Search results

So for this aspect we've built an external IPAM system, based on PHP IPAM and a few other tools. We then have DHCP providing the IPs and keeping the DNS / hostnames updated based on that. We use DHCP relay servers on each site to pass these requests forward. I don't really have much in the way...

We are currently using Ceph, though this decision was based around a desire for tight PMVE integration and a few other requirements, rather than raw IOPS. We're using commodity hardware, with many, many, 300GB disks. Performance is more than adequate.

We've been deploying something similar lately, and what we have ultimately used is a combination of CloudInit, template VMs and Ansible. We then have an exteral IPAM system handling much of the rest.

Is this a plan/thing? It's a feature we would really love to see, but haven't heard of until now :)

Sounds good! Thanks for the reply!

We're trying this now - looking very good so far! Out of interest (and sorry if this has been covered elsewhere!), but was there a reason for going for Nautilus rather than Mimic in Ceph? Keep up the great work! Jon

Very interesting! Thank you! So it's obviously not something specific to the FreeNAS code, since the instance you run that is presenting the storage from a SAN is fine, and OMV seems to perform well either way. I'll have to conduct some tests and see if we can pinpoint where the issue is...

Thanks for the info! 120TB is a bit more than I am playing with - as it's home use I'll be looking at nothing larger than 20TB, and that's if I upgrade things in the future. Is there any consensus on at which point it becomes ill-advised to use images? Jon

Hi all, I run quite a number of PMVE systems in production, with enterprise grade storage and such, and they all work just fine. However, I am currently looking for a solution to use at home, and as the scale of the whole thing is a bit different, I find myself with a few questions. I have an...

Also very interested in this - we're looking at migrating a large Xen estate that resides in two datacentres to Proxmox, and would love to be able to replace our DRBD storage with Ceph as part of this.

Seconded. This would be very useful.

We also see this - RAM use of upto 20GB on a host with 4 x 300GB HDDs. Looking at the documentation above it seems an adjustment is required for us - in Proxmox's deployment of Ceph are we to edit /etc/pve/ceph.conf to include the currently absent cache limit, and then restart the OSDs / Ceph...

OK, So TCPDump seems to suggest that the VMs are constantly trying to do discovery, to find out who has the router address. This would suggest that despite the auto-config having happened at somepoint, the RA packets are not arriving. I'm checking on our TOR switches, to ensure no strangeness...

Hi, Thanks for reviewing that - good to see there are no glaring errors! The auto-assigned IP comes from the upstream provider - we got them to enable this temporarily to assist with diagnostics. All systems are in the same prefix/subnet - the cross host communication is what has me really...

The new Ceph and cluster dashboards are very good! Nice work guys! The upgrade of our cluster went without a hitch. Jon

Hi, Please find the output from the VMs:

Hi, Please find output from the nodes below: Thanks, Jon

Hi, Thanks for the reply - I'll try provide answers to questions here, and output of the commands in a followup. We have 3 hosts that are co-located. Each has 6 NICs, in two bonds, that feed two bridges. We have a LAN bridge (vmbr0 - PMVE GUI, etc) and a WAN bridge (vmbr1). It is the WAN...

Sadly we still have this issue. Do the Proxmox staff have any ideas? Is anyone able to recommend someone we could hire in to investigate? Thanks Jon

This is where our situation differs somewhat. We're doing bridged, but on the WAN side - therefore using a secondary bridge we've added to the Proxmox host.