search

mesosphere / dcos-kubernetes-quickstart

Quickstart guide for Kubernetes on DC/OS
https://mesosphere.com/kubernetes/
Apache License 2.0
167 stars 54 forks source link

Error in Kubernetes-proxy log #75

Closed redpine42 closed 6 years ago

redpine42 commented 6 years ago

Running DCOS 1.11 kubernetes 1.0.1-1.9.6. After launching kubernetes I see the following error in the kubernetes-proxy log. Startup is looking for a directory /var/log/nginx to exist. This directory does not exist in a generic Centos 7 install. Creating this directory on the nodes resolves the error at next launch.

I0323 15:49:06.166402 7 executor.cpp:651] Forked command at 12 nginx: [alert] could not open error log file: open() "/var/log/nginx/error.log" failed (2: No such file or directory)

pires commented 6 years ago

I believe this is harmless, right @bmcstdio?

redpine42 commented 6 years ago

It doesn't cause any problems for me, but still an error.

pires commented 6 years ago

Not quite right. It just means the logfile doesn't exist and so it gets created.

redpine42 commented 6 years ago

Directory is missing and the log file does not get created, nor the directory. I tested this.

So unless the directory is created prior to startup the log file does not get created.

bmcustodio commented 6 years ago

The error is harmless. It happens because, when starting and before reading the configuration file, NGINX tries to open that file because it is the default path it was compiled with. After reading the configuration NGINX uses the log file specified in the error_log directive, which in this case is /dev/stderr. That means you will see every subsequent error that may happen in stderr logs for the kubernetes-proxy task. There is no need to create any "physical" file or directory.

I am closing this issue as there's nothing else to see here.