Issues with some cloudflare IP ranges in Portuguese ISPs

Hello,

I’m currently facing some issues with some services when proxied by Cloudflare.
Since a few hours ago, I started recieving complains about some of my websites/services not loading, after some troubleshooting I found out the issue is caused when the client is using google’s dns (8.8.8.8) to resolve my domain records.
In the following printscreen is shown the responses that I get from resolving one of my subdomains from 8.8.8.8 and 1.1.1.1


As you can see, if I ping the 188 range, the ping fails but to 104 or 172 it works normally.

I’ve also tried a traceroute and seems that there is no route to the 188 range.

I’m from Portugal and most of my users are also from Portugal, here we have 3 ISPs (MEO, NOS and Vodafone).
I’ve tested on my network (MEO) but also asked some friends from NOS and Vodafone and the issue is the same, when using google’s dns server and getting the ip from the 188 range, there is route to that IP address, failing to load the service.
I’ve also tested on my VPSs (from France and UK) and both of them seem to have routes to that IP address.
Since it is affecting all the ISPs here in Portugal and CF has POPs here in Lisbon, I would assume that is some kind of routing issue between those…

2 Likes

first complain about my service timing out was made at (April 17th) 9:30pm UTC+1 and at (April 18th) 1:30am UTC+1 the issue seems to be fixed

Also worth to mention, for what some users of my app said, that this seems the be a recurrent problem, but never affected me, since I use CF dns resolver…

having the same issue with the networks you mentioned, also from Portugal and MEO client…
I thought it was the first time but some users of my applications that I serve from Cloudflare, just told me that is already a recurrent problem, but since I don’t use google’s dns resolver, seems like I’m not affected.1.1.1.1 (and most of the default ISPs dns servers) return a different ip range thats works fine

also just opened a topic about this issue (I thought it was only happening here in PT), maybe should have searched better first…

1 Like

the following ips are currently not responding now from vodafone portugal…
188.114.96.0
188.114.97.0
and these just went dead now too
188.114.96.3
188.114.97.3
and
188.114.96.12
188.114.97.12
failing too…
and now also failing
188.114.96.7
188.114.97.7

mtr --report 188.114.96.0
Start: 2022-04-23T16:01:19+0100
HOST: mbd-zenbook Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 0.0% 10 8.9 5.9 1.2 30.8 9.3
2.|-- 3.128.189.46.rev.vodafone 0.0% 10 3.6 5.9 2.7 24.6 6.6
3.|-- 24.50.174.83.rev.vodafone 0.0% 10 3.7 7.4 3.7 29.1 7.9
4.|-- ??? 100.0 10 0.0 0.0 0.0 0.0 0.0

mtr --report 188.114.97.0
Start: 2022-04-23T16:01:38+0100
HOST: mbd-zenbook Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 0.0% 10 1.2 6.1 1.1 35.2 11.0
2.|-- 3.128.189.46.rev.vodafone 0.0% 10 56.7 11.2 3.5 56.7 17.0
3.|-- 24.50.174.83.rev.vodafone 0.0% 10 13.6 15.6 4.0 67.1 20.5
4.|-- ??? 100.0 10 0.0 0.0 0.0 0.0 0.0

mtr --report 188.114.96.3
Start: 2022-04-23T15:51:39+0000
HOST: raspberrypi Loss% Snt Last Avg Best Wrst StDev
1.|-- 192.168.1.1 0.0% 10 0.8 0.9 0.8 1.0 0.1
2.|-- 3.128.189.46.rev.vodafone 0.0% 10 2.9 2.9 2.8 3.1 0.1
3.|-- 24.50.174.83.rev.vodafone 0.0% 10 3.4 3.3 2.9 3.5 0.2
4.|-- ??? 100.0 10 0.0 0.0 0.0 0.0 0.0

mtr --report 188.114.97.3
Start: 2022-04-23T15:51:58+0000
HOST: raspberrypi Loss% Snt Last Avg Best Wrst StDev
1.|-- 192.168.1.1 0.0% 10 0.8 0.9 0.8 1.0 0.1
2.|-- 3.128.189.46.rev.vodafone 0.0% 10 2.9 2.9 2.7 3.0 0.1
3.|-- 24.50.174.83.rev.vodafone 0.0% 10 95.6 67.0 3.4 171.5 66.4
4.|-- ??? 100.0 10 0.0 0.0 0.0 0.0 0.0

mtr --report 188.114.96.12
Start: 2022-04-23T16:57:47+0000
HOST: raspberrypi Loss% Snt Last Avg Best Wrst StDev
1.|-- 192.168.1.1 0.0% 10 0.9 0.9 0.8 1.0 0.1
2.|-- 3.128.189.46.rev.vodafone 0.0% 10 2.6 2.8 2.5 3.2 0.2
3.|-- 24.50.174.83.rev.vodafone 0.0% 10 3.1 3.4 3.1 4.0 0.2
4.|-- ??? 100.0 10 0.0 0.0 0.0 0.0 0.0

