0 eth0: link down
Jun 26 11:53:21 arch kernel: IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready
Jun 26 11:53:24 arch dhcpcd[457]: eth0: carrier acquired
Jun 26 11:53:24 arch kernel: r8169 0000:02:00.0 eth0: link up
Jun 26 11:53:24 arch kernel: IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
Jun 26 11:53:24 arch systemd-networkd[474]: eth0: Gained carrier
Jun 26 11:53:24 arch dhcpcd[457]: eth0: adding address fe80::51fb:dd73:39f9:6d0
Jun 26 11:53:24 arch dhcpcd[457]: eth0: IAID 8a:c1:b3:ae
Jun 26 11:53:25 arch dhcpcd[457]: eth0: soliciting a DHCP lease
Jun 26 11:53:25 arch dhcpcd[457]: eth0: soliciting an IPv6 router
Jun 26 11:53:26 arch systemd-networkd[474]: eth0: Gained IPv6LL
Jun 26 11:53:26 arch dhcpcd[457]: eth0: Router Advertisement from nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn
Jun 26 11:53:26 arch systemd-networkd[474]: eth0: DHCPv6 address nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn/128 timeout preferred 1800 valid 3600
Jun 26 11:53:26 arch dhcpcd[457]: eth0: adding address nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn/64
Jun 26 11:53:26 arch dhcpcd[457]: eth0: adding route to nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn/64
Jun 26 11:53:26 arch dhcpcd[457]: eth0: adding default route via nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn
Jun 26 11:53:26 arch dhcpcd[457]: eth0: soliciting a DHCPv6 lease
Jun 26 11:53:28 arch systemd[1]: Started dhcpcd on eth0.
Jun 26 11:53:28 arch kernel: [UFW BLOCK] IN=eth0 OUT= MAC= SRC=XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX DST=XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX LEN=70 TC=0 HOPLIMIT=255 FLOWLBL=145361 PROTO=UDP SPT=5355 DPT=5355 LEN=30 
Jun 26 11:53:29 arch kernel: [UFW BLOCK] IN=eth0 OUT= MAC= SRC=XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX DST=XXXXXXXXXXXXXXXXXXXXXXXXXXXXXX LEN=70 TC=0 HOPLIMIT=255 FLOWLBL=145361 PROTO=UDP SPT=5355 DPT=5355 LEN=30 
Jun 26 11:53:29 arch kernel: [UFW BLOCK] IN=eth0 OUT= MAC= SRC=XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX DST=XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX LEN=70 TC=0 HOPLIMIT=255 FLOWLBL=145361 PROTO=UDP SPT=5355 DPT=5355 LEN=30 
Jun 26 12:08:01 arch systemd-networkd[474]: eth0: DHCPv6 address nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn/128 timeout preferred 1800 valid 3600
Jun 26 12:11:55 arch kernel: [UFW BLOCK] IN=eth0 OUT= MAC= SRC=nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn DST=nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn LEN=70 TC=0 HOPLIMIT=255 FLOWLBL=145361 PROTO=UDP SPT=5355 DPT=5355 LEN=30 
Jun 26 12:11:55 arch kernel: [UFW BLOCK] IN=eth0 OUT= MAC= SRC=nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn DST=nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn LEN=70 TC=0 HOPLIMIT=255 FLOWLBL=145361 PROTO=UDP SPT=5355 DPT=5355 LEN=30 
Jun 26 12:11:56 arch kernel: [UFW BLOCK] IN=eth0 OUT= MAC= SRC=nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn DST=nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn:nnnn LEN=70 TC=0 HOPLIMIT=255 FLOWLBL=145361 PROTO=UDP SPT=5355 DPT=5355 LEN=30 

# journalctl -b|grep -i ipv4

code 1
#  

(i blacked out what looked like sensitive data to me)
(disabling the firewall does not help either, but i wonder what those messages mean)

/etc/systemd/network/eth0.network:

[Match]
Name=eth0

[Network]
DHCP=ipv4

the browser still can't connect to 192.168.0.11, which is my router.

i have been reading some similar threads on the forums, esp. on the networking subforum, and gained a little insight, but no solutions.

Last edited by ondoho (2016-06-26 09:38:29)

Offline

#2 2016-06-26 10:27:06

ondoho
Member
Registered: 2013-04-30
Posts: 692
Website

Re: no more ipv4 on eth0 - ipv6 seems to be working fully

there has been an interesting development.

i edited /etc/dhcpcd.conf and changed duid to clientid, and commented out noipv4ll. i have only the faintest inkling as to what these options do, but now i get this (and no previous functionality seems to be broken):

# ping -c1 8.8.8.8
PING 8.8.8.8 (8.8.8.8) 56(84) bytes of data.
From 169.254.130.127 icmp_seq=1 Destination Host Unreachable

--- 8.8.8.8 ping statistics ---
1 packets transmitted, 0 received, +1 errors, 100% packet loss, time 0ms

which is a development to the previous "connect: Network is unreachable".
i can connect to my router now, and a few more webpages, but this still doesn't seem to be the end of the story.

i read about duid vs clientid on the archwiki, and it seems ok to change it.
but why is noipv4ll enabled by default, and why do i have to disable it.

