7.9 KiB
7.9 KiB
Deploying a VM with QSFS
Table of Contents
Prerequisites
First, make sure that you have your client prepared.
Code Example
import { FilterOptions, MachinesModel, QSFSZDBSModel } from "../src";
import { config, getClient } from "./client_loader";
import { log } from "./utils";
async function main() {
const grid3 = await getClient();
const qsfs_name = "wed2710q1";
const machines_name = "wed2710t1";
const vmQueryOptions: FilterOptions = {
cru: 1,
mru: 1, // GB
sru: 1,
availableFor: grid3.twinId,
farmId: 1,
};
const qsfsQueryOptions: FilterOptions = {
hru: 6,
availableFor: grid3.twinId,
farmId: 1,
};
const qsfsNodes = [];
const allNodes = await grid3.capacity.filterNodes(qsfsQueryOptions);
if (allNodes.length >= 2) {
qsfsNodes.push(+allNodes[0].nodeId, +allNodes[1].nodeId);
} else {
throw Error("Couldn't find nodes for qsfs");
}
const vmNode = +(await grid3.capacity.filterNodes(vmQueryOptions))[0].nodeId;
const qsfs: QSFSZDBSModel = {
name: qsfs_name,
count: 8,
node_ids: qsfsNodes,
password: "mypassword",
disk_size: 1,
description: "my qsfs test",
metadata: "",
};
const vms: MachinesModel = {
name: machines_name,
network: {
name: "wed2710n1",
ip_range: "10.201.0.0/16",
},
machines: [
{
name: "wed2710v1",
node_id: vmNode,
disks: [
{
name: "wed2710d1",
size: 1,
mountpoint: "/mydisk",
},
],
qsfs_disks: [
{
qsfs_zdbs_name: qsfs_name,
name: "wed2710d2",
minimal_shards: 2,
expected_shards: 4,
encryption_key: "hamada",
prefix: "hamada",
cache: 1,
mountpoint: "/myqsfsdisk",
},
],
public_ip: false,
public_ip6: false,
planetary: true,
cpu: 1,
memory: 1024,
rootfs_size: 0,
flist: "https://hub.grid.tf/tf-official-apps/base:latest.flist",
entrypoint: "/sbin/zinit init",
env: {
SSH_KEY: config.ssh_key,
},
},
],
metadata: "{'testVMs': true}",
description: "test deploying VMs via ts grid3 client",
};
async function cancel(grid3) {
// delete
const d = await grid3.machines.delete({ name: machines_name });
log(d);
const r = await grid3.qsfs_zdbs.delete({ name: qsfs_name });
log(r);
}
//deploy qsfs
const res = await grid3.qsfs_zdbs.deploy(qsfs);
log(">>>>>>>>>>>>>>>QSFS backend has been created<<<<<<<<<<<<<<<");
log(res);
const vm_res = await grid3.machines.deploy(vms);
log(">>>>>>>>>>>>>>>vm has been created<<<<<<<<<<<<<<<");
log(vm_res);
// get the deployment
const l = await grid3.machines.getObj(vms.name);
log(">>>>>>>>>>>>>>>Deployment result<<<<<<<<<<<<<<<");
log(l);
// await cancel(grid3);
await grid3.disconnect();
}
main();
Detailed Explanation
We present a detailed explanation of the example shown above.
Getting the Client
const grid3 = getClient();
Preparing QSFS
const qsfs_name = "wed2710q1";
const machines_name = "wed2710t1";
We prepare here some names to use across the client for the QSFS and the machines project
const qsfsQueryOptions: FilterOptions = {
hru: 6,
availableFor: grid3.twinId,
farmId: 1,
};
const qsfsNodes = [];
const allNodes = await grid3.capacity.filterNodes(qsfsQueryOptions);
if (allNodes.length >= 2) {
qsfsNodes.push(+allNodes[0].nodeId, +allNodes[1].nodeId);
} else {
throw Error("Couldn't find nodes for qsfs");
}
const vmNode = +(await grid3.capacity.filterNodes(vmQueryOptions))[0].nodeId;
const qsfs: QSFSZDBSModel = {
name: qsfs_name,
count: 8,
node_ids: qsfsNodes,
password: "mypassword",
disk_size: 1,
description: "my qsfs test",
metadata: "",
};
const res = await grid3.qsfs_zdbs.deploy(qsfs);
log(">>>>>>>>>>>>>>>QSFS backend has been created<<<<<<<<<<<<<<<");
log(res);
Here we deploy 8 ZDBs on nodes 2,3 with password mypassword, all of them having disk size of 10GB
Deploying a VM with QSFS
const vmQueryOptions: FilterOptions = {
cru: 1,
mru: 1, // GB
sru: 1,
availableFor: grid3.twinId,
farmId: 1,
};
const vmNode = +(await grid3.capacity.filterNodes(vmQueryOptions))[0].nodeId;
// deploy vms
const vms: MachinesModel = {
name: machines_name,
network: {
name: "wed2710n1",
ip_range: "10.201.0.0/16",
},
machines: [
{
name: "wed2710v1",
node_id: vmNode,
disks: [
{
name: "wed2710d1",
size: 1,
mountpoint: "/mydisk",
},
],
qsfs_disks: [
{
qsfs_zdbs_name: qsfs_name,
name: "wed2710d2",
minimal_shards: 2,
expected_shards: 4,
encryption_key: "hamada",
prefix: "hamada",
cache: 1,
mountpoint: "/myqsfsdisk",
},
],
public_ip: false,
public_ip6: false,
planetary: true,
cpu: 1,
memory: 1024,
rootfs_size: 0,
flist: "https://hub.grid.tf/tf-official-apps/base:latest.flist",
entrypoint: "/sbin/zinit init",
env: {
SSH_KEY: config.ssh_key,
},
},
],
metadata: "{'testVMs': true}",
description: "test deploying VMs via ts grid3 client",
};
const vm_res = await grid3.machines.deploy(vms);
log(">>>>>>>>>>>>>>>vm has been created<<<<<<<<<<<<<<<");
log(vm_res);
So this deployment is almost similiar to what we have in the vm deployment section. We only have a new section qsfs_disks
qsfs_disks: [{
qsfs_zdbs_name: qsfs_name,
name: "wed2710d2",
minimal_shards: 2,
expected_shards: 4,
encryption_key: "hamada",
prefix: "hamada",
cache: 1,
mountpoint: "/myqsfsdisk"
}],
qsfs_disks is a list, representing all of the QSFS disks used within that VM.
qsfs_zdbs_name: that's the backend ZDBs we defined in the beginningexpected_shards: how many ZDBs that QSFS should be working withminimal_shards: the minimal possible amount of ZDBs to recover the data with when losing disks e.g due to failuremountpoint: where it will be mounted on the VM/myqsfsdisk
Getting the Deployment Information
const l = await grid3.machines.getObj(vms.name);
log(l);
Deleting a Deployment
// delete
const d = await grid3.machines.delete({ name: machines_name });
log(d);
const r = await grid3.qsfs_zdbs.delete({ name: qsfs_name });
log(r);