Bug 1546195

Summary: [RFE] Use home directory instead of /tmp to copy host-deploy files and execute a process from it
Product: [oVirt] ovirt-engine Reporter: Arano-kai <captcha.is.evil>
Component: Host-DeployAssignee: Martin Perina <mperina>
Status: CLOSED DEFERRED QA Contact: Pavel Stehlik <pstehlik>
Severity: low Docs Contact:
Priority: low    
Version: 4.1.9CC: bugs, mperina
Target Milestone: ---Keywords: FutureFeature
Target Release: ---Flags: sbonazzo: ovirt-4.3-
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2018-07-03 10:20:39 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: Infra RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
/var/log/ovirt-engine/engine.log none

Description Arano-kai 2018-02-16 14:53:31 UTC
Created attachment 1397057 [details]
/var/log/ovirt-engine/engine.log

Description of problem:
This is common practice to mount `/tmp` (and some others) with `noexec`.
So, direct call of `/tmp/somescript.sh` will fail, and `/usr/bin/env bash /tmp/somescript.sh` will do the job.

Steps to Reproduce:
1. mount -t tmpfs -o noexec tmp /tmp
2. In WebUI: Hosts -> installation -> Check for upgrade

Actual results:
In events: `Failed to check for available updates on host ov1.buh.gu.kyiv.dsns.gov.ua with message 'Unexpected connection termination`.
In engine.log: see attachment.

Expected results:
Host check for upgrades successfully

Comment 1 Sandro Bonazzola 2018-02-19 11:59:23 UTC
This was part of the original design of the deployment process.
Martin, how much work is to change the call to theovirt-host-deploy command through ssh issued by the engine? Do we want to change it?

Comment 2 Martin Perina 2018-02-19 13:11:23 UTC
Check for upgrade/Upgrade flow in 4.2 is now performed by Ansible, which is using home directory (which causes issue with vintage node BZ1540101). So we should probably switch to use home directory instead of /tmp, but we will need have an exception for vintage node until we will get rid of it.

Comment 3 Yaniv Kaul 2018-02-28 16:26:34 UTC
(In reply to Martin Perina from comment #2)
> Check for upgrade/Upgrade flow in 4.2 is now performed by Ansible, which is
> using home directory (which causes issue with vintage node BZ1540101). So we
> should probably switch to use home directory instead of /tmp, but we will
> need have an exception for vintage node until we will get rid of it.

We will not support vintage node in 4.3, since it's < 4.2 cluster level.

Comment 4 Martin Perina 2018-07-03 10:20:39 UTC
Closing as deferred as vintage node will not be supported in 4.3