| #!/bin/sh -e | |
| hostname=$1 | |
| device=$2 | |
| file=$HOME/.dynv6.addr6 | |
| [ -e $file ] && old=`cat $file` | |
| if [ -z "$hostname" -o -z "$token" ]; then | |
| echo "Usage: token=<your-authentication-token> [netmask=64] $0 your-name.dynv6.net [device]" | |
| exit 1 | |
| fi | |
| if [ -z "$netmask" ]; then | |
| netmask=128 | |
| fi | |
| if [ -n "$device" ]; then | |
| device="dev $device" | |
| fi | |
| address=$(ip -6 addr list scope global $device | grep -v " fd" | sed -n 's/.*inet6 \([0-9a-f:]\+\).*/\1/p' | head -n 1) | |
| if [ -e /usr/bin/curl ]; then | |
| bin="curl -fsS" | |
| elif [ -e /usr/bin/wget ]; then | |
| bin="wget -O-" | |
| else | |
| echo "neither curl nor wget found" | |
| exit 1 | |
| fi | |
| if [ -z "$address" ]; then | |
| echo "no IPv6 address found" | |
| exit 1 | |
| fi | |
| # address with netmask | |
| current=$address/$netmask | |
| if [ "$old" = "$current" ]; then | |
| echo "IPv6 address unchanged" | |
| exit | |
| fi | |
| # send addresses to dynv6 | |
| $bin "http://dynv6.com/api/update?hostname=$hostname&ipv6=$current&token=$token" | |
| $bin "http://ipv4.dynv6.com/api/update?hostname=$hostname&ipv4=auto&token=$token" | |
| # save current address | |
| echo $current > $file |
Hi, im running this on a raspberry Pi behind my ISPs router. (...)
Is it not working? What error is coming up? Would you please send some more information, please?
Kind regards
//neph
Hi nephilim75,
the script is working nicely, however, the IP that is updated to dyn6c is the raspberrys local ip in my home networt (192.168.0.something).
However, I found a way, I substituted line 19 with
v4_address=$(curl ifconfig.me)
I am running this script on my Raspberry Pi 3.
ifconfig -a
pi@raspberrypi:~ $ ifconfig -a
eth0: flags=4099<UP,BROADCAST,MULTICAST> mtu 1500
ether b8:27:eb:49:d4:40 txqueuelen 1000 (Ethernet)
RX packets 0 bytes 0 (0.0 B)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 0 bytes 0 (0.0 B)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0
lo: flags=73<UP,LOOPBACK,RUNNING> mtu 65536
inet 127.0.0.1 netmask 255.0.0.0
inet6 ::1 prefixlen 128 scopeid 0x10
loop txqueuelen 1000 (Local Loopback)
RX packets 0 bytes 0 (0.0 B)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 0 bytes 0 (0.0 B)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0
wlan0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500
inet 192.168.1.15 netmask 255.255.255.0 broadcast 192.168.1.255
inet6 fe80::b6bc:2ed3:e647:93b9 prefixlen 64 scopeid 0x20
ether b8:27:eb:1c:81:15 txqueuelen 1000 (Ethernet)
RX packets 290 bytes 25171 (24.5 KiB)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 211 bytes 38365 (37.4 KiB)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0
When I run the script, I get this output:
pi@raspberrypi:~ $ token= ./dynv6.sh satvam.dynv6.net
no IPv6 address found
Am I doing something wrong here?
The IPv6 "address=…" line gives me the global temporary dynamic IPv6 host address, but what I need for DNS updates is the global dynamic (i.e., MAC-based) host IP, so I use a modified line like this:
address=$(ip -6 addr list scope global permanent dynamic $device | grep -v " fd" | grep -v " temporary" | sed -n 's/.*inet6 \([0-9a-f:]\+\).*/\1/p' | head -n 1)
for mac
address=$(ifconfig | grep inet | grep "autoconf temporary" | sed -E -n 's/.*inet6 ([0-9a-f:]+).*/\1/p' | head -n 1)
When I am trying to run
token=Q-4pOEg4UyAuq8mySeWgAJfiGWXHtQ ./dynv6.sh test123x.dynv6.net
With my own Token and hostname I get the following error.
sed: -e expression #1, char 10: unterminated `s' command
no IPv6 address found
Do I need to add a '-e' and If yes. Where do I please need to place it?
