You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello, I can see that fluent-bit (v0.11.2) constantly originates new tcp sessions to fluentd server.
For about one day it creates more than 2000 sessions:
[SERVICE]
Flush 1
Daemon Off
Log_Level debug
Log_File /fluent-bit/log/fluent-bit.log
Parsers_File parsers.conf
[INPUT]
Name tail
Path /mnt/*.log
Exclude_Path /mnt/fluentbit*
DB /tmp/files.db
Parser docker
Tag kube.*
Mem_Buf_Limit 5MB
[FILTER]
Name kubernetes
Match kube.*
[OUTPUT]
Match *
Name forward
Host 192.168.242.110
Port 24224
How can I eliminate such situation?
The text was updated successfully, but these errors were encountered:
In recent changes, the function call to release the upstream connection
when the data is flushed was commented, for hence the connection is leaked.
This patch fix the problem releasing the connection properly.
Signed-off-by: Eduardo Silva <[email protected]>
Hello, I can see that fluent-bit (v0.11.2) constantly originates new tcp sessions to fluentd server.
For about one day it creates more than 2000 sessions:
root@fluentbit-4zg6h:/# netstat -anpet |grep 24224 |wc -l
2186
It works in Kubernetes, config is like this:
How can I eliminate such situation?
The text was updated successfully, but these errors were encountered: