1. 04 Aug, 2014 1 commit
  2. 31 Jul, 2014 4 commits
  3. 30 Jul, 2014 2 commits
  4. 28 Jul, 2014 2 commits
  5. 25 Jul, 2014 7 commits
  6. 16 Jul, 2014 1 commit
    • Mike Hibler's avatar
      Fix to run slicefix after all images have been loaded. · 03bec3a7
      Mike Hibler authored
      Otherwise we might run slicefix after a base image and before a delta
      which could cause a train wreck in the FS if slicefix changes allocate
      or deallocate blocks.
      
      Also, grudgingly update some of the shell syntax to use constructs
      introduced in the early 1990s. Sigh...this script will no longer run
      on a PDP-11.
      03bec3a7
  7. 12 Jul, 2014 1 commit
    • Mike Hibler's avatar
      Make sure we use -i for sendmail when we pipe into it. · 355f5aa8
      Mike Hibler authored
      This prevents a line with a single "." from meaning EOF to sendmail.
      How arcane!
      
      I discovered this when I ran a create_image and I didn't get the complete
      log mailed to me. This is because create_image did a frisbee download of an
      image with a single chunk, which of course printed out:
      
        Using Multicast 235.252.1.187
        Joined the team after 0 sec. ID is 1586355915. File is 1 chunks (963200 bytes)
        .
      
      Fortunately, "arcane" is my middle name, so it didn't take me long to find
      this...
      355f5aa8
  8. 11 Jul, 2014 1 commit
  9. 09 Jul, 2014 2 commits
  10. 20 Jun, 2014 1 commit
  11. 12 Jun, 2014 1 commit
  12. 02 Jun, 2014 1 commit
  13. 30 May, 2014 5 commits
  14. 26 May, 2014 1 commit
  15. 25 May, 2014 1 commit
  16. 21 May, 2014 2 commits
  17. 20 May, 2014 1 commit
  18. 19 May, 2014 2 commits
  19. 16 May, 2014 2 commits
    • Mike Hibler's avatar
      Add a variable to control what space is included in the xen_vg VG. · 2fbfc5dd
      Mike Hibler authored
      Setting LVM_FULLDISKONLY will effectively tell it to not use extra
      space on the system disk when constructing the VG used for vnode disk
      creation. This can affect performace for everybody, especially if it
      uses multiple little partitions on the system disk. However, this is
      turned off by default since, on our d710 nodes, most of the extra
      space is on the system disk. Setting this should be plumbed through
      to the user somehow so they can choose.
      2fbfc5dd
    • Mike Hibler's avatar
      Check for standard static-network-up event in addition to our custom event. · 78137599
      Mike Hibler authored
      This will prevent the image from hanging inexplicably at boot when used
      external to Emulab control in a situation like, I don't know, maybe your
      machine is miles away at a datacenter and you would have to go stand
      around in a noisy machine room for hours to figure out what is wrong...
      78137599
  20. 15 May, 2014 1 commit
  21. 09 May, 2014 1 commit
    • Mike Hibler's avatar
      Change to use only geom and GPTs for FBSD 10 and above. · 0639a119
      Mike Hibler authored
      As of FreeBSD 10, I am tired of fighting the old MBR tools.
      So for whole disks (slice == 0) we are going to use GPT so that we
      can get good (1M) alignment and potentially big-ass partitions with
      a minimum of fuss.
      
      This means that you cannot image those partitions since imagezip
      does not yet (as of 05/2014) understand GPT. But we have no mechanism
      for capturing an image from anything but the system disk anyway, so
      we can live with this until GPT support is added.
      0639a119