Have you ever had a manager or department head who is a micromanager?

In the context of IT this often means being asked to provide documention accounting for how you spend your time.
I mean as a Sysadmin there is a fair amount of babysitting that goes on.

To name a few:

• No suspicious sign-ins?
• VPN Access is the regular expected users
• No suspicious security group changes?
• No unusual activity reported in sensitive file shares/folders?
• Snare Quite?
• Symantec quite?
• AlienVault quite?
• Application servers ping-able?
  • i.e ERP is accessible?

But what about the things that do require manual attention?

• Account creation/deletion
  • The power of macros is lovely but it would be irresponsible to distribute a macro enabled doc to HR. However you can make one for yourself and run it on the forms HR sends which amounts to the following process: table on Microsoft Word => macro => csv => powershell => ad user created/deleted.
• Setting up printers
  • Map the printers LOCALLY with group policy. But of course, we can only win so much.
    • User: Why doesn’t the printer doesn’t show up? I NEEED to printer this report or else the 4 Horseman Wrath, Pestilence, Sloth, And Death will be released (on me and therefore you on you because gravity = -1/2gt^2).
    • IT: Which printer?
    • User: The one in my office……?
    • IT: Did you get a new printer?
    • User: Yea I got for my office. All the other printers show up automatically but not this one. Oh, and I don’t want everyone to see this printer just me on my computer?
    • IT to self: He expects me to manage his personal printer, great.
    • IT to user: How about now? (After mapping the printer)
    • User: BTW everyone in my department is getting a personal printer in their office, so you might want to check in with them.
    • IT: ………………………………………………………………………………
  • So, you see we can only win so much.
  • Encountering this kind of this is what had led me to design organizational units around offices, for instance that we organize the network connections in our server room by office not user. With that in mind we have the following hierarchy for OU’s: domain => branch => department=> office. This way even if the numbers of users change or people start doubling up in offices, my work load remains the same.
• White listing emails external emails.
  • I haven’t really found a real way to automate this. It really is a case by case basis and most of The time I get asked this it is a legitamete request.
• Whitelisting Software Symantec flags
  • This is a legitimate request, but I usually come across it in illegitimate situations. And of course by illegitimate I mean I get asked to whitelist software Symantec has not flagged… Let that sink in.
• Firewall tweaks
  • Well you know a vendor that wants to have a login portal on a random port 13337 on an unencrypted(http) connection, the justification being a hacker would only ever look at ports 80 and 443. Security through obfuscation ……
  • Let’s not forgot that the pressure from other departments to just make it work (We make a lot of money with them so don’t rock the boat IT).

While the lists are not exhaustive you get the general idea.
There is also a small dichotomy worth mentioning(that a micro manager wouldn’t consider), it goes like this:
System running smooth => babysitting, system not running smooth => everyone not in IT dept is furious… at the IT dept.
So anyway back to micro management. How do you account for all this and satisfy your superiors, who think your job is chilling all day surfing the net or writing blog posts, while making this as low effort for yourself as possible? May I introduce the coin changing problem more generally known as the subset sum. Now we aren’t going to get into dynamic programming or any other algorithmic jazz. My main goal with this post is to point out applications of common algorithms you come across in CS that have practical applications, in this case dealing with management.

import random
import openpyxl
import os
from datetime import date
#############################################
# VARIABLES                                 #
#############################################
hour_combinations = []
task_categories = [
                    'task category 1: description',
                    'task category 2: description',
                    'task category 3: description',
                    'task category N: description',
                    'project 1',
                    'project 2',
                    'project 3',
                    'project N'
                  ]
workbook_save_location="C:\\Users\\adegraca\\Desktop\\crappycrap.xlsx"
worksheet_headers = ["category", "location", "hours"]
locations=["remote", "corporate", "field office", "blah blah blah"]
worksheet_data = []
#############################################
# FUNCTIONS                                 #
#############################################
'''
Generates a list numbers between start and stop in step increments.
For this I usually put the longest time a task has taken me
and the smallest as start and stop respectively.
'''
def get_task_increments(start,stop,step=0.25):
    numbers = []
    while start <= stop:
        numbers.append(start)
        start += step
    return numbers
'''
Algorithmic analysis coming in clutch.
Using the list generated in the function above,
We will generate a list of all possible combinations of task time increments
you can have in an 8 hour work day.
'''
def subset_sum(numbers, target, output, partial=[], debug='false'):
    s = sum(partial)
    # check if the partial sum is equals to target
    if s == target:
        if debug == 'true': 
            print ("sum(%s)=%s" % (partial, target))
        output.append(partial)
    if s >= target:
        return  # if we reach the number why bother to continue
    for i in range(len(numbers)):
        n = numbers[i]
        remaining = numbers[i+1:]
        subset_sum(remaining, target, output, partial + [n])
def createWorkSheetData():
    todays_task_hours = random.choice(hour_combinations)
    todays_task_categories = []
    todays_task_locations = []
    for i in range(0, len(todays_task_hours)):
        todays_task_categories.append(random.choice(task_categories))
        todays_task_locations.append(random.choice(locations))
    return todays_task_categories, todays_task_locations, todays_task_hours 
def createWorkdayLog(workbook_path):
    wb = NULL
    if os.path.exists(workbook_path):
        wb = openpyxl.load_workbook(workbook_path)
    else:
        wb = openpyxl.Workbook()
    # insert newest sheet in front
    sheetname = date.today().strftime("%B %d, %Y")
    ws = wb.create_sheet(sheetname,0) 
    wb.active = ws
    wb.active.append(worksheet_headers)
    data = createWorkSheetData()
    for i in range(0, len(data[0])):
        print(data[0][i], data[1][i], data[2][i])
        row = [data[0][i], data[1][i], data[2][i]]
        wb.active.append(row)
    wb.save(workbook_path)
#############################################
# MAIN                                      #
#############################################  
#print(get_task_increments(0.25,3.0,0.25))
subset_sum(get_task_increments(0.25,3.0,0.25), 8.5, hour_combinations)
createWorkdayLog(workbook_save_location)
#print(hr_combinations)

If you’re not savy at reading code:

• We create a list of tasks we are supposed to be doing
• break up our workday in pieces pieces of time
  • we use subset_sum get a list of all possible permutations of a work day, then we randomly choose a workday in the list.
• randomly assign tasks to each time segment in our workday.
• Then we save the current workday in a spreadhsheet.

The last step is to schedule this script to run as a task. For example I have this run M-F at 8:30 am on my work desktop.
If you wanted to be a golden boy (or girl!) you could even email your supervisor your work.
At my job for instance:

• C level exec: Zinhart we really appreciate you creating data that demonstrates the function of System Adminstration at our company.
• Zinhart: My pleasure.
• C level exec: How’s your current project? going.
• Zinhart: With the current deadline will prob need overtime to get it done.
• C level exec: How much overtime do you think you’ll need to complete your next project?
• Zinhart: Don’t have an estimate yet.
• C level exec: Ok let me know.
• Zinhart: Will do
• Zinhart to self: Its’s already done

( ͠≖ ͜ʖ͠≖)

We do not win by argueing or complaining.
We win by being smarter.
Now your PM/exec/whatever superior can have metrics they need to justify their teams productivity (and cost) and with your new found free time (I mean after all it takes alot of dilligent work to create an accurate representation of your time) you can look for another job or pursue a certification or write blog pots.