Thx,blub4747
@uhardegger Thanks a lot for sharing your script! I've modified it a bit for my Synology router and added system log prints (to /var/log/messages). The modified version can be found here. The corresponding ddns_provider.conf file is uploaded here.
#!/bin/bash
#Configure
Hostname="example.dynv6.net"
IPVERSION="6" #IPV4=4, IPV6=6
UpdateURL="http://dynv6.com/api/update?hostname=$Hostname&ipv6=auto&token=YOUR_TOKEN"
RecordType=$([ "$IPVERSION" == 4 ] && echo "A" || echo "AAAA")
while true; do
SavedIP=$(dig -$IPVERSION +short $RecordType $Hostname) #check IP which is stored on dynamic DNS
ActualIP=$(dig -$IPVERSION +short $RecordType myip.opendns.com @resolver1.opendns.com) #get actual IP
if [ "$SavedIP" != "$ActualIP" ] ; then
wget -qO- $UpdateURL &> /dev/null
if [[ $? -ge 0 ]]; then
SavedIP=$ActualIP
fi
fi
sleep 5s #run every 5 seconds
done
Very simple and neat script that runs every 5 seconds in a loop. Has some advantages over the script above. Should work on most clients, includes a reliable way to get the IP and can be adapted to most dynamic DNS providers.
Configure hostname, record type and update URL, done.
For dynv6 you have to add your token in the URL. If you want to use IPV4, change IPVERSION to 4 and change url to ipv4 version of the URL. Then, simply copy it into a file like ddnsupdate.sh, chmod +x ddnsupdate.sh to make it executable and execute it from a startup script like rc.local: nohup /home/ddnsupdate.sh > /dev/null 2>&1&
There is example of systemd unit
[Unit]
After=miredo.service
[Service]
ExecStart=/usr/local/bin/ddnsupdate.sh
[Install]
WantedBy=default.target
Adapted findings and requirements from this thread into a n updated (and currently working) script. Depends on ns1.google.com ability to return a TXT record for o-o.myaddr.l.google.com. @Beaving did a similar thing with opendns but it did stop working as the "myip service" has been discontinued.
Script
#!/bin/bash
if [ -z "$DYNV6_HOSTNAME" -o -z "$DYNV6_TOKEN" -o -z "$DYNV6_PROTO" ]; then
echo "Usage:"
echo "DYNV6_HOSTNAME=hostname DYNV6_TOKEN=token DYNV6_PROTO=[4|6] $0"
exit 1
fi;
record=$([ "$DYNV6_PROTO" == 4 ] && echo "A" || echo "AAAA")
resolved=$(dig +short $record $DYNV6_HOSTNAME )
actual=$(dig -$DYNV6_PROTO TXT +short o-o.myaddr.l.google.com @ns1.google.com | tr -d '"')
if [ "$resolved" != "$actual" ] ; then
if [ "$DYNV6_PROTO" == "4" ] ; then
curl "http://ipv4.dynv6.com/api/update?hostname=$DYNV6_HOSTNAME&ipv4=$actual&token=$DYNV6_TOKEN"
fi
if [ "$DYNV6_PROTO" == "6" ] ; then
curl "http://ipv6.dynv6.com/api/update?hostname=$DYNV6_HOSTNAME&ipv6=$actual&token=$DYNV6_TOKEN"
fi
fiExample:
DYNV6_HOSTNAME=ramalamadingdong.dynv6.net DYNV6_TOKEN=muchSecureSoSecret DYNV6_PROTO=6 ./update_dns.shPlease note that if you want to update v6 and v4 records, you will have to call the script twice.
like some other people before me realized: when the ipv6 has not changed, then the ipv4 address is not updated even if it has changed. This is due to 27, 32,40 where the script just exits.
