Search results

Since fall 2015 the default compression algorithm in ZOL is LZ4 and since choosing compression=on means activate compression using default algorithm then your pools are using LZ4 -> http://open-zfs.org/wiki/Performance_tuning#Compression

What results you get inside the VM is of no concern in regards to getting the numbers from your disk system on the host.

Rerun this test with direct=0 (ZFS does not support direct io)

I am pretty sure this is a ZOL specific problem! Try using below as input file while running fio on your system # This job file tries to mimic the Intel IOMeter File Server Access Pattern [global] description=Emulation of Intel IOmeter File Server Access Pattern [iometer]...

What I bad piece of advice!!! If the brakes is not working properly on your car so that your speed is limited would your advice then be to disconnect the brakes?

Hard to help you without access to the contents of the proper log files. Try run you preferred browser in web developer mode and see what is showing in the console window.

The article misses this in the 'location / { ..' section: add_header X-Frame-Options SAMEORIGIN;

Try what I wrote here: https://forum.proxmox.com/threads/vlan-to-kvm-guest-using-ovs.38167/#post-188644 I suspect there is something wrong with your router/router configuration.

You are certain that your router (VyOS) is not blocking ping? I would create a VM on 10.0.10.0/24 and test whether pings succeeds here.

Add 8021q to /etc/modules and reboot. I think OVS might fail to configure vlan when the module is missing when initial configuring the network.

Looks ok. Is this module loaded into the kernel - 8021q ? (lsmod | grep 8021q)

What does ip addr show? Is this package installed? openvswitch-switch Did you reboot the server after making the network changes?

Try to change the above line: ovs_options tag=88 To: ovs_options tag=88 vlan_mode=native-untagged

No.

This is wrong. RAID0 is not anywhere near the same as HBA mode!

If HA is a must your only option for the combination SAN and Proxmox is LVM.

If you install and run multipathd over the exposed disks via FC you could use ZFSPoolPlugin: https://pve.proxmox.com/wiki/Storage:_ZFS. But since this plugin does not guard against shared access to the pool I would not recommend HA.

Use ip_hash as distribution algorithm. See: http://nginx.org/en/docs/http/load_balancing.html (Session persistence)

A bit flip on the disk/storage some time in the future cannot be detected at a specific point in time but must be detected on a regularly basis.

The solution to your problems is to enable sticky sessions. Your problem is that your novnc some times makes connection to another node than the one you are currently connected to.