Offline

#3 2016-06-26 12:11:14

revellion
Member
From: Sweden
Registered: 2007-04-10
Posts: 54

Re: no more ipv4 on eth0 - ipv6 seems to be working fully

Tried running dhcpcd -T eth0 ?


Hurricane Electric Certified IPv6 Sage

CPU: Core i7-2600 @ 4.0Ghz | RAM: 16GB (4x4GB) | GFX: AMD Radeon R9 290 4096MB VRAM | HDD: 1x 120GB SATA3 Corsair SSD (~500MB/s RW), 1x SATA2 250GB, 1x SATA2 320GB, 1x 180GB SATA3 Intel SSD
*EDIT* Replaced Nvidia GTX 570 for a AMD Radeon R9 290, and added an extra SSD 180GB

Offline

#4 2016-06-26 13:55:40

\hbar
Member
Registered: 2014-03-15
Posts: 165

Re: no more ipv4 on eth0 - ipv6 seems to be working fully

IPV4LL means 'link-local IPv4', which does not enable you to connect to anything outside your local network.

Are you using dhcpcd, or systemd-networkd?

Offline

#5 2016-06-26 15:17:11

ondoho
Member
Registered: 2013-04-30
Posts: 692
Website

Re: no more ipv4 on eth0 - ipv6 seems to be working fully

thanks to both.
i have since been able to connect normally again.
changing duid to clientid in /etc/dhcpcd.conf seems to have helped.
noipv4ll is uncommented again.

\hbar wrote:

Are you using dhcpcd, or systemd-networkd?

i have been asking myself that.
i stopped & disabled dhcpcd.service, but after a reboot i'm still seeing this:

 systemctl status
● machine
    State: running
     Jobs: 0 queued
   Failed: 0 units
    Since: Sun 2016-06-26 13:44:34 EEST; 4h 20min ago
   CGroup: /
           ├─init.scope
           │ └─1 /sbin/init
           ├─system.slice
           │ ├─dbus.service
           │ │ └─484 /usr/bin/dbus-daemon --system --address=systemd: --nofork --nopidfile --systemd-activation
           │ ├─system-dhcpcd.slice
           │ │ └─dhcpcd@eth0.service
           │ │   └─552 /usr/bin/dhcpcd -q -w eth0
           │ ├─dhcpcd.service
           │ │ └─937 /usr/bin/dhcpcd -q -b
           │ ├─systemd-journald.service
           │ │ └─197 /usr/lib/systemd/systemd-journald
           │ ├─udisks2.service
           │ │ └─2664 /usr/lib/udisks2/udisksd --no-debug
           │ ├─systemd-resolved.service
           │ │ └─553 /usr/lib/systemd/systemd-resolved
           │ ├─systemd-timesyncd.service
           │ │ └─475 /usr/lib/systemd/systemd-timesyncd
           │ ├─systemd-logind.service
           │ │ └─483 /usr/lib/systemd/systemd-logind
           │ ├─systemd-networkd.service
           │ │ └─715 /usr/lib/systemd/systemd-networkd
           │ ├─systemd-udevd.service
           │ │ └─242 /usr/lib/systemd/systemd-udevd
           │ └─polkit.service
           │   └─2668 /usr/lib/polkit-1/polkitd --no-debug
           └─user.slice

which seems to indicate that both are somehow active; but which one takes precedence?

anyhow i'm beginning to suspect that my server, which hangs on the same router, might be under attack, causing the router to lock up.
now that my 2 arch machines are ok again, i cannot ssh into it anymore (connection refused), although ssh-agent is running just fine, and sshd on the server.
there's been other symptoms, even when accessing the server directly, physically. i just took it offline for now, i need a breather.

Offline

#6 2016-06-26 15:43:35

\hbar
Member
Registered: 2014-03-15
Posts: 165

Re: no more ipv4 on eth0 - ipv6 seems to be working fully

'dhcpcd@eth0'  and 'dhcpcd' are two different services. Are they both disabled?

I would suspect an interference between dhcpcd and systemd-networkd. It seems unlikely to me that an attack on your server would make the DHCP server on your router only give out ipv6 addresses. On the other hand, systemd-networkd was configured to only sollicit for an ipv4 address, which would explain why the ipv6 was left alone.

Offline

#7 2016-06-27 17:56:54

ondoho
Member
Registered: 2013-04-30
Posts: 692
Website

Re: no more ipv4 on eth0 - ipv6 seems to be working fully

\hbar wrote:

'dhcpcd@eth0'  and 'dhcpcd' are two different services. Are they both disabled?

yes, definitely.
could something else be re-enabling them? i might have to mask it if it comes up again, but it feels wrong to do so.

i also found out that both my desktop and my server are sharing the same internal ip address. i have to change that. i'm sure it can at least partly explain the weird behavior i've been experiencing.

Offline

Board footer

Powered by FluxBB