Closed hungnt61h closed 3 years ago
I tried to configure trino like previous version then it works now. These properties are required:
@hungnt61h This new mechanism requires the workers to use the IP address of the co-ordinator in the discovery service uri. It doesn't work if the IP address associated with the co-ordinator changes for some reason.
See the 3rd step under https://trino.io/docs/current/security/internal-communication.html#internal-tls-configuration where it mentions using an IP address for the discover service URI.
cc: @mosabua @dain Do you have more information?
@mosabua Maybe we add the point about using IP address as a NOTE too in https://trino.io/docs/current/security/internal-communication.html#internal-tls-configuration (even though it's mentioned in step 3 of the instructions).
@hashhar My cluster has internal DNS servers, so this discovery.uri
property must still require ip address instead of fqdn?
It does NOT work with fqdn .. IP address usage is required. Do you really think we need to point that out even more in the docs @hashhar ?
@mosabua with the existing wording it doesn't sound like it's required. It seems natural to assume that you can replace IP with FQDN since that's generally true.
This is what we decided in the doc PR with @dain and @electrum .. I can send an update though..
I’m clear for certain now! Thank you for supporting me! 😃
I tried to configure trino like previous version then it works now. These properties are required:
- internal-communication.https.keystore.path
- internal-communication.https.truststore.path
- node.internal-address-source Why this tutorial said that they must be removed?
I also encountered this problem, what are these values?
In fact, I just want to limit the client to use HTTP and force the use of HTTPS plus account passwords to limit its authority, but when I use HTTPS, there is a problem with the internal communication of the cluster. This is embarrassing.
Hi, I'm just beginner in setting up Trino cluster! I deployed Trino (version 360) cluster on vmware machines and got stuck at setting up internal TLS communication. Enabling TLS connection between client and coordinator works fine with me. I am able to log on Trino with my LDAP account and all workers are discovered. (Sensitive fields are ** or xx)
My first configuration here (no internal TLS communication) Coordinator**
Worker
With these configuration, Trino cluster works fine and smoothly. But when I configured internal TLS communication, the problem occured like this:
2021-08-14T17:42:33.887+0700 ERROR Announcer-0 io.airlift.discovery.client.Announcer Service announcement failed after 66.07ms. Next request will happen within 0.00s 2021-08-14T17:42:33.911+0700 ERROR Announcer-0 io.airlift.discovery.client.Announcer Service announcement failed after 17.11ms. Next request will happen within 1.00ms 2021-08-14T17:42:33.927+0700 ERROR Announcer-4 io.airlift.discovery.client.Announcer Service announcement failed after 13.21ms. Next request will happen within 2.00ms 2021-08-14T17:42:33.945+0700 ERROR Announcer-2 io.airlift.discovery.client.Announcer Service announcement failed after 13.28ms. Next request will happen within 4.00ms 2021-08-14T17:42:33.970+0700 ERROR Announcer-0 io.airlift.discovery.client.Announcer Service announcement failed after 16.54ms. Next request will happen within 8.00ms 2021-08-14T17:42:34.005+0700 ERROR Announcer-4 io.airlift.discovery.client.Announcer Service announcement failed after 17.71ms. Next request will happen within 16.00ms 2021-08-14T17:42:34.054+0700 ERROR Announcer-2 io.airlift.discovery.client.Announcer Service announcement failed after 16.39ms. Next request will happen within 32.00ms 2021-08-14T17:42:34.132+0700 ERROR Announcer-0 io.airlift.discovery.client.Announcer Service announcement failed after 14.19ms. Next request will happen within 64.00ms 2021-08-14T17:42:34.275+0700 ERROR Announcer-4 io.airlift.discovery.client.Announcer Service announcement failed after 14.14ms. Next request will happen within 128.00ms 2021-08-14T17:42:34.542+0700 ERROR Announcer-2 io.airlift.discovery.client.Announcer Service announcement failed after 9.37ms. Next request will happen within 256.00ms
with my configuration in
Coordinator
Worker
I followed this tutorial. Can anyone help me figure out what I missed or what's wrong, etc. Thank you guys in advance 😊