I would open an issue for that if it were not only a gist but a full-fledged github repo. In the meantime - could i just suggest to actually put this script into a full-fledged github repo and to encourage the many people linking to their own versions in the comments here to make an effort and merge their diffs into the script so it actually becoms usable for noobs?
Many thanks for considering this!
I would open an issue for that if it were not only a gist but a full-fledged github repo. In the meantime - could i just suggest to actually put this script into a full-fledged github repo and to encourage the many people linking to their own versions in the comments here to make an effort and merge their diffs into the script so it actually becoms usable for noobs?
+1
Nice service!
I use the following systemd timer/service templates to enable the update on my hosts:
# /etc/systemd/system/dyndns@.service
[Unit]
Description=Update dynv6 DNS Entries
[Service]
Type=oneshot
EnvironmentFile=/etc/conf.d/dynv6
ExecStart=/usr/local/bin/dynv6.sh %I.dynv6.net
[Install]
WantedBy=multi-user.target# /etc/systemd/system/dyndns@.timer
[Unit]
Description=Minutely dynv6 update
[Timer]
OnCalendar=minutely
Persistent=true
[Install]
WantedBy=timers.targetthen enable/start:
sudo systemctl enable dyndns@MYHOSTNAME.timer
sudo systemctl start dyndns@MYHOSTNAME.timer
Since the 2019-10-31 update this does no longer work.
There is a new 'zone' parameter. I tried a few things like myhostname.dynv6.net, dynv6.net or empty, but each time I get:
HTTP/1.1 422 Unprocessable Entity
...
Error 1146: Table 'dynv6.hosts' doesn't exist
What should the zone parameter be set to? The api doc just says: 'Your DynV6 zone (mandatory) '
Use the zones ipv4.dynv6.com or ipv6.dynv6.com to enforce a specific IP protocol version.
Bit I also felt
I have the same problem. Since yesterday the update does not work anymore.
The documentation on dynv6.com is weird and confusing. I think what they want to say is, that you should use "ipv4.dynv6.com" as the url to query - not as zone parameter - to enforce ipv4 (ipv4.dynv6.com has only A records).
If you replace the foreign "hostname" parameter by "zone" (as the documentation says) it complains about a missing hostname. The API seems buggy to me...
Same issue here. Also https://dynv6.com/docs/apis currently is borked.
Just got into the dynv6 here and same problem.
Seems to be fixed now, without changing the url.
Sorry, we introduced a bug that has been fixed in the meanwhile.
I have a raspi behind a fritzbox acting as owncloud server. That worked well until Monday morning, when the IPv4 address got updated to
127.0.0.1. I changed the script instead of ipv4=auto in the api update command i use the current external IPv4 address. That fixes the problem.
Any idea why this suddenly happens? (no hardware or software changes/updates)
Same here
You can use address=$(wget -O- "https://api6.ipify.org")
(For ipv4 use https://api.ipify.org)
Or better place it below
if [ -e /usr/bin/curl ]; then
bin="curl -fsS"
elif [ -e /usr/bin/wget ]; then
bin="wget -O-"
else
echo "neither curl nor wget found"
exit 1
fi
To allow using curl or wget, depending on your setup.
if [ -e /usr/bin/curl ]; then
bin="curl -fsS"
elif [ -e /usr/bin/wget ]; then
bin="wget -O-"
else
echo "neither curl nor wget found"
exit 1
fi
address=$($bin "https://api6.ipify.org")
Because we don't want to use crontab to execute the script, we hope to execute the update script through the /etc/ppp/ipv6-up script of the system after establishing the pppoe session. But I found that after my router established pppoe session, the IPv6 address acquisition would be delayed, resulting in the update failure when $address was empty. Therefore, the base script is modified. When it is found that $address is empty, wait for 1 minute and then assign the value again, The loop ends after 30 times.
Note: crontab is not recommended for this script, which will increase the operating system load.
https://gist.github.com/gaopfeng/0afc189bb79b3a6cc239d45af3ec6dbe
# When the address is empty, wait for 1 minute and then assign the value again. The loop ends after 30 times.
i=1
while [ $i -lt 30 ]
do
address=$(ip -6 addr list scope global $device | grep -v " fd" | sed -n 's/.*inet6 \([0-9a-f:]\+\).*/\1/p' | head -n 1)
if [ -z "$address" ]; then
sleep 1m
((i++))
else
break
fi
done
Here's a Python updater for IPv4 that updates to the IP of the PC the script is run on.
