CN106502855A - A kind of test the method for positioning PMC RAID card memory grain problems - Google Patents

Abstract

The invention discloses a kind of test the method that positions PMC RAID card memory grain problems, a kind of test the method for positioning PMC RAID card memory grain problems, concrete grammar is as follows：S1, testing tool cd-rom recorder is burnt in PMC RAID cards to be measured；S2, RAID card to be measured is installed on machine, and connects serial ports instrument to work computer；S3, start to test using test script in work computer；S4, repeatedly test, and collect test result, it is ensured that problem can reappear；S5, test result is checked, position the problem points of memory grain.The present invention can be used for testing and positioning PMC RAID card memory grain problems, for solving the problems such as caused RAID card is stuck as DDR reports an error and EEC reports an error, the service quality of company can be improved with quick positioning question point, acceleration problem process and the speed for solving.

Description

A method for testing and locating PMC-RAID card memory particle problems

technical field

The invention relates to the technical field of computers, in particular to a method for testing and locating PMC-RAID card memory particle problems.

Background technique

RAID (Redundant Array of Independent Disk, independent redundant disk array) card, as a key component in the server, can effectively protect the user's data security. However, if the RAID card itself cannot work stably, data security will be out of the question. At present, the testing of RAID cards in the industry puts more emphasis on performance, and the testing of stability is mostly limited to long-term stress testing.

A RAID card is a board used to implement the RAID function, and is usually composed of a series of components such as an I/O processor, a hard disk controller, a hard disk connector, and a cache. Different RAID cards support different RAID functions. Support RAlD0, RAID1, RAID3, RAID4, RAID5, RAID10 etc. The RAID card can allow many disk drives to transmit data at the same time, and these disk drives are logically a disk drive, so the use of RAID can achieve several times, dozens of times or even hundreds of times the speed of a single disk drive. This is also the problem that the RAID card originally wanted to solve. It can provide fault tolerance function, which is the second important function of RAID card.

Most RAID cards are equipped with a certain number of memory granules to be used as high-speed cache (Cache). The cache is the place where the RAID card exchanges data with the external bus. The RAID card first transfers the data to the cache, and then exchanges data between the cache and the external data bus. The cache has an extremely fast access speed, and is actually a buffer between a relatively low-speed hard disk platter and a relatively high-speed external device (such as memory). The size and speed of the cache are important factors directly related to the actual transmission speed of the RAID card. A large cache can greatly improve the data hit rate and thus improve the overall performance of the RAID card.

Contents of the invention

The technical task of the invention is to provide a method for testing and locating the problem of PMC-RAID card memory particles.

Technical task of the present invention is realized in the following manner, a kind of method of testing and positioning PMC-RAID card internal memory particle problem, concrete method is as follows:

S1, burn the test tool to the PMC-RAID card to be tested with a burner;

S2, install the RAID card to be tested on the machine, and connect the serial port tool to the working computer;

S3, use the test script on the work computer to start the test;

S4. Test multiple times and collect test results to ensure that the problem can be reproduced;

S5. Check the test results and locate the problem points of the memory particles.

Preferably, the specific steps of said S2 are: one end of the serial port tool is connected to the RAID card to be tested, the other end is connected to the working computer, the com port is confirmed, and the DDR test tool is used to open it according to the com port number.

Preferably, the S3 includes a total of eight test items to test and locate possible problems of the memory particles, and the test results are output to the formulated file.

Preferably, the eight test questions are specifically as follows: Runs level, margin, mem DMA, memDMC tests, Run pattern walk1, pattern walk0, pattern blink1, pattern blink0tests.

Compared with the prior art, a method for testing and locating PMC-RAID card memory particle problems of the present invention, the test items include eight items, the test results are comprehensive and accurate, and the problem points can be quickly located; the method can be used for testing and locating The problem of PMC-RAID card memory particles is used to solve the problems of RAID card stuck and EEC error caused by DDR error reporting, speed up the speed of problem processing and resolution, and improve the company's service quality.

