3vang

Figured it out. For anyone googling and getting this post, issue/fix was allowing Forged transmits and Promiscuous mode in ESXi vSwitch security settings.

Decided to virtualized one of my unraid in ESXi 7.0. USB and HBA passthrough. Everything seems to work fine but only my Docker containers have no internet/network access. eth0 is a VMXNET3 virtual NIC. Enable bonding: No. Enable bridging: Yes. IPv4 Only. DNS: 1.1.1.1. VLANs: No. Gateway: 10.1.1.1. IPv4 default 10.1.1.1 via br0 1 IPv4 10.1.1.0/24 br0 IPv4 10.41.41.2 wg0 1 IPv4 10.41.41.3 wg0 1 IPv4 10.41.41.4 wg0 1 IPv4 172.17.0.0/16 docker0 1 IPv6 ::1 lo 256 Any help is greatly appreciated.

I built my server on a Lenovo ST250. It has SAS 12Gbps backplane and cables. I dropped in a SAS 9300-8i HBA, but have been using it with 6TB shucked drives. Recently decided to add in two 12Gbps 1TB SAS SSD as cache pool. The SATA spinning drives and SAS SSD all run off the 9300-8i. Anyone knows if the HBA is smart enough to run my array with SATA3 but run the cache pool with SAS3? Will my cache pool run at 12Gbps? Any ways to actually benchmark the cache pool to verify its running at 12Gbps?

You are correct and I just wanted to provide an update. Change board and CPU to a 3.7 Ghz, SMB read and write are 1100 MB/s now.

Yes, the low 2.1 Ghz maybe cause, but this cpu is most definitely multithreaded if OS and application can utilizes.

Is this the correct way to connect to disk share directly, \\host-ip\mnt\cache\share-name (in Windows) and host-ip:/mnt/cache/share-name (in Linux)? using cache/disk as opposed to mnt/user.

I recently upgrade my network and equipment to a 10GbE network. I know I can fully saturate the link because I can read from my cache at over 1100 MB/s, but I can only write at about 600 MB/s. I have two Samsung PM981 1TB NVme SSD (OEM 970 EVO equivalent) in RAID1 with btrfs. I don't see why the write speed can't be fully saturated over the network like the read speeds. I an no expert on btrfs, so can it be a limitation of that file system, or certain settings I am missing, or it is truly my hardware limitation? Unraid is running on a Supermicro X10SDV-TLN4F, Xeon D-1541, 64 GB DDR4 RAM at 2133.

"Device Contents Emulated Click to spin down device" That is the disk being rebuild. I am assuming that is normal status for the replaced disk that is currently rebuilding.

No errors on Main. No VMs, Docker or IO activity at all. All disks are onboard SATA on a Supermicro X10SDV-TLN4F board, with 2 NVMe as cache. No SMART warnings.

All disks are 5400 RPM, but there's no way that is the reason for 15 MB/s. I recently added those parity drives and average speed were 150 MB/s. I did not preclear as I'm new and am not even sure that that is. The 6TB rebuilding disk came from a Freenas server. Disk was wipe by Freenas before pulling and moving it to unraid.

New to unraid. I'm swapping out a data disk and I am getting between 15-20 MB/sec speed. ETA of over 3 to 4 days. Seems like that's not normal. Can someone help me figure out what is going on? unraid-diagnostics-20200722-1541.zip