[GH-ISSUE #244] Huge delay in sending new file content from harvester to server #197

New issue

Open

opened 2026-02-26 05:31:34 +03:00 by kerem · 0 comments

kerem commented

2026-02-26 05:31:34 +03:00

Owner

Originally created by @MichalBiel on GitHub (Jul 3, 2020).
Original GitHub issue: https://github.com/NarrativeScience-old/log.io/issues/244

I have an issue that make log.io unusable. After fresh reboot of server and haresters everything seems ok, but after about an hour, there is a 30 min delay in delivering messages ie. logs created at 12:30 are delivered to the server at 13:00. I have 4 harvesters watching application logs producing about 500-1000 lines per file per minute. All application servers and log.io server are located on the same host (different VM's) so network bandwith is not a problem here. In my opinion it's harvester side issue, after restarting harvester it works fine for couple of minutes and then the delay grows and grows. Could you please tell me how to find the bottleneck ?
My server is installed on CentOS, harvesters are working on Windows Server 2012. Harvester config file :
{
"messageServer": {
"host": "xx.xx.xx",
"port": 6689
},
"inputs": [
{
"source": "XXX",
"stream": "XXXX",
"config": {
"path": "D:\xxx\xxx\logs\SystemOut.log"
}
}
]
}

Originally created by @MichalBiel on GitHub (Jul 3, 2020). Original GitHub issue: https://github.com/NarrativeScience-old/log.io/issues/244 I have an issue that make log.io unusable. After fresh reboot of server and haresters everything seems ok, but after about an hour, there is a 30 min delay in delivering messages ie. logs created at 12:30 are delivered to the server at 13:00. I have 4 harvesters watching application logs producing about 500-1000 lines per file per minute. All application servers and log.io server are located on the same host (different VM's) so network bandwith is not a problem here. In my opinion it's harvester side issue, after restarting harvester it works fine for couple of minutes and then the delay grows and grows. Could you please tell me how to find the bottleneck ? My server is installed on CentOS, harvesters are working on Windows Server 2012. Harvester config file : { "messageServer": { "host": "xx.xx.xx", "port": 6689 }, "inputs": [ { "source": "XXX", "stream": "XXXX", "config": { "path": "D:\\xxx\\xxx\\logs\\SystemOut.log" } } ] }