# Broken/unstable host/stack config

**URL:** <https://forums.suse.com/t/broken-unstable-host-stack-config/7994>\
**Category:** Rancher 1.x\
**Created:** [December 5, 2017, 9:17am UTC](https://forums.suse.com/t/broken-unstable-host-stack-config/7994 "2017-12-05T09:17:05Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![cthiebault](https://sea2.discourse-cdn.com/flex022/user_avatar/forums.suse.com/cthiebault/32/2378_2.png) [@cthiebault](https://forums.suse.com/u/cthiebault)\
**Post date:** [December 5, 2017, 9:17am UTC](https://forums.suse.com/t/broken-unstable-host-stack-config/7994/1 "2017-12-05T09:17:05Z")

</div>

I’ve made a mistake by bypassing Rancher and removed a container manually and now I always see this error ☹

```auto
Failed to allocate instance [container:1i4730]: Bad instance [container:1i4730] in state [error]:
Allocation failed: volume [6369] must have exactly these pool(s): [25], valid host(s) [28]

```

Here are the steps (using Rancher **1.6.12** ):

1. I had a stack deployed on a host with volume mapping local dir to container dir:

```auto
services:
  slave:
    volumes:
      - /var/jenkins:/home/jenkins

```

1. I created a new version of this stack replacing this local volume by a Docker volume

```auto
services:
  slave:
    volumes:
      - jenkins-slave:/home/jenkins
volumes:
  jenkins-slave:
    driver: local

```

1. As I had an error on stack upgrade, I logged in on the host and manually removed (!!!) stack container ☹
2. Since then, I can’t use this host within my environment… Host is well added but my stack can’t be deployed to it…  
I’ve tried to reinstall a fresh OS (but kept IP) but with same results…

```auto
Failed to allocate instance [container:1i4730]: Bad instance [container:1i4730] in state [error]:
Allocation failed: volume [6369] must have exactly these pool(s): [25], valid host(s) [28]

```

**Is there a way to fix/reset Rancher database for a host?**

**UPDATE 1:**  
It’s even worse than I thought: adding new fresh host gives me the same error!  
It seems that it’s the stack/container/volume config that is broken 😱  
Any idea of what should I do to fix my mistakes?

**UPDATE 2:**  
Same error with a fresh new environment & clean hosts ☹

- I’ve evacuated & removed my hosts from broken environment
- I’ve deleted this environment
- I’ve removed all containers & volumes from my hosts (just to be sure there is no trace of previous stuff)
- I’ve created a fresh new environment
- I’ve added these clean hosts to my new env  
–\> Jenkins master & one slave is OK, but Rancher is not able to deploy slave to other host ☹

I really don’t understand as I started from a clean env with clean hosts…

```auto
Failed to allocate instance [container:1i4992]: Bad instance [container:1i4992] in state [error]: 
Allocation failed: valid host(s) [48], volume [7098] must have exactly these pool(s): [45]

```