uploading stemcell fails with EOF about bosh-lite HOT 6 CLOSED

You might want to double check the integrity of the downloaded stemcell. The digest can be found from http://bosh.io/stemcells/bosh-warden-boshlite-ubuntu-trusty-go_agent and for 3363.12 should be a72b63c5b17ab2fd8e9e92c4a5019298765d8dc1.

You can also double check you have enough disk space (both on the vm via vagrant ssh and on the host).

Another suggestion would be running vagrant reload, particularly if you have moved the bosh-lite directory after doing the initial vagrant up (see #109).

from bosh-lite.

Thanks for your help. I followed your suggestion to narrow down the problem. Checksum is ok. vagrant reload didnt change anything. For the diskspace, I am not sure. While there is plenty of space in general, there is limited space under /var/vcap/data/sys/run . Can that be the problem? If not, what else can I try to analyse the issue.

vagrant@agent-id-bosh-0:~$ df -h
Filesystem                   Size  Used Avail Use% Mounted on
udev                         2.9G  4.0K  2.9G   1% /dev
tmpfs                        597M  1.4M  596M   1% /run
/dev/mapper/ubuntu--vg-root   78G  2.9G   71G   4% /
none                         4.0K     0  4.0K   0% /sys/fs/cgroup
none                         5.0M     0  5.0M   0% /run/lock
none                         3.0G     0  3.0G   0% /run/shm
none                         100M     0  100M   0% /run/user
/dev/sda1                    236M   82M  142M  37% /boot
tmpfs                        1.0M   52K  972K   6% /var/vcap/data/sys/run
vagrant                       35G   15G   20G  44% /vagrant
cgroup                       3.0G     0  3.0G   0% /tmp/garden-/cgroup

from bosh-lite.

That's odd. Try recreating the vagrant VM from scratch vagrant destroy && vagrant up just to be sure everything is in a clean state. Also, make sure you're running recent versions of You could also try manually extracting the stemcell on the vagrant vm with tar -xzf stemcell-file.tgz and see if it fails (if so, you may be able to enable verbose mode; it'd be weird for it to fail though since checksum matches).

Tried locally and it seems to work...

$ wget --content-disposition https://s3.amazonaws.com/bosh-core-stemcells/warden/bosh-stemcell-3363.12-warden-boshlite-ubuntu-trusty-go_agent.tgz
$ BOSH_CLIENT=admin BOSH_CLIENT_SECRET=admin gobosh -e https://192.168.50.4:25555 --ca-cert ca/certs/ca.crt us bosh-stemcell-3363.12-warden-boshlite-ubuntu-trusty-go_agent.tgz
Using environment 'https://192.168.50.4:25555' as client 'admin'

######################################################### 100.00% 142.37 MB/s 2s
Task 1

23:36:45 | Update stemcell: Extracting stemcell archive (00:00:02)
23:36:47 | Update stemcell: Verifying stemcell manifest (00:00:00)
23:36:47 | Update stemcell: Checking if this stemcell already exists (00:00:00)
23:36:47 | Update stemcell: Uploading stemcell bosh-warden-boshlite-ubuntu-trusty-go_agent/3363.12 to the cloud (00:00:12)
23:36:59 | Update stemcell: Save stemcell bosh-warden-boshlite-ubuntu-trusty-go_agent/3363.12 (28bbbe8a-3124-49b4-57f0-c374ae0a9d7e) (00:00:00)

Started  Fri Mar 17 23:36:45 UTC 2017
Finished Fri Mar 17 23:36:59 UTC 2017
Duration 00:00:14

Task 1 done

Succeeded

from bosh-lite.

vagrant destroy / up didn't help but I could get it to work after lowering the number of cpus in the virtualbox ! My Vagrantfile now looks like (the line with ### is added):

config.vm.provider :virtualbox do |v, override|
    override.vm.box_version = '9000.137.0' # ci:replace
    v.cpus = 1 ###
  end

After that, the bosh upload works. Can there be some kind of racecondition? untgz-ing starting before the upload is finalized ?

from bosh-lite.

Not sure why CPUs would be affecting things, but that makes me think your host may simply be underpowered. If running on a cloud workstation, perhaps try scaling to something more powerful.

From the code side, ruby isn't invoked until the stemcell upload was fully written to disk, so I can't think of a race condition like that.

Do you see any difference if doing bosh upload-stemcell https://s3.amazonaws.com/bosh-core-stemcells/warden/bosh-stemcell-3363.12-warden-boshlite-ubuntu-trusty-go_agent.tgz as opposed to uploading a local file?

from bosh-lite.

I can't follow up further on this issue. My suspicion is there is something odd and very specific to my setup. Closing therefore.

from bosh-lite.