"There is not enough space on the disk" Error - Over 100 GB Free Space RRS feed

  • General discussion

  • I am evaluating Service Fabric for use at my company.  I went to the portal and spun up a service fabric cluster and then deployed the default template Service Fabric WebAPI service to via Visual Studio.

    I then opened the `Service Fabric Explorer` and I saw it had an error in the System Node.  The error was under `fabric:/System/FaultAnalysisService`.  Under that node there was a guid(ish) value of 00000000-0000-0000-0000-000000005000 (for the partition I think).

    Under that it had `Primary (_MainNode_0)`.  An error was listed in that node.  The only understandable part of the error was "There is not enough space on the disk. (Exception from HRESULT: 0x80070070)"

    I was confused by this because I had picked on of the "recommended" VM types.  So I had figured it would "just work".

    Here is where it gets interesting: 

    I figured out how to remote desktop to the machine (to ensure that it really did have disk space).  Sure enough if had 111GB free on the C: drive.

    But when I looked at the error message again, the Node had changed.  It was now saying `Primary (_MainNode_1)`!

    So I logged into that one and found that it too had 111GB of free space.  After I did that, the error changed again.  Now it was showing `Primary (_MainNode_2)`.

    So, I tried did a remote desktop session to MainNode02 (It had 113GB free).

    Once I did that the error went back to _MainNode_1!

    So here are my questions:

    1. What can I do to fix this error?
    2. What is going on to cause this error?
    3. This kind of unpredictability does not seem like something I really want to use for my production systems.  Is there some way to prove that this is a weird occurrence and that Service Fabric is actually a stable platform?


    So I think it was the "D" drive.  From what I can see, the VM is setup with tons of free space on the C drive, but only 7 GB total on the D Drive.  (And the Service Fabric Logs use up 4 GB of that).

    I can't seem to find a way to allocate more space to the D drive....  

    I am guessing that this is just a issue with the Portal leading me to failure rather than success.  Service Fabric its self is probably fine.

    Friday, April 21, 2017 5:13 PM

All replies