kube generate -> play fails due name already in use

[ikke-t]

Is this a BUG REPORT or FEATURE REQUEST? (leave only one on its own line)

/kind bug

Description

podman fails to start pod with pod name same as container name. This is from yaml podman generate kube created.

Steps to reproduce the issue:

1. create pod

sudo podman run -d  --name postgres \
 -e POSTGRES_USER=pg \
 -e POSTGRES_PASSWORD=pg \
 -e POSTGRES_DB=test \
 -e PGDATA="/var/lib/postgresql/data/pgdata" \
 postgres:9.6

2. generate kube

sudo podman generate kube postgres > postgres.yml

3. Delete pod

$ sudo podman pod rm -f postgres
b13caa435916e79f45131fe816270eb93ce893e064b3c226702f82775721308c

4. try playing back the kube yaml

sudo podman play kube postgres.yml 

Describe the results you received:

$ sudo podman  rm -f postgres
d9aeba5043379cfc1db9ffeb722b8fa9a998150a90add978186acc2b24334604
$ sudo podman play kube postgres.yml 
b13caa435916e79f45131fe816270eb93ce893e064b3c226702f82775721308c
Error: name postgres is in use: container already exists

Describe the results you expected:

postgres to start

Additional information you deem important (e.g. issue happens only occasionally):

Output of podman version:

sudo yum install https://kojipkgs.fedoraproject.org//packages/podman/1.0.1/38.dev.gitcf52144.fc31/x86_64/podman-1.0.1-38.dev.gitcf52144.fc31.x86_64.rpm

Output of podman info --debug:

debug:
  compiler: gc
  git commit: ""
  go version: go1.12rc1
  podman version: 1.0.1-dev
host:
  BuildahVersion: "1.7"
  Conmon:
    package: podman-1.0.1-38.dev.gitcf52144.fc31.x86_64
    path: /usr/libexec/podman/conmon
    version: 'conmon version 1.12.0-dev, commit: b3ddd847a8d3802a2dcbf0e3fc632fb136a569c6'                                                        
  Distribution:
    distribution: fedora
    version: "29"
  MemFree: 196767744
  MemTotal: 8052445184
  OCIRuntime:
    package: runc-1.0.0-68.dev.git6635b4f.fc29.x86_64
    path: /usr/bin/runc
    version: |-
      runc version 1.0.0-rc6+dev
      commit: ef9132178ccc3d2775d4fb51f1e431f30cac1398-dirty
      spec: 1.0.1-dev
  SwapFree: 8017932288
  SwapTotal: 8199860224
  arch: amd64
  cpus: 4
  hostname: ohuska.localdomain
  kernel: 4.20.6-200.fc29.x86_64
  os: linux
  rootless: false
  uptime: 72h 14m 29.43s (Approximately 3.00 days)
insecure registries:
  registries: []
registries:
  registries:
  - docker.io
  - registry.fedoraproject.org
  - quay.io
  - registry.access.redhat.com
  - registry.centos.org
store:
  ConfigFile: /etc/containers/storage.conf
  ContainerStore:
    number: 3
  GraphDriverName: overlay
  GraphOptions:
  - overlay.mountopt=nodev
  GraphRoot: /var/lib/containers/storage
  GraphStatus:
    Backing Filesystem: extfs
    Native Overlay Diff: "true"
    Supports d_type: "true"
    Using metacopy: "false"
  ImageStore:
    number: 11
  RunRoot: /var/run/containers/storage

Additional environment details (AWS, VirtualBox, physical, etc.):

[ikke-t]

I was actually trying to reproduce the bug that podman couldn't create pod from it's own generated kube yaml. Yaml is in here: #2303 (comment)

This was supposed to be simple reproducal for @baude, who got this result from yaml playback:

18:16:17 | <@baude> | ikkeT, the postgres container dies with:
-- | -- | --
18:16:23 | <@baude> | /usr/local/bin/docker-entrypoint.sh: line 75: initdb: command not found
18:17:25 | <@baude> | ikkeT,  the rabbit one fails with :
18:17:28 | <@baude> | /usr/local/bin/docker-entrypoint.sh: line 345: rabbitmq-plugins: command not found
18:17:35 | <@baude> | /usr/local/bin/docker-entrypoint.sh: line 381: exec: rabbitmq-server: not found
18:18:03 | <@baude> | and memcached failed with:
18:18:06 | <@baude> | can't run as root without the -u switch

[ikke-t]

@baude fixed this by changing the name of pod during play if there is name collission. Pod becomes named "-pod" at the end of name to make names different. I tested this and it works for me. Thanks