|
OSP and RHEL possible Bug
Hello,
I had these problems earlier with OSP14 and RHEL 7.6; Code:
[root@localhost network-scripts]# systemctl status network -lCode:
[root@localhost log]# Code:
[root@amer network-scripts]# Quote:
the PC was accidentally disconnected and caused the last punch of errors, which caused crashdump, and I think it's the RabbitMQ heartbeat as it showed in my logs, and it seems that there is a problem when the server disconnects that causes RabbitMQ to halt and crash the system with Kernel loop.:( |
It’s not clear to me what’s the problem. You can’t start your network, but have you tried to find more log messages in the journal? There is an unlimited number of possible causes for network failures. Also, is the network really down?
Do the “CPU stuck” messages cause any real problem or are they just annoying entries in the log? Where are these messages generated, on a compute node, controller or VM? In which log do you see the message in red font? You allude to problems with RabbitMQ. Can you elaborate? You say that “the server disconnects”. Which server disconnects from what? Finally, what is the significance of the PC you mention in the last paragraph? |
Quote:
https://access.redhat.com/solutions/2073603 Those messages typically revolve around high system load and/or insufficient resources for the RHEL system to function properly. You say nothing about what's running on that machine, if this is a new installation/problem or an existing one, etc. Red Hat support can easily analyze an SOS report for you, to get you more details. |
Quote:
https://bugs.launchpad.net/nova/+bug/1813446 https://bugzilla.redhat.com/show_bug.cgi?id=1666598 I have the Kdump crash image and ABRT tool snapshot for reference, also I want to say that the firmware of VMs is not compatible with the new Kernels of Linux. |
You must admit that your problem descriptions are a bit lacking. For example, the Nova bug description starts like this:
Quote:
The Bugzilla description mentions something about bridges not being set up correctly without providing details. In this Linuxquestions thread you say that systemctl status network reports an error, but you are not clear which computer’s or VM’s network has problems, nor if they are really problems.You then mention the CPU stuck messages without clarifying how they are related to the network problem, before jumping to RabbitMQ problems, again without saying how that is supposed to be related. I know a little bit about OpenStack and would be delighted to be of help, but at this point, I don’t even know what your problem is. Is your VM host crashing? Does the VM have network connectivity problems? Anything else? Without a clear problem statement (Note: random log messages are not problems), nobody can help you. |
Quote:
as you can see, I was most of the time talking to no body in these channels, they are supposed to answer, and more than once, well also you don't see a Guru in my name at all. and I found out later that there is people working in both sites, and I reporting to them the same problem, first that they will say, this site is not responsible for this, try the other site, and on the other site they say, this is not a bug, blah blah. my VM crashed, I have a snapshot of the crash, and Horizon Gui gave, report this bug to launchpad, launchpad member said that this is not launchpad's this is bugzilla's, bugzilla's did respond by the same man that this is not a bug without so many replies and decided to close saying that this not a Bug, no explanation at all. (Duhhh) thanks for the reply https://ibb.co/ZzQ8Bfb https://ibb.co/MV5RTV2 https://ibb.co/4SvwRVg |
Quote:
Quote:
12GB of RAM and 8 cores is plenty to run RHEL, but you still don't say what ELSE is running on that VM, or what kind of VM you're running. The phrase "the firmware of VMs is not compatible with new kernels of Linux" also makes little sense, since there are many articles on the VMWare site that have resolutions listed, specifically for such issues. Again, if you're actually PAYING FOR VMWare, you can download and apply those patches. |
Quote:
Quote:
|
Quote:
Thanks |
Quote:
Quote:
Quote:
Quote:
Can you confirm that your problem is a VM in OpenStack that crashes, possibly due to CPU lockups? If so, your compute host might not have enough power to run this VM (https://ask.openstack.org/en/questio...ft-lockup-cpu/). Can you share the compute node's specs, and how much load is on that node? As an aside, RabbitMQ is required to launch an instance, but it is not required for an instance to continue running once its launch was successful. I therefore guess that your RabbitMQ problems are separate from the VM crash, but if your compute node is under-powered, this may well cause RabbitMQ timeouts. |
Quote:
I have deleted the VM with the logs, and I started a fresh installation with Minimal conf in the answer file. I had an All in one installation, not multi-node. Best regards |
Quote:
Does OSP support single-host deployments? I would guess not, which might explain the reluctance of Red Hat Support to help you. |
Quote:
anyway thanks for your swift reply |
Try Tripleo instead of OSP.
|
describing the solution
Quote:
|
| All times are GMT -5. The time now is 10:12 PM. |