A little background; I work as a consulting/helpdesk errand boy with noob-starter knowledge on linux (RHEL mostly). I often get thrown in a new customer environment with little or no documentation or the guy/gal who knows is not there. Often there are half a dozen servers named CentOS-dev0x or similar. Typical tasks are just checking why LDAP-sync doesn't work against Windows servers or to confirm that server has proper network connectivity and proper mount points work as they should, restarting services occasionally. I usually get root-equivalent admin access on these servers, but I would like to avoid installing new software on them if built-in tools are available. For scripting purpose it would be nice to use existing commands, one script per linux distro. Also, not every developer-box have internet connection.
The thing; what would be a good set of commands to run on these servers that would help me identify what the server is for and is it currently in use? I was planning to make a small script that would run the commands and vomit them as a text file, so I would get simple documentation at the same time.
(physical or virtual? CPU, memory, disks?) Processes? running services/daemons? Cronjobs or other schedulers? Disk mounts? open network ports, ssh tunnels etc? current traffic? Logs of recent happenings?
What are your steps when connecting to an uncharted server to get to know it?
[link][10 comments]