-Prerequisites
Python (duh)
requests the python module
pip install requests
Once done with the prerequisites, edit the script so the variables match your zone (hostname) and token.
Then run the script and the IPv4 address for your zone should update every 30 seconds!
I created two versions of the updater; an auto and a manual.
Manual Version:
https://gist.github.com/flid0/8784baff35ec061f00432a9db3525c20
Auto Version:
https://gist.github.com/flid0/91bd9a3e15b61a8fd5797c325b3451a0
DM me on Discord if you have any problems:
flido#0001
Only for IPV6 DDNS
filename: dynv6.sh
path:/usr/bin
chmod +x dynv6.sh
#!/bin/sh -e
GETIPV6="https://ip.ddnspod.com"
hostname="xxxxx.v6.rocks"
token="xxxxxxxxxxxxxxxxxxxxxxxxxx"
file=$HOME/.dynv6.addr6
[ -e $file ] && OldIP=cat $file
echo "[Old IP]:$OldIP"
NewIP=$(curl -6 -s -k $GETIPV6)
echo "[New IP]:$NewIP"
if [ "$OldIP" == "$NewIP" ];then
echo "IP IS SAME,SKIP UPDATE."
exit
fi
#send addresses to dynv6
curl "http://dynv6.com/api/update?hostname=$hostname&ipv6=$NewIP&token=$token"
#curl "http://ipv4.dynv6.com/api/update?hostname=$hostname&ipv4=auto&token=$token"
#save current address
echo $NewIP > $file
[root@C921 ~]# crontab -e
*/1 * * * * dynv6.sh &>/dev/null 2>&1
@andi34
the following script was out of work
address=$(wget -O- "https://api6.ipify.org")
"api64.ipify.org " always returns ipv4 address
@tolshao nope, works fine. Typo on your end? api64 vs api6?
I don't like caching the "old" IP in a file because if it was changed manually on dynv6.com or the update went wrong, the script doesn't notice.
So this variant works very well for me:
#!/bin/sh
HOSTNAME_DYNV6="****.dynv6.net"
TOKEN_DYNV6="****"
IP4ADDR=$(curl -s http://ipecho.net/plain)
IP6ADDR=`ip addr show eth0 | grep 'scope global dynamic' | grep -Po 'inet6 \K[0-9a-fA-F:]+'`
if [ "$IP4ADDR" = "" ]
then
echo "Error: unable to determine IPv4 address" 1>&2
fi
if [ "$IP6ADDR" = "" ]
then
echo "Error: unable to determine IPv6 address" 1>&2
fi
if [ "$IP4ADDR" != "" ]
then
ping $HOSTNAME_DYNV6 -4 -c 1 > null # a little dirty - needed to update dns-cache
IP4ADDR_DYNV6=$(dig $HOSTNAME_DYNV6 A +short)
if [ "$IP4ADDR" != "$IP4ADDR_DYNV6" ]
then
echo "IPv4 adress has changed -> update ..."
curl -s "https://ipv4.dynv6.com/api/update?hostname=$HOSTNAME_DYNV6&token=$TOKEN_DYNV6&ipv4=auto"
echo "---"
fi
fi
if [ "$IP6ADDR" != "" ]
then
ping $HOSTNAME_DYNV6 -6 -c 1 > null # a little dirty - needed to update dns-cache
IP6ADDR_DYNV6=$(dig $HOSTNAME_DYNV6 AAAA +short)
if [ "$IP6ADDR" != "$IP6ADDR_DYNV6" ]
then
echo "IPv6 adress has changed -> update ..."
curl -s "https://ipv6.dynv6.com/api/update?hostname=$HOSTNAME_DYNV6&token=$TOKEN_DYNV6&ipv6prefix=auto"
echo "---"
fi
fi
Hi, im running this on a raspberry Pi behind my ISPs router. Is there a way to use this script to update the external/public IP?