Ovirt is Hosted Engine and version 4.2 installed a week ago.
OS is Centos7.4 on hosted engine and hosts
Peter Harman – Systems and Safety Cordinator | Homeyer Precision Manufacturing
[Description: C:\Users\gruether\AppData\Local\Temp\Temp1_Homeyer Logo (2).zip\Homeyer
Logo\Homeyer Logo.jpg]
16051 State Hwy 47, Marthasville, MO 63357| E pharman(a)homeyertool.com
|<mailto:pharman@homeyertool.com%20%7C> P 636.433.2244 | F 636.433.5257
From: Nir Soffer <nsoffer(a)redhat.com>
Sent: Thursday, May 24, 2018 3:39 PM
To: Peter Harman <pharman(a)homeyertool.com>; Yedidyah Bar David
<didi(a)redhat.com>
Cc: users(a)ovirt.org; Daniel Erez <derez(a)redhat.com>
Subject: Re: [ovirt-users] Upload Image Error
On Thu, May 24, 2018 at 10:57 PM Peter Harman
<pharman@homeyertool.com<mailto:pharman@homeyertool.com>> wrote:
Ovirt Users,
What version are you running?
I am running into a strange problem with uploading images through the webUI. When I test
the connection on an upload I get “Connection to ovirt-imageio-proxy service has failed.
Make sure the service is installed, configured, and ovirt-engine certificate is registered
as a valid CA in the browser.” I have conducted this operation on several computers using
both chrome and firefox and ensuring the certs were loaded into the browsers.
Are you sure you import the certificate correctly info the browser?
I went to this page:
https://ovirt.org/develop/release-management/features/infra/pki/ to
find cert info and checked both the engine and hosts for the certificates and the identity
of the certificates – everything seemed to match up.
This issue means that the browser refuse to communicate
with the proxy because the proxy certificate does not match
the browser certificates.
Is it possible that you changed engine fqdn and regenerated
engine certificates?
Didi, how can we regenerate all certificates to make sure everything
is configured correctly?
Or verify that the certificates in a host are correct?
Next thing I looked at was the ovirt-imageio-proxy service. I checked it and restarted it
below is a status output from one of the failed operations:
[root@hpm-engine ~]# systemctl status ovirt-imageio-proxy
● ovirt-imageio-proxy.service - oVirt ImageIO Proxy
Loaded: loaded (/usr/lib/systemd/system/ovirt-imageio-proxy.service; enabled; vendor
preset: disabled)
Active: active (running) since Thu 2018-05-24 13:25:38 CDT; 1h 13min ago
Main PID: 21239 (ovirt-imageio-p)
Tasks: 2
CGroup: /system.slice/ovirt-imageio-proxy.service
└─21239 /usr/bin/python /usr/bin/ovirt-imageio-proxy
May 24 13:25:38 hpm-engine.server.local systemd[1]: Starting oVirt ImageIO Proxy...
May 24 13:25:38 hpm-engine.server.local systemd[1]: Started oVirt ImageIO Proxy.
May 24 14:38:02 hpm-engine.server.local ovirt-imageio-proxy[21239]: 127.0.0.1 - -
[24/May/2018 14:38:02] "PUT /tickets/ HTTP/1.1" 200 0
This means proxy is running, and engine is able to communicate
with it. Unfortunately, this does not mean that the browser is able to
communicate with the proxy.
ovirt-imageio-proxy service seems to check out. So, next step was checking out VDSM
process output is below (NOTE: I redacted a bunch of unrelated warnings):
...
Vdsm is not related to proxy connection errors.
Did you know that you can upload using the SDK? It is also much
faster since you can upload directly to the host, instead of via the proxy.
Here is an example:
https://github.com/oVirt/ovirt-engine-sdk/blob/master/sdk/examples/upload...
Nir