Why a disk/SR/VDI performance graph is not available on stats page for host/VM?
Thanks
Posts made by Paolo
-
Disk performance on Stats
-
RE: Suggestions for new servers
I've done some benchmarks with my new servers and want to share results with you.
Server:
- CPU: 2 x Intel Gold 5317
- RAM: 512GB DDR4-3200
- XCP-ng: 8.3-beta2
fio parameters common to all tests:
--direct=1 --rw=randwrite --filename=/mnt/md0/test.io --size=50G --ioengine=libaio --iodepth=64 --time_based --numjobs=4 --bs=32K --runtime=60 --eta-newline=10VM Debian: 4 vCPU, 4 GB Memory, tools installed
VM Windows: 2/4 vCPU, 4 GB Memory, tools installedResults:
- First line: Troughtput/Bandwidth (MiB/s)
- Second line: IOPS: KIOPS (only linux)
Considerations:
- On bare metal I get full disk performance: approx. double read speed due to RAID1.
- On VM the bandwidth and IOPS are approx 20% of bare metal values
- On VM the bottleneck is tapdisk process (CPU at 100%) and can handle approx 1900 MB/s
-
RE: XCP-ng 8.3 beta 🚀
I suppose i can install new servers with xcp-ng-8.3.0-beta2-test3.iso and then update to rc/stable release with yum update.
Am i correct?Regards
Paolo -
RE: Suggestions for new servers
My new servers will be available in the next week. The full specs are:
- 2 x CPU Xeon Gold 5317
- 384 GB DDR4-3200 RAM
- 2 x 10Gb Intel X550 cards
- Adaptec SmartRAID 3102E-81
- 2 x 480 GB SDD
- 2 x 12.8 TB NVMe
Planned setup (KISS):
- XCP installed on 2x480 GB SDD in RAID1 with HW controller
- SR on local storage using NVMe units with SW RAID1
- no HA
- single-host pools with VM replication between hosts/pools
- backup on external NFS file server
- networking 10 Gbe (xcp-ng hosts, backup file server)
Questions / your opinion about:
- Is 8.3 version suitable for production use?
- How can i download beta2 ISO?
- I'm thinking to use ZFS/RAID1 instead of mdadm/EXT4 for local storage (with at least 16 GB on Dom0). Is a good choice?
- bonded links (2 x 10Gb) for xcp-ng server and backup file server (with the switch) are useful or the max. backup speed remain 10 Gb?
- any other suggestion?
Thanks
-
RE: Suggestions for new servers
Tanks a lot @olivierlambert and @planedrop for your answers.
IMHO with NVMe (PCIe devices) RAID1 / RAID10 configuration the mdadm overhead on the CPU should be negligible, and also reliability should be very good.
Anyway I'll investigate the cost of hardware RAID controller. -
Suggestions for new servers
I'm planning to buy new servers with NVMe (U.2) based local SR (and install xcp-ng 8.3).
I've some questions:
- can i use a single software (mdadm) RAID1 or RAID10 volume for both OS and SR?
- Is better to use two distinct (mdadm) volumes for OS (2xSSD RAID1) and SR (2xNVMe RAID1)?
- Is VROC an option (I think no because it is a software raid)?
- Is hw controller an option to evaluate?
- Is there any size limit (I'm thinking to use 2 x 7.68 TB U.2 NVMe Samsung PM9A3 DC Series)?
- I will opt for a CPU (2 x Intel Gold 5317) with good single-thread performance to maximize tapdisk throughput. Is it a good idea?
- Other parts: Intel X550T2 2x10 gbe, 256 GB DDR4-3200 RAM, redundant power supply
Any other suggestion about storage for a new server?
Thanks
-
RE: Not enough server memory when start VM
Hi Olivier,
The host is a single host (no other hosts in the pool) with local storage (EXT) only.
I've done some other tests:-
Tried to detach disks, re-create VM, re-attach disks: no success
-
the VM starts if memory is reduced from 16GB to 10GB
-
I've created a new VM (W2019 template) and the behaviour is the same: with 16GB memory the error is "not enought memory"; with 10GB the VM starts
-
Starting one of the 10GB memory VM, no other VM can be started (also with only 1 GB memory)
It's like the host has lost 15GB of free memory...
SOLVED: Restarting Toolstack the problem is solved: now i can start 16GB VM and another 9GB VM to use all available memory (25,5 GB).
I'd be curious to know what happened...
-
-
RE: Not enough server memory when start VM
[11:40 host1 ~]# xe host-list params=memory-free memory-free ( RO) : 27513360384
[11:40 host1 ~]# xe vm-compute-maximum-memory uuid=bc3a1a8e-1e64-fd0b-ecc8-ae8e1bd7a6e0 total=`xe host-list params=memory-free --minimal` 27292336128
Thanks for your help...
& Merry Christmas to all XCP-ng community! -
RE: Not enough server memory when start VM
@olivierlambert Also from command line the VM cannot be started:
[17:50 host1 ~]# xe vm-start uuid=bc3a1a8e-1e64-fd0b-ecc8-ae8e1bd7a6e0 There are no suitable hosts to start this VM on. The following table provides per-host reasons for why the VM could not be started: host1 : Cannot start here [Not enough free memory] There were no servers available to complete the specified operation.
-
Not enough server memory when start VM
xcp-ng 8.2
host server total memory: 64 GB
host server available memory: 25.5 GB
Already running VMs: 7
Dom0 memory: 8 GB
Dom0 top info:KiB Mem : 8110196 total, 5590192 free, 550848 used, 1969156 buff/cache KiB Swap: 1048572 total, 1048572 free, 0 used. 7355948 avail Mem
VM memory parameters
[11:50 host1 ~]# xe vm-param-list uuid=bc3a1a8e-1e64-fd0b-ecc8-ae8e1bd7a6e0 | grep " memory-" memory-actual ( RO): 0 memory-target ( RO): 0 memory-overhead ( RO): 140509184 memory-static-max ( RW): 17179869184 memory-dynamic-max ( RW): 17179869184 memory-dynamic-min ( RW): 17179869184 memory-static-min ( RW): 1073741824
When starting this VM with 16GB RAM (fixed) using with XCP-ng Center the following error is displayed:
"not enough server memory is available to perform this operation"When starting the same VM with XOA the error is NO_HOSTS_AVAILABLE.
I've tried to decrease the VM memory to 8 GB but the error persists.
I think there is a lot of available memory...
What i should check?Thanks.