When making a change to an nginx web server running on Ubuntu, I tested the nginx configuration and received the following warning:
nginx: [warn] “ssl_stapling” ignored, host not found in OCSP responder “ocsp.int-x3.letsencrypt.org/ ”
I’d set up the web server a few months ago and had been updating the nginx configuration without seeing this warning until yesterday.
I started poking around and tested OCSP using openssl.
I ran the following command:
sudo openssl ocsp -issuer chain.pem -cert tm.pem -text -url http://ocsp.int-x3.letsencrypt.org/
with the following results
OCSP Request Data:
Version: 1 (0x0)
Requestor List:
Certificate ID:
Hash Algorithm: sha1
Issuer Name Hash: 7EE66AE7729AB3FCF8A220646C16A12D6071085D
Issuer Key Hash: A84A6A63047DDDBAE6D139B7A64565EFF3A8ECA1
Serial Number: 03BD1754FBA80442EFE1CE2735DC96C598CF
Request Extensions:
OCSP Nonce:
0410DC5C7354CF0C9838F66506F58A081AE3
Error connecting BIO
Error querying OCSP responder
140605810423448:error:2006A066:BIO routines:BIO_get_host_ip:bad hostname lookup:b_sock.c:146:host=ocsp.int-x3.letsencrypt.org
The website seems to work fine.
I can’t find much info about the error. I don’t know what changed.
Thanks.
cpu
April 28, 2017, 12:20pm
2
Hi @turnipcyberveggie ,
I’m not sure what the cause of your Nginx problem is, but I can help you fix your openssl ocsp
test command. You’re missing a few arguments:
-no_nonce
- we don’t use OCSP nonces
-VAfile chain.pem
- for validating the response signature
-header Host ocsp.int-x3.letsencrypt.org
- the OpenSSL ocsp utility doesn’t send an HTTP host header, causing issues with our CDN.
So, all told it should look like:
openssl ocsp -no_nonce -issuer chain.pem -cert tm.pem -VAfile chain.pem -text -url http://ocsp.int-x3.letsencrypt.org/ -header Host ocsp.int-x3.letsencrypt.org
Does that work for you?
1 Like
@cpu : Same results. (And thanks for the help)
cpu
April 28, 2017, 12:30pm
4
Is the DNS on your test host working? What does dig ocsp.int-x3.letsencrypt.org
look like?
Hm… it would appear not:
; <<>> DiG 9.10.3-P4-Ubuntu <<>> ocsp.int-x3.letsencrypt.org
;; global options: +cmd
;; connection timed out; no servers could be reached
cpu
April 28, 2017, 12:35pm
6
Seems like the root of the problem!
Are you able to diagnose this yourself from here out? I would probably try replacing the upstream DNS servers with a well known public resolver as a first step (assuming you aren’t using one already). If that still fails it would probably point to a network issue.
I got it. I’ll track it down from here. I don’t think its a Lets Encrypt issue. Thanks for the info @cpu .
1 Like
cpu
April 28, 2017, 12:38pm
8
Happy to help! Good luck!
To wrap things up for anyone who ends up here: I rebooted my server and all my problems went away. This was the 4th-ish reboot since the problem started so I’m not really clear what happened. Perhaps it was an issue with the host’s DNS.
system
Closed
May 28, 2017, 2:25pm
10
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.