Skip to content

graylog-labs/apollo

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

30 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

The Apollo collector allows our support customers to easily submit bundles of information to the Graylog team. This way we can streamline support and dig deep into metrics without having to request more and more information via email or phone.

Read before you run Apollo

  • The machine you run Apollo from must be able to reach the REST APIs of all graylog-server instances in your Graylog cluster. If you are unsure we recommend not to run it from your workstation but from the machine your graylog-server master is running on.
  • You must run Apollo with a Graylog admin user. This can either be the built-in administrator user or a custom user with the administrator ruleset applied. The bundle extraction will fail if you run it with a reader user.
  • The information collected is usually not containing any sensitive information and never any messages you sent into Graylog. We will however be able to see stream and extractor names. Handling of the bundles falls under the support contract conditions and will thus never be shared and stored securely. You can look at the source code or unzip the generated bundle if you want to make sure.
  • You only need to provide the REST API URL of one graylog-server instance. Apollo will auto-discover the other graylog-server instances in the cluster.
  • Future version will allow an automatic transmission of the bundle to us.
  • Get the current latest release of apollo for your OS

Using Apollo on Linux

  • Download the binary: curl -OL https://github.com/Graylog2/apollo/releases/download/0.5/apollo_linux_386
  • Make the binary executable: chmod +x apollo_linux_386
  • Run the binary (replace username, password and hostname of graylog-server instance): ./apollo_linux_386 -user hans -password secret -url http://graylog.example.org:9000/api
  • Send us the generated .ZIP bundle file via email. (Located in same folder from where you executed Apollo. Called something like graylog_apollo_bundle-2015-09-23T22-05-54.zip)

Using Apollo on OSX

  • Download the binary: curl -OL https://github.com/Graylog2/apollo/releases/download/0.5/apollo_osx_386
  • Make the binary executable: chmod +x apollo_osx_386
  • Run the binary (replace username, password and hostname of graylog-server instance): ./apollo_osx_386 -user hans -password secret -url http://graylog.example.org:9000/api
  • Send us the generated .ZIP bundle file via email. (Located in same folder from where you executed Apollo. Called something like graylog_apollo_bundle-2015-09-23T22-05-54.zip)

Using Apollo on Microsoft Windows

  • Download the binary from the Releases.
  • Run the binary (replace username, password and hostname of graylog-server instance): c:\apollo_windows_386.exe -user hans -password secret -url http://graylog.example.org:9000/api
  • Send us the generated .ZIP bundle file via email. (Located in same folder from where you executed Apollo. Called something like graylog_apollo_bundle-2015-09-23T22-05-54.zip)

FAQ

The run of apollo and the output looks similar to the following:

./apollo_linux_386 -user admin -password <password> -url http://graylog-web-interface:9000
2017/10/20 08:39:51.515719 apollo.go:70: Starting up.
2017/10/20 08:39:51.517787 apollo.go:209: Successfully read 1988 bytes [system/cluster/nodes].
2017/10/20 08:39:51.517826 apollo.go:247: invalid character '<' looking for beginning of value

Make sure to use a graylog-server REST API URL (Default to graylog-web-interface/api) and not the URL of a graylog-web-interface as -url parameter.

I am getting a HTTP 404 and Apollo exits

Make sure to use a graylog-server REST API URL (Default to graylog-web-interface/api) and not the URL of a graylog-web-interface as -url parameter.

I am getting a HTTP 401 and Apollo exits

Make sure to use a user with administrator permissions in the -user and -password parameters of Apollo.

What data is included in the bundle?

The information collected is usually not containing any sensitive information and never any messages stored in Graylog. We will however be able to see stream and extractor names. Handling of the bundles falls under the support contract conditions and will thus never be shared and stored securely. You can look at the source code or unzip the generated bundle if you want to make sure.

Who will be able to see the bundles?

Only Graylog employees will be able to see data from the bundles.