单进程CPU总是99%，这是正常的吗？
单进程CPU总是99%，这是正常的吗？
提问于 2018-12-19 06:38:31
每当我在Ubuntu上运行我的nodeJS应用程序时，来自"top“命令的mongodb进程显示99%。这是mongodb本身的日志。
2018-12-19T06:27:08.085+0000 I COMMAND [conn4022] command mydb.City command: find { find: "City", filter: { country: { $regex: "^United States$", $options: "i" }, state: { $regex: "^New York$", $options: "i" }, city: { $regex: "^New York$", $options: "i" } }, returnKey: false, showRecordId: false, lsid: { id: UUID("19d70135-fb6f-4729-8e8a-46958c127479") }, $db: "mydb" } planSummary: IXSCAN { country: 1, state: 1, city: 1 } keysExamined:82420 docsExamined:1 cursorExhausted:1 numYields:643 nreturned:1 reslen:301 locks:{ Global: { acquireCount: { r: 644 } }, Database: { acquireCount: { r: 644 } }, Collection: { acquireCount: { r: 644 } } } protocol:op_query 106ms
2018-12-19T06:27:08.190+0000 I COMMAND [conn4021] command mydb.City command: find { find: "City", filter: { country: { $regex: "^United States$", $options: "i" }, state: { $regex: "^New York$", $options: "i" }, city: { $regex: "^New York$", $options: "i" } }, returnKey: false, showRecordId: false, lsid: { id: UUID("19d70135-fb6f-4729-8e8a-46958c127479") }, $db: "mydb" } planSummary: IXSCAN { country: 1, state: 1, city: 1 } keysExamined:82420 docsExamined:1 cursorExhausted:1 numYields:643 nreturned:1 reslen:301 locks:{ Global: { acquireCount: { r: 644 } }, Database: { acquireCount: { r: 644 } }, Collection: { acquireCount: { r: 644 } } } protocol:op_query 104ms
2018-12-19T06:27:08.297+0000 I COMMAND [conn4022] command mydb.City command: find { find: "City", filter: { country: { $regex: "^United States$", $options: "i" }, state: { $regex: "^New York$", $options: "i" }, city: { $regex: "^New York$", $options: "i" } }, returnKey: false, showRecordId: false, lsid: { id: UUID("19d70135-fb6f-4729-8e8a-46958c127479") }, $db: "mydb" } planSummary: IXSCAN { country: 1, state: 1, city: 1 } keysExamined:82420 docsExamined:1 cursorExhausted:1 numYields:643 nreturned:1 reslen:301 locks:{ Global: { acquireCount: { r: 644 } }, Database: { acquireCount: { r: 644 } }, Collection: { acquireCount: { r: 644 } } } protocol:op_query 106ms
2018-12-19T06:27:08.402+0000 I COMMAND [conn4021] command mydb.City command: find { find: "City", filter: { country: { $regex: "^United States$", $options: "i" }, state: { $regex: "^New York$", $options: "i" }, city: { $regex: "^New York$", $options: "i" } }, returnKey: false, showRecordId: false, lsid: { id: UUID("19d70135-fb6f-4729-8e8a-46958c127479") }, $db: "mydb" } planSummary: IXSCAN { country: 1, state: 1, city: 1 } keysExamined:82420 docsExamined:1 cursorExhausted:1 numYields:643 nreturned:1 reslen:301 locks:{ Global: { acquireCount: { r: 644 } }, Database: { acquireCount: { r: 644 } }, Collection: { acquireCount: { r: 644 } } } protocol:op_query 104ms
2018-12-19T06:27:08.506+0000 I COMMAND [conn4022] command mydb.City command: find { find: "City", filter: { country: { $regex: "^United States$", $options: "i" }, state: { $regex: "^New York$", $options: "i" }, city: { $regex: "^Rochester$", $options: "i" } }, returnKey: false, showRecordId: false, lsid: { id: UUID("19d70135-fb6f-4729-8e8a-46958c127479") }, $db: "mydb" } planSummary: IXSCAN { country: 1, state: 1, city: 1 } keysExamined:82420 docsExamined:1 cursorExhausted:1 numYields:643 nreturned:1 reslen:302 locks:{ Global: { acquireCount: { r: 644 } }, Database: { acquireCount: { r: 644 } }, Collection: { acquireCount: { r: 644 } } } protocol:op_query 103ms我的NodeJS应用程序只是在mongodb搜索国家、城市和州的记录。
MongoClient.connect(url, {
useNewUrlParser: true,
reconnectTries: 60,
reconnectInterval: 1000
}, (err, client) => {
if (err) {
console.error(`Connection Error: ${err}`)
return process.exit(1);
}
clientc = client;
client.topology.on('close', function(event) {
console.log(JSON.stringify(event, null, 2));
return process.exit(1);
});
client.topology.on('reconnect', function(event) {
console.log(JSON.stringify(event, null, 2));
});
console.log(new Date().toString() + '>Connection successful.');
db = client.db('mydb');
collectionCity = db.collection('City');
collectionCountry = db.collection('Country');
})
const findRecords = (collection, myobj) => {
return new Promise((resolve, reject) => {
collection.find(myobj).toArray((err, data) => {
if (err) {
console.error(`Cannot find records: ${err}`)
resolve('error');
} else {
console.log(`Record found: ${data.length}`)
resolve(data)
}
})
});
}这是循环函数中搜索国家、州和代码的代码：
let cityquery = {
country: {
'$regex': country_s,
$options: 'i'
},
state: {
'$regex': state_s,
$options: 'i'
},
city: {
'$regex': city_s,
$options: 'i'
}
};
let cityrecord = await findRecords(collectionCity, cityquery);城市收藏索引：
> db.City.getIndexes()
[
{
"v" : 2,
"key" : {
"_id" : 1
},
"name" : "_id_",
"ns" : "mydb.City"
},
{
"v" : 2,
"unique" : true,
"key" : {
"country" : 1,
"state" : 1,
"city" : 1
},
"name" : "country_1_state_1_city_1",
"ns" : "mydb.City"
}
]我不知道是什么让CPU保持99%。我在AWS上使用MongoDBv4.0.3在Ubuntu上运行。
回答 1
Stack Overflow用户
回答已采纳
发布于 2018-12-20 14:45:51
正如我所看到的，您使用了不区分大小写的regexp搜索('i‘选项)。
来自MongoDB文档：
不区分大小写的正则表达式查询通常不能有效地使用索引。$regex实现不支持排序规则，无法利用不区分大小写的索引.
https://docs.mongodb.com/manual/reference/operator/query/regex/#index-use
因此，似乎您的索引根本不被使用，并执行全面扫描集合。
建议1:如果城市名称在数据库中大小写，那么只需删除“i”选项即可。
建议2:可能尝试“$text”搜索https://docs.mongodb.com/manual/reference/operator/query/text/
页面原文内容由Stack Overflow提供。腾讯云小微IT领域专用引擎提供翻译支持
原文链接：
https://stackoverflow.com/questions/53845819
复制相关文章
相似问题
腾讯云开发者
