CN109408293B

CN109408293B - Method for monitoring RACK cabinet power supply in real time

Info

Publication number: CN109408293B
Application number: CN201811298688.0A
Authority: CN
Inventors: 刘东伟
Original assignee: Zhengzhou Yunhai Information Technology Co Ltd
Current assignee: Zhengzhou Yunhai Information Technology Co Ltd
Priority date: 2018-11-02
Filing date: 2018-11-02
Publication date: 2022-02-18
Anticipated expiration: 2038-11-02
Also published as: CN109408293A

Abstract

The invention provides a method for monitoring a RACK cabinet power supply in real time, which comprises the following steps: RMC polls each PSU to read the output current value of the PSU; wherein, the RMC reads the value of the PSU output current by PMbus polling each PSU. RMC outputs the log that PSU current is too low; if the RMC outputs the times that the current of the same PSU is too low log to reach the set time threshold, the RMC sends a power supply restart command to the PSU, and the method specifically comprises the following steps: for each PSU, counting the times that the RMC outputs the log of the PSU with too low current; if the times of the same PSU current output by the RMC which is too low log reaches a set time threshold; the RMC issues a power restart command to the PSU.

Description

Method for monitoring RACK cabinet power supply in real time

Technical Field

The invention relates to the technical field of server power monitoring, in particular to a method for monitoring a RACK cabinet power supply in real time.

Background

The RACK complete machine is powered by a group of power supplies in a centralized manner, the power consumption of the complete machine is very high, and the requirement on power supply is severe. With the requirement for increasingly stable operation of data centers, the power supply stability of the RACK complete machine becomes more important.

When the PSU fails, flag bits corresponding to overvoltage, overcurrent, overtemperature and the like in the PSU are set, the RMC judges whether the power supply fails or not by detecting the flag bits, but the flag bits are not set when the power supply fails, so that the RMC cannot find that the power supply fails.

Disclosure of Invention

In order to overcome the defects in the prior art, the invention provides a method for monitoring the power supply of a RACK cabinet in real time, so as to solve the technical problems.

The technical scheme of the invention is as follows:

a method for monitoring RACK cabinet power supply in real time comprises the following steps:

RMC polls each PSU to read the output current value of the PSU;

comparing and judging the read current value;

the RMC outputs a log of the PSU current too low;

and if the RMC outputs the times that the current of the same PSU is too low log to reach the set time threshold, the RMC sends a power supply restart command to the PSU.

Further, the step RMC of polling each PSU to read the output current value of the PSU specifically includes:

the RMC reads the value of the PSU output current by PMbus polling each PSU.

Further, the step of comparing and judging the read current value includes:

and respectively comparing the read PSU output current value with a set current threshold value, and if the PSU output current value is smaller than the set current threshold value, executing the step RMC to output the log of the low PSU current.

Further, if the RMC outputs the number of times that the current of the same PSU is too low log to reach the set number threshold, the RMC issues a power restart command to the PSU, including:

for each PSU, counting the times that the RMC outputs the log of the PSU with too low current;

if the times of the same PSU current output by the RMC which is too low log reaches a set time threshold;

the RMC issues a power restart command to the PSU.

Further, the step RMC issues a power restart command to the PSU, including:

the RMC issues a closing command to the PSU and starts timing;

if the time reaches a set time threshold, the RMC issues an opening command to the PSU;

and reading the PSU output current value, and judging the read current value.

Further, the step of reading the output current value of the PSU and judging the read current value includes:

if the read current value is smaller than the set current threshold value, the RMC issues a closing command to the PSU and starts timing;

reading the PSU output current value, and judging the read current value;

and if the PSU cannot be recovered to be normal after the operation of restarting the PSU is continuously carried out for three times, outputting an alarm log corresponding to the PSU.

Further, the method further comprises:

PSU anomaly logs were automatically collected and saved.

Further, the method comprises the steps of automatically collecting and storing the PSU abnormal log, and specifically comprises the following steps:

acquiring a PSU abnormal log;

determining the sequence number of an event corresponding to the abnormal log according to a returned result code and the node ID in the PSU abnormal log;

and storing the PSU abnormal log into a storage queue of a log table in the RMC according to the result code and the sequence number.

Further, the step of determining the sequence number of the event corresponding to the abnormal log according to the result code returned from the PSU abnormal log and the node ID includes:

using the result code and the node ID as keywords to inquire whether an event corresponding to the abnormal log exists in the current storage queue;

if the event exists, acquiring a serial number of the event existing in the storage queue, and distributing the acquired serial number to an event corresponding to a newly generated PSU abnormal log;

and if the PSU exception log does not exist, allocating a new sequence number to an event corresponding to the newly generated PSU exception log, wherein the new sequence number is the current existing maximum sequence number plus 1.

Further, the step of counting, for each PSU, the number of times the RMC outputs the log of the PSU current that is too low specifically includes:

using the result code and the node ID as keywords to inquire the serial number in the current storage queue;

and counting the occurrence times of the same serial number, namely the times that the current output by the PSU is too low log.

According to the technical scheme, the invention has the following advantages: the method fundamentally ensures that the power supply can be detected after failure, can try to repair the power supply to a certain extent, ensures safe and stable operation of a data center, and ensures reliability and stability of power supply of a cabinet.

In addition, the invention has reliable design principle, simple structure and very wide application prospect.

