New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Log message for 'Connection reset by peer' could be improved #5183
Comments
We're also seeing a lot of these. Using |
Where we can find more info about this and all other errors, what is causing them and how to get more info about them? |
I've checked on master and I see that the issue is already solved. Now it looks like this: 2020.04.16 22:31:17.903683 [ 4920 ] {} ServerErrorHandler: Code: 210, e.displayText() = DB::NetException: Connection reset by peer, while reading from socket ([::ffff:127.0.0.1]:35372), Stack trace (when copying this message, always include the lines below):
|
Two changes:
|
But it looks like you tested some other scenario (it was already working properly before).
|
Steps to reproduce: # open a http connection
exec 3<>/dev/tcp/127.0.0.1/8123
# start sending request
echo "GET / HTTP/1.0" >&3
# now kill the connection (to mimic connection reset)
sudo ss -K dst 127.0.0.1 dport = 8123 |
@filimonov can you please tell me what does
|
A large number of the following logs are printed on our logs, what are the solutions to solve this
|
Seeing a ton of these messages in all our clusters (a few every second per node). Would it make sense to change these to trace log level instead of error? (also running
|
Same here tons of these messages, running 20.6.3. what can be possible reason for it??? and can we move it to trace logs. |
The error message was improved in the newer versions (most probably starting from in 21.3), and now it includes the ip.
|
Still happening, see examples in #23858 |
We see a lot of messages like
[ 23 ] {} <Error> ServerErrorHandler: Poco::Exception. Code: 1000, e.code() = 104, e.displayText() = Connection reset by peer (version 19.5.3.1)
. It is almost impossible to trouble shoot them now.It would be great if remote port:ip will be logged with this message.
The text was updated successfully, but these errors were encountered: