Cloud foundry on Google Compute engine can't create container

Question

I am very new with Cloud foundry. I have added cloud foundry for google compute engine platform by this guides source1 and source2.

Terraform was used for creating needed infrastructure. It seemed all was fine I didn't get any errors during deployment cloud foundry itself and bosh cck command returns that there are no any problems. But when I tried to deploy my hello world app, I got next error message in terminal after cf push command:

Creating container Failed to create container

FAILED

Error restarting application: StagingError.

After checking log files I found next message:

{
   "timestamp":"1474637304.026303530",
   "source":"garden-linux",
   "message":"garden-linux.loop-mounter.mount-file.mounting",
   "log_level":2,
   "data":{
      "destPath":"/var/vcap/data/garden/aufs_graph/aufs/diff/08829a3252c1d60729e3b5482b0fb109652c9ab5beff9724e4e4ae756a0bc3ce",
      "error":"exit status 32",
      "filePath":"/var/vcap/data/garden/aufs_graph/backing_stores/08829a3252c1d60729e3b5482b0fb109652c9ab5beff9724e4e4ae756a0bc3ce",
      "output":"mount: wrong fs type, bad option, bad superblock on /dev/loop0,\n       missing codepage or helper program, or other error\n       In some cases useful info is found in syslog - try\n       dmesg | tail  or so\n\n",
      "session":"2.276"
   }
}{
   "timestamp":"1474637304.026949406",
   "source":"garden-linux",
   "message":"garden-linux.pool.acquire.provide-rootfs-failed",
   "log_level":2,
   "data":{
      "error":"mounting file: mounting file: exit status 32",
      "handle":"ec6e7469-0ef0-48a8-bcd0-82f4a2ea173f-5de2e641d9284aeea209ca447ffffb6d",
      "session":"9.545"
   }
}
{
   "timestamp":"1474637304.027062416",
   "source":"garden-linux",
   "message":"garden-linux.garden-server.create.failed",
   "log_level":2,
   "data":{
      "error":"mounting file: mounting file: exit status 32",
      "request":{
         "Handle":"ec6e7469-0ef0-48a8-bcd0-82f4a2ea173f-5de2e641d9284aeea209ca447ffffb6d",
         "GraceTime":0,
         "RootFSPath":"/var/vcap/packages/rootfs_cflinuxfs2/rootfs",
         "BindMounts":[
            {
               "src_path":"/var/vcap/data/executor_cache/6942123d3462ad9d21a45729c3cae183-1474475979582384649-1.d",
               "dst_path":"/tmp/lifecycle"
            }
         ],
         "Network":"",
         "Privileged":true,
         "Limits":{
            "bandwidth_limits":{

            },
            "cpu_limits":{
               "limit_in_shares":512
            },
            "disk_limits":{
               "inode_hard":200000,
               "byte_hard":6442450944,
               "scope":1
            },
            "memory_limits":{
               "limit_in_bytes":1073741824
            }
         }
      },
      "session":"11.44187"
   }
}{
   "timestamp":"1474637304.034646988",
   "source":"garden-linux",
   "message":"garden-linux.garden-server.destroy.failed",
   "log_level":2,
   "data":{
      "error":"unknown handle: ec6e7469-0ef0-48a8-bcd0-82f4a2ea173f-5de2e641d9284aeea209ca447ffffb6d",
      "handle":"ec6e7469-0ef0-48a8-bcd0-82f4a2ea173f-5de2e641d9284aeea209ca447ffffb6d",
      "session":"11.44188"
   }
}

And meantime in dmesg | tail I got next:

[161023.238082] aufs test_add:283:garden-linux[7681]: uid/gid/perm /var/vcap/data/garden/aufs_graph/aufs/diff/d350dcd30f6d6f8b37eabe06a3b73bcea0a87f9aff4edf15f12792269fc9f97c 4294967294/4294967294/0755, 0/0/0755 [161023.238109] aufs au_opts_verify:1597:garden-linux[7681]: dirperm1 breaks the protection by the permission bits on the lower branch [161023.413392] device wtj3qdqhig0t-0 entered promiscuous mode

I'm not sure that this issues connected or that it is issue at all, but I post them here in order to be sure, that I didn't miss anything.

I don't know how to fix this problem and where, should I look solution for terraform scripts or for bosh manifest files. We have micro service architecture with three nodes on node js and one on ruby, so deployment is very important question for us.

here is my application manifest.yml file:

---
applications:
- name: hello_cloud
  memory: 128M
  buildpack: https://github.com/cloudfoundry/nodejs-buildpack
  instances: 1
  random-route: true
  command: "node server.js"

My goal is to be able deploy applications using cloud foundry. If you have any additional questions or I wrote something unclear feel free to write me.

You might consider opening a GitHub issue on the repo containing the docs you followed: github.com/cloudfoundry-incubator/bosh-google-cpi-release/…. They would probably be in the best position to help you understand if something is misconfigured in relation to what they have documented. — Amit Kumar Gupta

Jeff Johnson Jeff Johnson · Accepted Answer · 2016-09-28T17:34:01

This issue is related a conflict between garden and the 4.4 Linux kernel. To use the example cloudfoundry manfest, use the follow stemcell:

bosh upload stemcell https://bosh.io/d/stemcells/bosh-google-kvm-ubuntu-trusty-go_agent?v=3262.19
bosh deploy

You may need to delete your cf deployment before re-deploying due to quota issues.

Cloud foundry on Google Compute engine can't create container

1 Answers