mtr --report 188.114.97.12
Start: 2022-04-23T17:03:51+0000
HOST: raspberrypi Loss% Snt Last Avg Best Wrst StDev
1.|-- 192.168.1.1 0.0% 10 0.9 0.9 0.8 1.1 0.1
2.|-- 3.128.189.46.rev.vodafone 0.0% 10 2.9 2.9 2.3 3.3 0.3
3.|-- 24.50.174.83.rev.vodafone 0.0% 10 3.6 10.4 3.4 72.2 21.7
4.|-- ??? 100.0 10 0.0 0.0 0.0 0.0 0.0

mtr --report 188.114.96.7
Start: 2022-04-23T17:16:43+0000
HOST: raspberrypi Loss% Snt Last Avg Best Wrst StDev
1.|-- 192.168.1.1 0.0% 10 1.0 0.9 0.7 1.0 0.1
2.|-- 3.128.189.46.rev.vodafone 0.0% 10 2.9 3.0 2.9 3.2 0.1
3.|-- 24.50.174.83.rev.vodafone 0.0% 10 3.4 3.4 3.2 3.7 0.1
4.|-- ??? 100.0 10 0.0 0.0 0.0 0.0 0.0

mtr --report 188.114.97.7
Start: 2022-04-23T17:17:06+0000
HOST: raspberrypi Loss% Snt Last Avg Best Wrst StDev
1.|-- 192.168.1.1 0.0% 10 0.8 0.9 0.8 1.1 0.1
2.|-- 3.128.189.46.rev.vodafone 0.0% 10 2.9 2.9 2.8 3.1 0.1
3.|-- 24.50.174.83.rev.vodafone 0.0% 10 3.3 8.3 3.2 52.6 15.6
4.|-- ??? 100.0 10 0.0 0.0 0.0 0.0 0.0

Right after I wrote this post, I made a script that resolves my most important DNS records in multiple DNS providers (1.1.1.1, 8.8.8.8 and MEO resolver). After resolving the records it pings those 4 times and sends a message on discord if something fails.
It is been running since then and so far it didn’t happen again for me (MEO client)…

without resolving any domains, because different domains can resolve to different ips, can you reach 188.114.96.0
188.114.97.0
188.114.96.3
188.114.97.3
188.114.96.12
188.114.97.12
188.114.96.7
188.114.97.7
?


Well… not anymore…
Services down for some users…
188.114.96.0/24
188.114.97.0/24
Both are unreachable…

i can reach
mtr --report 188.114.97.13
Start: 2022-04-23T17:12:06+0000
HOST: raspberrypi Loss% Snt Last Avg Best Wrst StDev
1.|-- 192.168.1.1 10.0% 10 1.0 0.9 0.8 1.0 0.1
2.|-- 3.128.189.46.rev.vodafone 30.0% 10 3.0 3.0 2.8 3.1 0.1
3.|-- 24.50.174.83.rev.vodafone 0.0% 10 3.4 3.5 3.1 4.4 0.3
4.|-- ??? 100.0 10 0.0 0.0 0.0 0.0 0.0
5.|-- 188.114.97.13 0.0% 10 3.5 3.3 3.0 3.5 0.2

so I wouldn’t say /24 is failing but /32 on the 188.114.96.0
188.114.97.0
188.114.96.3
188.114.97.3
188.114.96.12
188.114.97.12
188.114.96.7
188.114.97.7
list

ips have started to come back to life for vodafone portugal network. Not all @ the same time, but they are returning to service

and starting to fail again…

So far they are still working for me, currently testing those IPs you mentioned yesterday.
Also since yesterday that I removed Cloudflare proxy from some of my services, since some users couldn’t access them because routes being broken… I’ve been using Cloudflare to proxy my services, for a few years and never had this issues…
Since it probably doesn’t affect only us (from 2 different ISPs), Cloudflare should take a look at this issue…

today (as of now) I’m only not able to reach 188.114.96.0 and 188.114.97.0
and now (Apr 24 20:56:23) also
188.114.97.3
188.114.96.3

and there we go again… 1 week later since the start of this post (sunday @ 9pm UTC+1), 188.114.96.3 and 188.114.97.3 are both unreachable…
last time was 188.114.96.12 and 188.114.97.12, which for now are still up

Update:
(Apr 24 10:46pm UTC+1) 188.114.96.3 is back for MEO
(Apr 24 10:54pm UTC+1) 188.114.97.3 is back for MEO

and from Vodafone Portugal they are starting to respond again…

here we go again…
188.114.97.0 and 188.114.96.0 stopped responding to Vodafone Portugal

you can add 188.114.97.15 to the list if ips that fail from vodafone when the service interruption occurs

This topic was automatically closed 15 days after the last reply. New replies are no longer allowed.