Search results

Hi there, thank you for all the hard work you are putting into Proxmox! I've been a happy user for 3 years now. I've even managed to convince my previous employer to pay for a subscription for our 5 servers. :D I am very sorry but now I have to express my complaint about the vertical menu...

So here it is: https://github.com/ScOut3R/cd-proxmox-ansible-vm-deployment It was created ages ago and I was still a rookie with Ansible so forgive me for anything stupid you might find there. I have removed any confidential information (at least I hope so), but our whole infrastructure has...

Thank you very much for making this reply on your holiday! I would be very interested in that howto if you would have time to make it sometime. Enjoy your holiday! :)

Well, I use a backup system from the outside. I found the integrated backup solution too slow for us (we are running KVM based VMs only on Ceph storage). So what I do is I have a dedicated backup server, ZFS on it, and using http://www.dirvish.org to do the hard work. I was using dirvish before...

Yeah that can be done with ansible or any other config management framework. I will scan through the module and clean out any private stuff and upload it to github. If you are interested in a more robust way of deploying new VMs you might want to take a look at http://fai-project.org and...

Create a VM and convert it to a template. ;) I have an in-house ansible module which uses the Proxmox API to create a VM from this template and make some configurations. If you are interested I can share it with you.

First of all thank you for the amazing work! This update is huge, but the ones before this are not a tiny bit inferior! :) I would like to clarify my planned upgrade method from 3.4 to 4.0 with you, if you don't mind. I am running a 5 node cluster, running all KVM based VMs with Ceph backend...

Thanks again Dietmar for the quick answer! As it seems I have misunderstood your answer or my OP question wasn't clear enough. My "vms" attribute contained the current members, appended by the new ID and this was wrong. Only the new ID(s) should be appended. I missed the "delete" flag. :)

Just one more question please. What does the "pve-vmid-list" format exactly mean? It says that it is a string so I tried to send the attribute as a concatenated string, the ids seperated but a comma and space but it does not work.

Thank you Dietmar!

Hi, The usage of the PUT /api2/json/pools/{poolid} API endpoint is not clear to me. I would like to add a newly created VM to a specific pool but I am not sure about the content of the "vms" parameter. Should it contain only the given VM ID or the list of the Pool members appended by the given...

Thanks! That solved my issue!

Hi, I am running a Ceph Hammer RBD backend for Proxmox and I tried to do a storage migration from one pool to another. The progression goes up to 100% then just loops there forever with "busy: false". Please see the attached log output. I haven't experienced this kind of error so far. Does...

For the sake of clarity, I am running 1 OSD per HDD. I have two different node setups. One with 6 OSDs, 3 TB each, and one with 4 OSDs, 1 TB each. From the 3 TB node I have 3, that means 18 x 3 TB OSD in total. From the 1 TB node I have 5, that means 20 x 1 TB OSD in total. Originally, the...

I updated my 8 node cluster today without any trouble. It was seamless and relatively fast. Now my cluster is moving data to compensate for the CRUSH bug. :) I don't experience any performance improvements but the CRUSH bugfix is a big hit for us because we have different OSD (HDD) sizes and the...

Thanks!

Hi, yes, I was careless, I meant 0.94.1. :) I would like to update because of the CRUSHMAP fix which isn't in Giant and I think it's affecting my cluster. I just afraid to upgrade because I'm not sure how the KVM stack would handle the new version and I don't have a test infrastructure at hand...

Havint the same issue since running a newer kernel version than 3.10.0-1. Using the ipxe ISO is an acceptable workaround but there's definitely something broken in Proxmox.

Hi, I'm running the latest Proxmox 3.4 from the subscription repo, backed by a Ceph cluster running Giant. Hammer was recently released and I'm planning to upgrade my Ceph cluster but I would like to ask around if anyone has Proxmox running with Hammer already?

I've upgraded one of the nodes with the packages from pvetest and migration to that node seems to be fine for now. Anyway, could unsychnorized system clocks in the VMs cause this migration issue?