Difference between revisions of "Technical infrastructure checklist"

From FreekiWiki
Jump to navigation Jump to search
(Mainly moving stuff that I do to a separate section. Also added some helpful hints.)
Line 1: Line 1:
 
Resources on this list should be checked regularly to see that they are in working order.
 
Resources on this list should be checked regularly to see that they are in working order.
 +
= During "walkthrough" =
 
== Services ==
 
== Services ==
 
''italicized text is checked on a per room basis''
 
''italicized text is checked on a per room basis''
Line 5: Line 6:
 
* mailing list moderation
 
* mailing list moderation
 
** reply to subscription requests to determine sentience
 
** reply to subscription requests to determine sentience
* rootmail cleanup/checkup
+
* Check each server up and up-to-date? http://monitor/nagios2 look at tactical overview. A report is also emailed to asswatch every night at midnight. If you want to use the command line, there are three scripts: list_problems (lists everything that has a problem, and what that problem is), down (lists down servers), and needs_upgrading (lists servers that need to be "aptitude dist-upgrade"ed). These scripts live on monitor.
* Check each server
 
** listed in purpose
 
** up and up-to-date? http://monitor/nagios2 look at tactical overview
 
** maintenance check
 
** backups working (check lucien:/bum/HOSTNAME/var/log/syslog for current date)
 
 
* web services administration (wiki, todo, lessons, area, dev/svn, dev/projects, web, web/mail, www, others...)
 
* web services administration (wiki, todo, lessons, area, dev/svn, dev/projects, web, web/mail, www, others...)
** integrate this with nagios?
+
** integrate this with nagios? [[User:Ryan52|Ryan]] will look into what can and needs to be integrated.
 +
* check if there are any open security announcement tickets. Search for open RT tickets created by ass-security@our_domain. If you like the command line, then you can use this (mainly self explanatory, just run it) script: ./ass/scripts/security/show_security
  
 
=== Servers & Security ===
 
=== Servers & Security ===
Line 26: Line 23:
 
** Balanced load
 
** Balanced load
 
** Battery installation within 3 years
 
** Battery installation within 3 years
* internet is connected and working
 
  
 
=== Phones ===
 
=== Phones ===
Line 33: Line 29:
  
 
=== Printers ===
 
=== Printers ===
 +
[[User:Ryan52|Ryan]] will integrate the "on and ready" and "stalled jobs" into nagios.
 
* Are all printers on and ready? http://scribble:631
 
* Are all printers on and ready? http://scribble:631
 
** are there stalled jobs?
 
** are there stalled jobs?
Line 71: Line 68:
 
** terminals and images.
 
** terminals and images.
  
 
+
= Taken care of by somebody/something else =
 
+
* watch all of the asswatch scripts, and makes sure that they are working. ([[User:Ryan52|Ryan]] does this)
 +
* rootmail cleanup/checkup ([[User:Ryan52|Ryan]])
 +
* check each server is listed in purpose (asswatch)
 +
* backups working on each server (asswatch)
 +
* check each server with maintenance notes and script, check rootmail is working ([[User:Ryan52|Ryan]] does this every so often, probably doesn't need to be done often)
  
 
[[Category:Technocrats]]
 
[[Category:Technocrats]]

Revision as of 18:22, 20 November 2008

Resources on this list should be checked regularly to see that they are in working order.

During "walkthrough"

Services

italicized text is checked on a per room basis

Software and Systems

  • mailing list moderation
    • reply to subscription requests to determine sentience
  • Check each server up and up-to-date? http://monitor/nagios2 look at tactical overview. A report is also emailed to asswatch every night at midnight. If you want to use the command line, there are three scripts: list_problems (lists everything that has a problem, and what that problem is), down (lists down servers), and needs_upgrading (lists servers that need to be "aptitude dist-upgrade"ed). These scripts live on monitor.
  • web services administration (wiki, todo, lessons, area, dev/svn, dev/projects, web, web/mail, www, others...)
    • integrate this with nagios? Ryan will look into what can and needs to be integrated.
  • check if there are any open security announcement tickets. Search for open RT tickets created by ass-security@our_domain. If you like the command line, then you can use this (mainly self explanatory, just run it) script: ./ass/scripts/security/show_security

Servers & Security

  • Security cameras
    • Software set up correctly and running.
    • pointing in correct direction
  • visual inspection
    • look for things like bad fans on servers
    • dust/vaccuum servers, switches
    • look dangerous cable tangles, stress/tension on punched-down cables
  • UPS check
    • Tests successfully
    • Balanced load
    • Battery installation within 3 years

Phones

  • Outgoing Phone Message
  • All listed phones working

Printers

Ryan will integrate the "on and ready" and "stalled jobs" into nagios.

  • Are all printers on and ready? http://scribble:631
    • are there stalled jobs?
    • are there paper jams (walk around)
    • are they free of non-standard paper

hubs/switches/networking

  • check network ports around room
  • check for bad fans

Rooms

Meeting Room

  • Projector and computer
  • Network hub (plugged into wall correctly)
  • Phone (plugged in?)

Classroom

  • All terminals
  • Printer
  • networking around room
  • Wireless access point

Reception

  • Printer (toner low? paper?)
  • Terminals and monitors functioning well?

upstairs and downstairs office

  • terminals
  • networking
  • phones

Production

  • Advanced testing
    • lots of networking
    • testers functioning
  • build (+mac +laptop +enterprise)
    • networking
    • terminals and images.

Taken care of by somebody/something else

  • watch all of the asswatch scripts, and makes sure that they are working. (Ryan does this)
  • rootmail cleanup/checkup (Ryan)
  • check each server is listed in purpose (asswatch)
  • backups working on each server (asswatch)
  • check each server with maintenance notes and script, check rootmail is working (Ryan does this every so often, probably doesn't need to be done often)