Recent content by Maxnet

We have also experienced issues with iPXE and large image files on some systems, while it does work fine on others. E.g. Vmware vSphere 5 also comes with a very large ramdisk. When you try to pxe boot that on a problematic system, iPXE/syslinux either fails with an "out of resources" error...

Correct. My own web application logs into Proxmox as root through the API. My application is also responsible for access control. For normal actions like "power on" and "power off" this is easy: 1) User requests the action through my webinterface 2) My webapplication verifies if the user is...

In short: I only want to use Proxmox for its backend, and use my own webinterface as frontend instead of the one that comes with Proxmox. I also would prefer to handle authentication in my own application instead of the user having to type an extra Proxmox password.

Chrome complains that one is old. And if you wait too long to click away that message and the certificate errors, you get: What about a pure Javascript client like noVNC?

Is it possible for root to impersonate a less powerful user through the API, without knowing the user's password?

When I use the API (nodes/$node/qemu/$vmid/vncproxy) to make a console connection, I receive a ticket back. I assume I need to use that as VNC password? Can this token only be used to access the console of that particular VM, or can it be (ab)used for other management actions as well? Just...

Console is also unusable here under Kubuntu 11.10 with Firefox and Chrome. With Firefox, the console does start but Java consumes 100% CPU. Firefox also complains about the JavaSCRIPT consuming too much resources, and suggests to kill it. If you do so, the entire browser will crash. Chrome...

Was just wondering whether it sends a negligible amount, e.g. just a packet a second for each node to see if all nodes are still up. If that's the case, I wouldn't mind if servers that are not part of the cluster receive the traffic as well. "a lot" meaning things like fancy distributed file...

The Wiki mentions: Was wondering if I could get around having to mess with the switch configuration, by telling Corosync to use the broadcast address for its communication instead of multicast. I recall it has an option for that. Does the Proxmox cluster stuff generate a lot of traffic, or...

On the mailinglist it was suggested to change the IO scheduler to "deadline" on the host system. And it indeed improves random IO a lot. After doing on the host system: echo deadline > /sys/block/[disk device]/queue/scheduler Virtio: Version 1.03c ------Sequential Output------...

I'm having the problem that the performance of VirtIO is nowhere near native performance, and even slower than IDE. Especially with random IO (seeks). Occurs with both a raw image on the standard "directory" storage, as well as LVM on a seperate drive. Tested with both Ubuntu 9.04 as the new...

I'm having the same problem with both 1.3 and 1.4b2 on a system with a Nvidia SATA chipset. As a work-around I just installed Debian, and added the proxmox packages.