ovirtsdk4.Error: Fault reason is "Operation Failed". Fault detail is "[]". HTTP response code is 400.

[yongshengma]

Hello,

I'm installing a self hosted engine on a ovirt node ng 4.5.5. I chose nfs when specify the storage:

Please specify the storage you would like to use (glusterfs, iscsi, fc, nfs)[nfs]:

However it reports an error and then shows up choosing storage again:

[ INFO ] TASK [ovirt.ovirt.hosted_engine_setup : Activate storage domain] [ ERROR ] ovirtsdk4.Error: Fault reason is "Operation Failed". Fault detail is "[]". HTTP response code is 400. [ ERROR ] fatal: [localhost]: FAILED! => {"changed": false, "msg": "Fault reason is \"Operation Failed\". Fault detail is \"[]\". HTTP response code is 400."} Please specify the storage you would like to use (glusterfs, iscsi, fc, nfs)[nfs]:

Any idea please? Thanks.

Yongsheng

[yongshengma]

This issue still exists. It shows up when hosted-engine --deploy --4 comes to
[ INFO ] TASK [ovirt.ovirt.hosted_engine_setup : Activate storage domain].

I'm pretty sure both DNS and NFS share work correctly. DNS is on the same private network with ovirt host and engine; NFS share is on another network specifically for storage.

All the steps look fine until Please specify the storage you would like to use (glusterfs, iscsi, fc, nfs)[nfs]:. I found the nfs share was mounted correctly when I checked using df -h such as

192.168.20.67:/mnt/data/engine 7.3T 1006G 6.3T 14% /rhev/data-center/mnt/192.168.20.67:_mnt_data_engine

But the deploy got stuck on the error [ ERROR ] ovirtsdk4.Error: Fault reason is "Operation Failed". Fault detail is "[]". HTTP response code is 400. [ ERROR ] fatal: [localhost]: FAILED! => {"changed": false, "msg": "Fault reason is \"Operation Failed\". Fault detail is \"[]\". HTTP response code is 400."}

BTW, I run deploy with root login.

Yongsheng

[agibson684]

I got this issue too. hope it can be fixed soon or a workaround?

[yongshengma]

One more detail of this issue.

When dnf install ovirt-engine-appliance -y, I always got gpg error:
oVirt upstream for CentOS Stream 8 - oVirt 4.5 2.8 MB/s | 2.9 kB 00:00 GPG key at file:///etc/pki/rpm-gpg/RPM-GPG-KEY-oVirt-4.5 (0xFE590CB7) is already installed The GPG keys listed for the "oVirt upstream for CentOS Stream 8 - oVirt 4.5" repository are already installed but they are not correct for this package. Check that the correct key URLs are configured for this repository.. Failing package is: ovirt-engine-appliance-4.5-20231201120252.1.el8.x86_64 GPG Keys are configured as: file:///etc/pki/rpm-gpg/RPM-GPG-KEY-oVirt-4.5 The downloaded packages were saved in cache until the next successful transaction. You can remove cached packages by executing 'dnf clean packages'. Error: GPG check FAILED

So --nogpgcheck has to be appended.

[gocallag]

The GPG is a different unrelated issue IMO - it only applies to Centos 8 - I would suggest creating a new issues based on that.

[gocallag]

This issue still exists. It shows up when hosted-engine --deploy --4 comes to [ INFO ] TASK [ovirt.ovirt.hosted_engine_setup : Activate storage domain].

I'm pretty sure both DNS and NFS share work correctly. DNS is on the same private network with ovirt host and engine; NFS share is on another network specifically for storage.

All the steps look fine until Please specify the storage you would like to use (glusterfs, iscsi, fc, nfs)[nfs]:. I found the nfs share was mounted correctly when I checked using df -h such as

192.168.20.67:/mnt/data/engine 7.3T 1006G 6.3T 14% /rhev/data-center/mnt/192.168.20.67:_mnt_data_engine

But the deploy got stuck on the error [ ERROR ] ovirtsdk4.Error: Fault reason is "Operation Failed". Fault detail is "[]". HTTP response code is 400. [ ERROR ] fatal: [localhost]: FAILED! => {"changed": false, "msg": "Fault reason is \"Operation Failed\". Fault detail is \"[]\". HTTP response code is 400."}

BTW, I run deploy with root login.

Yongsheng

I was able to re-create this issue Http Resp Code 400 with my QNAP acting as a NAS. I got the same issue with a standalone engine trying to add the same NFS mount. but I was able to circumvent the issue by setting 'squash all users' on the QNAP side and making sure the user I was squashing to had access to the 'share' on the QNAP side. I still have this issue with the self-hosted engine and will explore when I get a chance.

[CyberPunkXXX]

I have tried it and still keep getting the error even with Squash all user's set. Any fix?

[gocallag]

I have tried it and still keep getting the error even with Squash all user's set. Any fix?

What does the vdsm.log say? My issue was definitely a problem with my NFS backend. The vdsm.log on your ovirt node will tell you what the problem is (in vague terms). I certainly agree that the error UX leads a lot to be desired.