As a remote I have had zendev timeout and serviced keep running in the background while serviced is pulling something from the vpn or upstream. 60s is too short.

zendev build serviced needs to exist.

Ubuntu 14.04
Docker 1.0
Increase limits in /etc/security/limits.conf

The following errors occurs when running zendev box create --type ubuntu europa with VAGRANT_DEFAULT_PROVIDER=vmware_fusion.

Bringing machine 'default' up with 'vmware_fusion' provider...
==> default: Box 'ubuntu-13.04-docker-v1' could not be found. Attempting to find and install...
    default: Box Provider: vmware_desktop, vmware_fusion, vmware_workstation
    default: Box Version: >= 0
==> default: Adding box 'ubuntu-13.04-docker-v1' (v0) for provider: vmware_desktop, vmware_fusion, vmware_workstation
    default: Downloading: http://vagrant.zendev.org/boxes/ubuntu-13.04-docker-v1.box
The box you attempted to add doesn't match the provider you specified.

Provider expected: vmware_desktop, vmware_fusion, vmware_workstation
Provider of box: virtualbox
The provider for this Vagrant-managed machine is reporting that it
is not yet ready for SSH. Depending on your provider this can carry
different meanings. Make sure your machine is created and running and
try again. Additionally, check the output of `vagrant status` to verify
that the machine is in the state that you expect. If you continue to
get this error message, please view the documentation for the provider
you're using.

Easy enough to work around, but I do use Fusion as my default provider for Vagrant.

The error message looks like this:

ZENDEV: Environment 'X' is not defined

where X is the name of the first zendev env that is dropped.

got

kernel: [  608.236848] init: docker main process ended, respawning
kernel: [  608.451873] init: docker main process (4230) terminated with status 1
kernel: [  608.451906] init: docker main process ended, respawning
kernel: [  608.676417] init: docker main process (4280) terminated with status 1
kernel: [  608.676445] init: docker respawning too fast, stopped

trying to start docker service,
to fix this, you need to install apparmor at step 3)

$ sudo apt-get install apparmor

Still refers to "resetserviced", among other things.

Update the repos

sudo apt-get update

right before the apt-get install -y lxc-docker command otherwise it wont find the package.

SSH in to a box (without -X), source .bashrc, and run: zendev use foo. When you perform sync (and some repos auth over ssh) you get prompted for the passphrase of the ssh key.

==> Cloning repositories
==> All repositories are cloned!
==> Checking for remote changes
Enter passphrase for key '/home/dgarcia/.ssh/id_rsa': Enter passphrase for key '/home/dgarcia/.ssh/id_rsa': Enter passphrase for key '/home/dgarcia/.ssh/id_rsa': Enter p
assphrase for key '/home/dgarcia/.ssh/id_rsa': Enter passphrase for key '/home/dgarcia/.ssh/id_rsa': Enter passphrase for key '/home/dgarcia/.ssh/id_rsa': Enter passphra
se for key '/home/dgarcia/.ssh/id_rsa': Enter passphrase for key '/home/dgarcia/.ssh/id_rsa': Enter passphrase for key '/home/dgarcia/.ssh/id_rsa': Enter passphrase for 
key '/home/dgarcia/.ssh/id_rsa': Enter passphrase for key '/home/dgarcia/.ssh/id_rsa': Enter passphrase for key '/home/dgarcia/.ssh/id_rsa': Enter passphrase for key '/h
ome/dgarcia/.ssh/id_rsa': Enter passphrase for key '/home/dgarcia/.ssh/id_rsa': Enter passphrase for key '/home/dgarcia/.ssh/id_rsa': Enter passphrase for key '/home/dga
rcia/.ssh/id_rsa': Enter passphrase for key '/home/dgarcia/.ssh/id_rsa': Enter passphrase for key '/home/dgarcia/.ssh/id_rsa': Enter passphrase for key '/home/dgarcia/.s
sh/id_rsa': Enter passphrase for key '/home/dgarcia/.ssh/id_rsa': Enter passphrase for key '/home/dgarcia/.ssh/id_rsa': Enter passphrase for key '/home/dgarcia/.ssh/id_r

To workaround this issue an ssh-agent must be started see: http://mah.everybody.org/docs/ssh

The ubuntu box needs to have docker login credentials for zenossqa and quay.io in order to build/run

It's hardcoded to 5 seconds. We should have checks.

