Product introduction--X0

Comprehensively analyze and restore the real transaction scenario from the seven dimensions of database dimension, connection dimension, table dimension, SQL dimension, transaction dimension, cluster SQL dimension and cluster transaction dimension, and directly locate the scenario when the fault occurs; Then the root cause of the fault is analyzed through the return code.

Product function

Performance monitoring

Resource monitoring

Node monitoring
- 01
- 02
- 03
- 04
- 05
- 06
- 07
- 08
TPS

Number of transactions per second (number of transactions), set threshold alarm for TPS.

RSPT

Average response time of SQL statements in the database, and set threshold alarm for RSPT.

Response rate

The response proportion of the database and the response rate threshold alarm.

Output rate

The normal response proportion of the database can represent the service capacity of the database.

Proportion of abnormal sqlcode

The percentage of SQL statements that failed to execute in the total executed SQL.

Rollback rate

The ratio of transactions executing rollback to total transactions.

Slow SQL statement

The database executes the slowest SQL statement of TOP10.

Uncommitted transactions

The total number of uncommitted transactions in the current database.
- 01
- 02
- 03
- 04
- 05
“hang”Status monitoring

The database is not responding.

“down”Status monitoring

The database is down or the application has not been requested for a long time.

“Bump”Status monitoring

Database response time and its instability.

“Inferior”Status monitoring

The draw response time of the database tends to slow down.

“Slow”Status monitoring

The average response time of the database is greater than 1s.
4
- 01
- 02
- 03
- 04
- 05
SQLCODEclassification

Classify and monitor the return codes of failed SQL statements.

Affect row count monitoring

The number of records in the SQL statement operation table.

High risk statement monitoring

DDL statements, such as create, alter,...

Channel not submitted for monitoring

The application under the same connection has not executed commit / rollback for a long time.

Channel not responding to monitoring

The database has not answered for a long time under the same connection.

1
- 01
- 02
- 03
- 04
- 05
- 06
- 07
- 08
- 09
- 10
TPS low diagnosis

Determine whether the low TPS is caused by the database.

Sqlcode classification analysis

Classify and analyze the causes of SQL statement execution failure.

Worst performing statement types

Analyze the top 10 statement types that are slow to execute and affect the database.

Worst performing statement analysis

Analyze whether the statement type of slow executing statements is slow.

Channel uncommitted diagnosis

Analyze and count uncommitted transactions and similar transactions.

Large transaction diagnosis

The addition, deletion and modification statements in a transaction affect the transaction analysis with a large number of rows.

High frequency rollback transaction diagnosis

Analyze the transaction types with the highest rollback rate of top 5.

Application anomaly diagnosis

After database upgrade or migration, you can query the missing application IP and new application IP.

Hot spot data diagnosis

Analyze hot data queued application transactions.

Supercomputing diagnosis of large data sets

Analyze supercomputing statements that operate on large data sets
4
5
6

- 01
- 02
Promotion Escort

Determine whether there are promotion activities and analyze them.

Upgrade Escort

Determine whether the application is upgraded, monitor the resource changes and performance of the database when the application is upgraded, and evaluate whether the upgrade is risky.
3
4
5
6

- 01
- 02
Security audit

Identify, record, store and analyze relevant information of safety related activities in the system.

Development audit

Identify, record, store and analyze relevant information related to program development.
3
4
5
6

monitor

Abnormal diagnosis

Campaign escort

Audit

Architecture Deployment

Relevant cases

One hundred sets of DB2 / Oracle container perspective diagnosis

/ Case introduction With the widespread use of database in all walks of life, its importance is self-evident. At the same time, the industry also has a great demand for database monitoring. This process must be accompanied by database monitoring tools. At present, such tools are also scarce. How to find a good database monitoring tool has become a hot topic for major companies.X0 perspective diagnosis tool provides a "God's perspective", comprehensively analyzes and evaluates the operation status of the database through multiple dimensions, timely alerts the possible performance and other problems of the database, and helps DBAs solve the thorny problems that can only be solved by historical experience. X0 is designed to liberate DBAs. / Project background A joint-stock bank hopes to monitor the performance and resources of hundreds of DB2 / Oracle databases. With the increase of data volume and the complexity of application, DBA is required to timely find the operation faults of the database, conduct root cause analysis, trace the causes of faults and solve the database faults in time. / Customer pain points 1.The deployment of hundreds of database monitoring systems is time-consuming and labor-consuming;
2.Maintenance cost is difficult to control;
3.The consumption of system resources is difficult to control. / Solution 1. Flow control By deploying the soft tap tool on the packet capturing machine, preprocess the traffic of hundreds of databases imported from the tap port to the packet capturing machine, divert the traffic of the physical network card according to each database and distribute it to the virtual network card of the specific virtual bridge, process the traffic received on the virtual network card and forward it to the analysis machine,Then go to the virtual network card of the analysis machine to capture traffic messages according to the library, so as to realize the ability of traffic diversion according to the library.Thus, the x0 container on the analysis machine can capture packets according to the library, analyze and monitor the real-time state of the database.

2. One click deployment Container deployment based on kubernetes, and rapid deployment of hundreds of x0 perspective diagnosis systems through peripheral control, as well as rolling upgrade, gray release and fallback of x0 version.

3. Distributed storage The cluster mode is adopted for storage. The data stored by the three master nodes form a full data set. Whenever a batch of data needs to be stored, the cluster will automatically allocate these data to the appropriate nodes. Each master node is configured with slave nodes for component data backup.

/ Program advantages 1.One click deployment：Compared with traditional virtual machine or physical machine, large-scale container deployment not only solves the dependence of development environment, but also provides great convenience for deployment and management; 2.Reduce resource consumption：Through traffic control, the situation of repeated traffic replication caused by repeated packet capture can be avoided, and the effect of reducing resource consumption and packet loss rate is obvious; 3.Improve operation and maintenance efficiency：each database needs a set of x0 for monitoring, and the large-scale deployment of x0 has been very time-consuming and energy-consuming.The containerization of x0 greatly improves the efficiency of deployment and reduces the difficulty of operation and maintenance; 4.High availability of data：distributed storage is adopted to solve the problem of data inaccessibility caused by single point of failure, which greatly improves the high availability of data.

Deployment details

Background Interface Display