Acme-v02.api.letsencrypt.org down

Help
1

requests.exceptions.ReadTimeout: HTTPSConnectionPool(host='acme-v02.api.letsencrypt.org', port=443): Read timed out. (read timeout=45)

seems down?

2

Step 1: Ensure DNS is resolving the IP correctly:

  • nslookup acme-v02.api.letsencrypt.org
  • dig acme-v02.api.letsencrypt.org

Step 2: trace to that IP using HTTPS port

  • traceroute -T -p 443 acme-v02.api.letsencrypt.org
4 Likes
3

I can confirm that this is not resolving properly from residential Rogers (Ottawa, ON, Canada):

  1     1 ms    <1 ms    <1 ms  [redacted for privacy]
  2    19 ms    12 ms    17 ms  7908b9.watebrain.com [149.115.69.129]
  3    21 ms    17 ms    18 ms  38.147.246.134
  4    25 ms    25 ms     *     38.74.42.137
  5     *        *        *     Request timed out.
(etc etc with timeout)

Nor is it routable from thje datacenter my server is on (Montreal, QC, Canada):

root@git:~# traceroute -T -p 443 acme-v02.api.letsencrypt.org
traceroute to acme-v02.api.letsencrypt.org (172.65.32.248), 30 hops max, 60 byte packets
 1  38.143.59.193 (38.143.59.193)  0.521 ms  0.458 ms *
 2  * 198-98-100-13.beanfield.net (198.98.100.13)  1.201 ms *
 3  be3926.ccr22.ymq01.atlas.cogentco.com (154.54.45.45)  0.903 ms * *
 4  * * *
[etc, etc]

Hurricane Electric routes it eventually, but it took 45s which is very long for a plain traceroute from an internet backbone.

4

I don't have any problems connecting from my residence or an AWS US east coast center.

And, this site shows many world-wide locations are fine: Check website performance and response : Check host - online website monitoring

I'm not discounting certain locales might be having problems but this doesn't look like full outage at LE center. If it is an LE outage at all.

These are some other tests to run to help narrow your connection problem

curl -i -m10 https://acme-v02.api.letsencrypt.org/directory
curl -i -m10 https://acme-staging-v02.api.letsencrypt.org/directory
curl -I https://google.com
curl -4 https://ifconfig.io
curl -6 https://ifconfig.io

The -4 will fail if you don't have IPv4. -6 if no IPv6

4 Likes
5

Its likely not a LE outage but rather one of their transit providers, is my guess.

6

Follow up: on my end this is routing again!

7

We have seen many problems reported like the first post over the years. Most of the time, by far, it is a problem with all outbound requests from that location. @rg305 suggested a first round of commands that often point to the problem. And, the commands I suggested help narrow down the problem if those don't surface the issue.

2 Likes
8

Now it seems fine on my side.
I’m able to resolve the DNS, and the traceroute works fine.
Strange, because yesterday it just ran into timeouts for no apparent reason — and every “is it down” website confirmed it.

Thanks a lot! :slightly_smiling_face:

2 Likes