Search results

I don't suppose anyone has any experiences, or know if the additional investment in NVMe is worthwhile when used with Ceph? Clearly NVMe is faster, but Ceph has non-negligible overhead which may make this moot (but I don't have any benchmarks to know), and since I've never used NVMe, its...

Sounds similar to https://forum.proxmox.com/threads/proxmox-4-x-is-killing-my-ssds.29732/

We're currently building out a new 3-node datacenter that uses shared Ceph + Proxmox nodes. We've done this in the past using a classic LSI SAS3 RAID controller attached to a SAS3 backplane extender using all Intel DC S3700 SSDs (in JBOD mode) and have had good success. Now, with the new...

What network architecture/speed? What VM disk subsystem? What filesystem for OSDs? What are you using to benchmark in the VM?

I'm a little hesitant to try this since I'm pretty sure I won't have network connectivity once I purge openvswitch to then do the re-installation of openvswitch. All interfaces are bound to the bridge ... upgrading to OVS 2.5 did require a system reboot to bring the network back up, so it...

We've been using the same network configuration for about 2 years now with no changes, we use Open vSwitch. This issue appears to be intermittent, but sometimes on reboot, the bridge does not come up with the requested MTU size of 8996, instead it uses the default of 1500. This also means our...

Hmm, we haven't hit this bug ... yet ... on our cluster nodes with SuperMicro X10DRI-O and 2x Xeon E5-2630v3 ... but we also fairly recently did the migration from Proxmox v3 to v4, and our load isn't ramped up either. But I did notice that there is a maintained ppa for ubuntu kernels...

You list a 12Gbps SAS controller, but the chassis has a 6Gbps SAS backplane. They changed the connector between 12Gbps and 6Gbps. I'm a huge supermicro fan ... you're not going to be happy when you try to put that solution together :) I should also mention that JBOD is not the same as RAID0...

Sorry for the confusion, my reply was a question not a statement. I was planning on testing the upgrade to OVS 2.5 soon and ran across this thread, so I was trying to determine how severe the issue is before proceeding in order to try to save myself some stress. I haven't personally hit this...

So even after a reboot any "ovs_type OVSIntPort" entries in /etc/network/interfaces won't come back online? I can handle a temporary outage during an upgrade, not a permanent outage. Any workarounds for starting at boot properly?

Well, ring and star topology are the same thing when you have 3 nodes :) At 4 nodes, the topologies diverge, whereas each node is always connected to 2 other nodes with ring topology, but with star topology it would be connected to every other node. Star topology really doesn't scale easily...

Even for multiple connections, between 2 separate hosts, will only use 1 physical link in most instances, so even if you have 100 connections, you'll be limited to 1Gbps. Linux can do balance-rr for outbound which would alleviate this concern, but the switch will do source or destination...

LACP bonding probably doesn't work the way you think it does anyhow. Typically switches will use a destination hash to determine what physical port to send traffic out on. So if you have 8 physical links, and 3 servers, you'll use at most 2 of those physical links for communicating between...

Just curious if Open vSwitch 2.5 (the current LTS) is on the roadmap for PVE 4.2? It would definitely solve some issues for me (like rstp and multicast snooping). Thanks! -Brad

Curious to see if Open vSwitch 2.5 testing went ok. I don't see it in the pvetest repo yet, so perhaps it didn't go so well. Thanks! -Brad

Looks like OVS 2.5 was released about a week ago, and is indeed an LTS release :) http://openvswitch.org/releases/NEWS-2.5.0 Here's hoping we see OVS 2.5 in proxmox in the near future.

I've been trying to get this topology to work: http://pve.proxmox.com/wiki/Open_vSwitch#Example_4:_Spanning_Tree_.28STP.29_-_1Gbps_uplink.2C_10Gbps_interconnect However, I'm having issues where the proxmox to proxmox 10G interconnects aren't showing up as STP_BLOCK on all interfaces that it...

If your ISP has MAC restrictions, then it probably doesn't make sense for your physical ethernet interface to be part of a bridge at all. You should just assign your ip address to the physical ethernet interface. With MAC restrictions, your VMs wouldn't be able to be on that vlan either...

sure, if you're mac restricted at your provider's switch, that could cause major issues. Also, it's possible your vlan_mode would be more appropriately 'access' like: ovs_options tag=1 vlan_mode=access for your eth0 definition

Can you elaborate on what way it is not working? Are all the interfaces visible in ifconfig -a? Also, instead of using 2 ovs_options lines, you could combine it into one, perhaps having 2 entries is causing only one to be used: ovs_options tag=1 vlan_mode=native-untagged