The following sequence will yield a stack trace:

$ zendev add some_legitimate_manifest.json
$ zendev status

The problem can be resolved by

$ zendev sync

Need to prefer closer matches. difflib.get_close_matches() is probably the way to do it.

diff --git a/makefile b/makefile
index e88340d..6c130c4 100644
--- a/makefile
+++ b/makefile
@@ -106,6 +106,7 @@ devimg: $(SRCROOT)
-v $$ZENHOME:/opt/zenoss
-v $$SRCROOT:/mnt/src
-v $$PWD:/mnt/build \

•           -v $$HOME/.m2:/home/zenoss/.m2  \
          -w /mnt/build            \
          -e SRCROOT=/mnt/src      \
          zendev/devimg /bin/bash -c "exec /tmp/install_core.sh"
  

When I create a cluster and try to run serviced on a VM using 'zendev serviced', zenoss will not run. It seems the uid/gid on the global.conf is set to the zenoss user in the VM, not the container.

zenoss@foo01:~/dev$ serviced service attach zenping
bash-4.2# ls -l /opt/zenoss/etc/global.conf
-rw-rw---- 1 900 900 639 Nov 19 14:39 /opt/zenoss/etc/global.conf
bash-4.2# id zenoss
uid=1000(zenoss) gid=1000(zenoss) groups=1000(zenoss)
bash-4.2# exit
exit
zenoss@foo01:~/dev$ id zenoss
uid=900(zenoss) gid=900(zenoss) groups=900(zenoss),4(adm),24(cdrom),27(sudo),30(dip),46(plugdev),109(lpadmin),110(sambashare),999(docker)

In the logs I can see the following:

I1119 14:39:48.095270 00001 controller.go:226] Wrote config file /opt/zenoss/etc/global.conf
I1119 14:39:48.103532 00001 controller.go:197] Successfully ran command:'&{/usr/bin/chown [chown zenoss:zenoss /opt/zenoss/etc/global.conf] []  <nil>   [] <nil> 0xc2080e06a0 exit status 0 ?reflect.Value? true [0xc20804c0b8 0xc20804c0d8 0xc20804c0d8] [0xc20804c0b8 0xc20804c0d8] [0xc20804c0d0] [0x6b1e00] 0xc2080c40e0}' output:
I1119 14:39:48.112093 00001 controller.go:197] Successfully ran command:'&{/usr/bin/chmod [chmod 660 /opt/zenoss/etc/global.conf] []  <nil>   [] <nil> 0xc2080e0b20 exit status 0 ?reflect.Value? true [0xc20804c198 0xc20804c1b8 0xc20804c1b8] [0xc20804c198 0xc20804c1b8] [0xc20804c1b0] [0x6b1e00] 0xc2080c4230}' output:

In bootstrap.sh no temp file is created for the ls and drop commands so that sourcing the temp file raises an error and abort the script in some environments where set -e is executed, e.g., Jenkins build jobs.

serviced requires the cgroups-lite package to be installed so that the correct lxc-groups are created at boot time.

control-center/serviced@395292e

'zendev add' does not provide success/failure feedback. As a result the user cannot tell if the add was successful.

$ zendev add nonexistent.manifest.json
$ echo $?
0

Cleanup to feature/cluster branch required before merging:

• Cluster does not restart after shutdown. "hostname -f" fails. https://gist.github.com/jafcrocker/d243840e1b7df3d18811
  This is a problem with the way that we do the /etc/hosts file. The file is not available until vagrant mounts /vagrant directory, but we need it before that.
• The vagrantfile is getting too big - we should split out the bash initialization into a separate file instead of embedding it in the Vagrantfile.
• We write into .bashrc, but code is not idempotent, so we add the same changes each time we bring up the vm. That SCRIPT should probably be added as /etc/init.d/first-boot.sh which would need to remove itself after running
• fix 'zendev box' to properly create btrfs volumes
• add --cpus and --fssize to cluster and box