Description of drawings

Accompanying drawing 1 is a schematic flow chart of a method for testing and locating memory particle problems of a PMC-RAID card.

detailed description

First of all, it is necessary to burn the test tool to the PMC-RAID card to be tested with a burner. This tool is coded according to the memory particles of the existing RAID card of PMC and can only be used for the RAID card of PMC.

Secondly, install the RAID card to be tested on the machine and connect it to a serial port tool to monitor the RAID card information and operate the memory particle test. Connect the other end of the serial port tool to the working computer, confirm the com port, and use the DDR test tool to open it according to the com port number.

Finally, use the test script to start the test, including a total of 8 test items to test the possible problems of locating memory particles, and output the test results to the specified file for easy observation. Because some memory particle problems are more difficult to reproduce, generally the test will be repeated several times, and all PASS means there is no problem. If there is Failed, you can locate the point where the problem occurred.

The memory particle test mainly includes the following 8 tests:

- Runs level, margin, mem DMA, mem DMC tests

- Run pattern walk1, pattern walk0, pattern blink1, pattern blink0 tests

Example 1:

The following is the test results of three tests in a test we extracted. Among them, the mem DMA and mem DMC tests are failed, while the Walk1 pattern test is passed. We will focus on analyzing the mem DMA and mem DMC based on the test results. Two parts.

Running mem DMA test...

Writing at 0x0

memcmp failed rv = 0x0

Errors: Addr = 0xbe04ea04 0x9a7a448 != 0x9a708a6

Addr = 0xbe04ea0c 0x269e912 != 0x269c26a

Addr = 0xbe04ea14 0xd09a7a45 != 0xd09a309a

Addr = 0xbe04ea1c 0x8c269e90 != 0x8c268c27

Addr = 0xbe04ea44 0xc1a309a6 != 0xc1a30000

Addr = 0xbe04ea4c 0xe068c268 != 0xe0680000

Addr = 0xbe04ea54 0x381a309a != 0x381a0000

Addr = 0xbe04ea5c 0xde068c27 != 0xde068000

Test FAILED!!

Running mem DMC test...

If the following memory init does not complete within 2 seconds

The SPD on the cache module may be incorrect.

Initializing DDR for ECC (size=2048 MiB)

....... DDR fill done.

Writing at 0x0

Test FAILED!!

Running Walk1 pattern test...

Writing to DDR of size 0x80000000...

at 0x0

at 0x10000000

at 0x20000000

at 0x30000000

at 0x40000000

at 0x50000000

at 0x60000000

at 0x70000000

Done

Reading from DDR and comparing...

at 0x0

at 0x10000000

at 0x20000000

at 0x30000000

at 0x40000000

at 0x50000000

at 0x60000000

at 0x70000000

Done

Test Passed

Through the above specific implementation manners, those skilled in the technical field can easily realize the present invention. However, it should be understood that the present invention is not limited to the above-mentioned several specific implementation manners. On the basis of the disclosed embodiments, those skilled in the art can arbitrarily combine different technical features to achieve different technical solutions.

Claims (4)

1. a method for testing and locating the PMC-RAID card internal memory particle problem, is characterized in that, concrete method is as follows: S1, burn the test tool to the PMC-RAID card to be tested with a burner; S2, install the RAID card to be tested on the machine, and connect the serial port tool to the working computer; S3, use the test script on the work computer to start the test; S4. Test multiple times and collect test results to ensure that the problem can be reproduced; S5. Check the test results and locate the problem points of the memory particles. 2. a kind of test according to claim 1 and the method for positioning PMC-RAID card internal memory particle problem, it is characterized in that, the detailed steps of described S2 are, one end of serial port tool connects the RAID card to be tested, and the other end connects Go to the working computer, confirm the com port, and use the DDR test tool to open it according to the com port number. 3. a kind of test according to claim 1 and the method for positioning PMC-RAID card internal memory particle problem, it is characterized in that, comprise eight test items altogether in the described S3, the problem that test positioning internal memory particle may occur, test The results are output to a specified file. 4. a kind of test according to claim 3 and the method for positioning PMC-RAID card internal memory particle problem, it is characterized in that, described eight test questions, specifically as follows: Runs level, margin, mem DMA, mem DMC tests ,Run pattern walk1, pattern walk0, pattern blink1, pattern blink0 tests.