This software is called Google Updater, and it secretly “calls home” on a regular basis and downloads updates to your Google software without either asking before, or notifying you after, doing so. In Developer circles, this is considered very shady practice. Users should be asked for consent and informed when software makes changes to either itself or the user’s computer, and ideally those notifications should tell the user what has been changed and how the changes could impact them.
Before I beat this drum any harder, however, I owe you at least the other side of the story. If I worked for Google, I’d probably come up with this response: “Hey look, a major source of computer virus and malware infections is that users are often using out-of-date software that hasn’t been patched to combat newly-discovered exploits. No matter how much we tell users to keep ther software up-to-date, the truth is the majority don’t. We provide an automatic updater so that users don’t have to worry about it, and can be assured they’re always using the latest and safest version of our software”.
I’ve heard this argument so many times, I don’t doubt it’s something close to what Google would actually argue. My problem with this is that while automatic updates can be a good thing if they’re security related, it’s not at all clear why an app should be updating itself automatically for any other reason, or why it’s updating itself without providing notifications about when and what updates were made.
If an independent developer did that, they’d almost certainly find their software labelled as “suspicious” at best, and “dangerous” at worst. The fact that Google is a multinational, global enterprise with a stranglehold on the internet, and which is often tangling with the law in countries throughout the world, may make you feel more or less confident that they can be trusted more than independent developers, whose income depends very much on their reputation. I’ll leave that one for the reader to decide. 😉
Do I have Google Updater?
To see if you’ve got Google Updater hiding on your system, try this quick test in Terminal. Triple click the line of code below to highlight it.
defaults read com.google.Keystone.Agent
If you’ve previously installed my Terminal workflow, just hit control-opt-cmd-T or right/control click and choose “Services > Run in Terminal” from the contextual menu. Alternatively, if you have my free utility app FastTasks 2, the Analyser’s Profile view will show you if Google Updater is installed (see ‘Locate Google Updater’ below for the locations to check in the profile view). Elsewise, manually copy and paste it into a Terminal window.
If the result comes back as
Domain com.google.Keystone.Agent does not exist
you’re fine. Google Updater has not found its way into your system. Anything else and you’re going to need to decide whether you want to remove it or not. If you’re a regular Chrome user, keeping Updater might prove convenient, though you’ll have to live with the idea that the app is updating itself in ways over which you have no control. If you rarely or never use Chrome, there’s no reason to have this hidden process regularly calling home to Google every time you’re connected to the net.
How do I remove it?
You have two options. You can either disarm it or you can nuke it. Disarming it is simplest, it’s a one-line Terminal command:
defaults write com.google.Keystone.Agent checkInterval 0
This command tells the Updater how often to “call home”. A value of 0 basically means ‘never’. Disarming it is probably better than nuking it if you still keep Chrome on your system and use it occasionally. You can temporarily set it back to something like ‘once a week’ from time to time to check for security updates with
defaults write com.google.Keystone.Agent checkInterval 604800
Nuking the Google Updater is a bit more complex. You’ll want to run some uninstaller commands, and then you’ll want to go and clear up the crud that is still left behind. And before you can do either of those, you need to find out where it’s hiding. So, we have a three-step process.
1. Locate Google Updater
Triple click the first of these two lines, and choose ‘Services > Reveal in Finder’ from the contextual menu (that’s another right-click or control-click on the selected line), and then repeat for the second line:
You will likely get the error message “The operation can’t be completed because the item can’t be found” from one of these lines, but not the other. Note that the difference is all in the presence or absence of the tilde ~. Make a note of which one worked, and run the appropriate commands in step 2.
2. Run the uninstaller commands
Run these in Terminal (again, triple clicking to highlight and doing the usual trick afterwards with shortcut key or Services menu if you have my workflow installed), one at a time:
If the Updater was in your user library (with the tilde ~), then first do this (it’s all one line):
Update: please read the Comments for latest commands
If the Updater was in your domain library (no tilde ~), then first do this (it’s all one line):
sudo python /Library/Google/GoogleSoftwareUpdate/GoogleSoftwareUpdate.bundle/\
and enter your Admin password (note that you won’t see any indication of your password being typed in the Terminal window). Then do this:
sudo touch /Library/Google/GoogleSoftwareUpdate
3. Clear up the crud
If the updater was in your user library, open that now and go to
and delete the folder called ‘GoogleSoftwareUpdate’. If you don’t use any other Google software (I don’t), you can just delete the entire ‘Google’ parent folder.
If the updater was in your domain library, search for the same folder and send it to the trash. You will need to give Finder your admin password to authorise the move.
Next, let’s just check the uninstaller was successful. Look for the following. If you don’t find them, good (the installer did its job). If you do, help them on their way to oblivion by sending them to the trash:
If you’ve deleted Chrome from your Applications folder too, then you might as well hunt down and exterminate its prefs list while you’re at it:
The following sources were used in researching this post:
‘Don’t be evil’ picture was remediated from here.
I recently helped a user track down a keylogger on his computer. While this case was certainly one of being ‘snooped’ on (the keylogger had been installed on his own machine by a third party), this got me to thinking: are keyloggers ever ethical?
While they can be used for nefarious purposes (spying on your partner or colleagues, for example), they can also help you track down a stolen laptop, aid companies in detecting illegal behaviour or corporate espionage by employees, and help parents protect their children from internet pests.
Feeling conflicted, I thought the best people to ask would be you! So what do you think? Click on one of the options below and hit ‘vote’ to take the poll, and/or leave your thoughts in the Comments!