Old conf not reading the new

I have a telégraf running in a Windows server 2022
I had a 1.23 versión
And in my conf had a kafka output with a broker dns
Server1292 but It was an error connect to dns and i need to use ip broker instead
I delete all the files and install new telegraf 1.30 with new conf broker (ip instead dns)
Still i have and error with dns. Is like still reading the old configuration, cache problem?
Some hace the same error?

Do a DNS look up on your system and see if DNS is still showing the old hostname. This sounds like a DNS issue if it is the case.

thanks for anwer

if i do this in a server with no problem and similar telegraf conf:
C:>nslookup server1292
Server: Bancadc01.lu

server with the problem:

C:>nslookup server1292
DNS request timed out.
timeout was 2 seconds.
Server: UnKnown
telegraf error:

! [agent] Error writing to outputs.kafka: kafka: Failed to produce message to topic m-wintel: dial tcp: lookup server1292 getaddrinfow: This is usually a temporary error during hostname resolution and means that the local server did not receive a response from an authoritative server.

line for telegraf conf: output kafka
brokers = [“”]

Thanks, that looks like the DNS is not configured on that system correctly or the server itself is down or not working. There isn’t really anything Telegraf can do about a bad or down DNS server.

thank you! then is not a telegraf issue

I didnt understand that If i using a ip like brokers = [“”] telegraf turn me an error with server dns name lookup server1292 for this reason i thought it was a telegraf issue

1 Like