Hey folks,
I'm having a really weird problem with my host machine. I have two containers, let's say container A and B. My container A is used only for utilities and it contains the full ansible configurations to configure my servers. I run ansible in A to configure B. Both servers are running CentOS.
This is working fine, but one time, when I ran ansible-playbook, the server stopped responding. I checked the other container, and it was not responding too. Then I went to my host machine and it was offline. I was really confused, so I thought it could be some network problem, but I checked with my provider and everything was OK. I hard rebooted the server and it started working again.
I checked the logs in my host machine and found nothing. Nothing at all. So I thought it was an isolated situation, but after a few hours it happened again. In the same freaking task.
I checked the logs again and found nothing at all. This was yesterday. Today it happened again, on the exact same task, and no trace at all on the logs. This is not happening every time, it's pretty random... I ran a bunch of hardware tests and it's all fine with my hardware, so it must be something on the software.
I thought I could be a kernel panic, but again, I don't see anything on the logs... But what can happen in a container that would crash the whole server? Aren't containers supposed to be "sandboxed"?
Here is the Ansible task that it's running when the server crashes:
It's basically crashing on the Install IUS task. The script it is running it's available here: https://setup.ius.io/ It's only a few `yum` commands...
I might be looking in the wrong place for logs, I'm no expert on debian based systems, so I would love some assistance. This problem is driving me nuts and I really need to have a stable machine, as I have some critical services running on it right now...
Thanks!
I'm having a really weird problem with my host machine. I have two containers, let's say container A and B. My container A is used only for utilities and it contains the full ansible configurations to configure my servers. I run ansible in A to configure B. Both servers are running CentOS.
This is working fine, but one time, when I ran ansible-playbook, the server stopped responding. I checked the other container, and it was not responding too. Then I went to my host machine and it was offline. I was really confused, so I thought it could be some network problem, but I checked with my provider and everything was OK. I hard rebooted the server and it started working again.
I checked the logs in my host machine and found nothing. Nothing at all. So I thought it was an isolated situation, but after a few hours it happened again. In the same freaking task.
I checked the logs again and found nothing at all. This was yesterday. Today it happened again, on the exact same task, and no trace at all on the logs. This is not happening every time, it's pretty random... I ran a bunch of hardware tests and it's all fine with my hardware, so it must be something on the software.
I thought I could be a kernel panic, but again, I don't see anything on the logs... But what can happen in a container that would crash the whole server? Aren't containers supposed to be "sandboxed"?
Here is the Ansible task that it's running when the server crashes:
Code:
- name: Download IUS Project install script
get_url: url=https://setup.ius.io/
dest=/tmp/ius.sh
- name: Install IUS
shell: bash /tmp/ius.sh
ignore_errors: yesIt's basically crashing on the Install IUS task. The script it is running it's available here: https://setup.ius.io/ It's only a few `yum` commands...
I might be looking in the wrong place for logs, I'm no expert on debian based systems, so I would love some assistance. This problem is driving me nuts and I really need to have a stable machine, as I have some critical services running on it right now...
Thanks!
