The destination for incoming CrashPlan backups on my server (11.04) is


(SeagateBig is the volume name of my 2TB USB drive).

When the server boots, two things happen: 1) SeagateBig auto-mounts and 2) CrashPlan starts.

The problem is, that often these two things don't happen in that order. Then I get:

  • Crashplan starts
  • looks for /media/SeagateBig
  • doesn't find it
  • instead of waiting for it, CREATES IT

Now it's backing up onto my / filesystem. NOT COOL.

Meanwhile, when SeagateBig finally gets around to mounting, it finds that /media/SeagateBig already exists, shrugs, and creates /media/SeagateBig_ as its mount point.

What I need is a way for the order to be enforced - where SeagateBig mounts and then and only then the CrashPlan service is started.

Unless I learn that CrashPlan can be told to wait for its destination directory, never to create it... which I am also investigating. But the CrashPlanEngine script is installed by the product so I am loath to modify it, as I know I could by having it loop until df greps successfully for "SeagateBig".

UPDATE: More research has led to asking a better question. See this one.

| improve this question | | | | |

I am not sure how Crashplan works, but here are some tips:

  • If it has an automatically startup plan, disable it, and find out which is the executable for Crashplan.
  • Then create a shell script that looks like this:

    sleep 30 && >name of executable<;

Note: This will stop the execution for 30 seconds. You can use whatever time

name your script start_crashplan.sh

Make it executable:

chmod +x start_crashplan.sh

Then add your script in the startup.

Sorry if I was of the subject, but I think this way you have maximum control of what you start and when.

| improve this answer | | | | |
  • This is kind of where I don't want to go -- I am trying not to break CrashPlan so that when they send an update it still works. I see that the script that starts CrashPlan has $local_fs and $remote_fs in the LSB but it's running ahead of the mounting of devices in /etc/mtab ANYWAY. – DavidF Dec 1 '11 at 13:42
  • 1
    Just heard from CrashPlan support -- they said, "Modify the script." sigh – DavidF Dec 7 '11 at 18:47
  • :) What can you do, sometimes you need to get your hands dirty. – bioShark Dec 7 '11 at 19:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.