Therefore, compared with the prior art, the invention has prominent substantive features and remarkable progress, and the beneficial effects of the implementation are also obvious.

Drawings

Figure 1 is a flow chart of a method for monitoring RACK power in real time.

Detailed Description

The invention provides a method for monitoring RACK cabinet power supply in real time, which judges whether the power supply fails by continuously reading the output current value of a PSU (power supply unit) through an RMC (remote metering controller), meanwhile, the RMC can print an abnormal log in real time, tries to recover a power supply restart action after the power supply fails, and automatically stores the abnormal log of the PSU.

In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

Example one

As shown in fig. 1, an embodiment of the present invention provides a method for monitoring a RACK power supply in real time, including the following steps:

s1: RMC polls each PSU to read the output current value of the PSU;

it should be noted that, a plurality of PSU modules are arranged in the RACK cabinet, in this step, the RMC polls each PSU through a PMbus (Power Management Bus) to READ the value of PSU Page 00x8C (READ _ IOUT), that is, the value of PSU output current;

s2: comparing and judging the read current value;

firstly, the read PSU output current value can be classified and stored according to different PSUs; simultaneously comparing the read PSU output current value with a set current threshold, where the set current threshold may be 0.5A, and if the PSU output current value is less than 0.5A, performing step S3;

s3: the RMC outputs a log of the PSU current too low;

the specific implementation commands of the steps are as follows:

RMC polls through PMC bus command to read the value of 0x8C, i.e. the PSU output current value:

the # # # # # # # # ### # is polled and READ-IOUT 0x8C numerical value # # # # # # # # # # # # # ############# - #

～#PSU_CLI A1 0 0X8C

--------------------------

READ_IOUT(L16N＝-7):0.250000(Amps)

--------------------------

RAW BYTES(read):

0x20 0x00

As above, the PSU output current is 0.25A, RMC judges that the value is less than 0.5A, and the following log is reported under RMC:

0912 08/23/2018 13:54:26 123456:critical PSU A1output current was0.250000，0001

s4: and if the times that the output current of the same PSU is lower than the set current threshold reach the set time threshold, the RMC sends a power supply restart command to the PSU.

Firstly, counting the times that the output current of each PSU is lower than a set current threshold value, namely, the times that the RMC outputs the log of the low current of the PSU;

when the RMC repeatedly detects that the current of the PSU is less than 0.5A for 3 times, the RMC actively issues a stop PSU command to the PSU, restarts the PSU after 20s, and then detects whether the output current of the PSU is normal. If the PSU still can not be recovered to normal after three times of stop/start PSUs, the RMC stops recovering the PSU and outputs a corresponding alarm log.

The switching on and off PSU operates as follows:

stop Power/PSUA 1// RMC automatic turn-off PSUA1

Start Power/PSUA 1// RMC automatic turn-on PSUA1

If the PSU output current returns to normal, the following log is reported under RMC:

0914 08/23/2018 13:54:43 123456:info PSU A1output current got back to normal 0002

if the PSU is turned on and off three times and still cannot be recovered to be normal, the following logs are reported under the RMC:

092508/23/201813: 57: 45123456: info PSU A1auto recovery fail 0003 the method further comprises the step S5: PSU anomaly logs were automatically collected and saved.

When the RMC fails to attempt to repair the PSU, the RMC automatically collects the PSU abnormal log and stores the PSU abnormal log under the RMC for subsequent problem analysis and judgment.

The RMC auto-collect PSU exception log commands are as follows:

～#get_psu_log A1

log saved in file:/conf/rmc/log/psuA1_log_all.txt

the PSU problem can subsequently be located by looking at the psuA1 log all.

Example two

The embodiment of the invention provides a method for monitoring a RACK cabinet power supply in real time, which comprises the following steps:

s1: RMC polls each PSU to read the output current value of the PSU;

s2: comparing and judging the read current value;

s3: the RMC outputs a log of the PSU current too low;

it should be noted that the output log needs to carry relevant information, such as node ID in RACK enclosure, RACK enclosure ID, and returned result code;

wherein, the result code may be hexadecimal number, RACK enclosure ID is enclosure unit, may be A, B, C, etc., since multiple PSUs are provided in RACK enclosure, each PSU is provided with a node ID, and the IDs of nodes cannot be repeated, a1, a2, B1, B2, etc., RACK enclosure ID and node ID are mainly used to explicitly indicate which node in which enclosure the result code is specifically returned.

The specific implementation commands of the steps are as follows:

～#PSU_CLI A1 0 0X8C

--------------------------

READ_IOUT(L16N＝-7):0.250000(Amps)

--------------------------

RAW BYTES(read):

0x20 0x00

0912 08/23/2018 13:54:26 123456:critical PSU A1 output current was 0.250000，0001

when the current of the PSU is detected to be less than 0.5A for 3 times, a stop PSU command is issued to the PSU actively, the PSU is restarted after 20s, and then whether the output current of the PSU is normal or not is detected. If the PSU still can not be recovered to normal after three times of stop/start PSUs, the RMC stops recovering the PSU and outputs a corresponding alarm log.

The switching on and off PSU operates as follows:

stop Power/PSUA 1// RMC automatic turn-off PSUA1

Start Power/PSUA 1// RMC automatic turn-on PSUA1

In step S5, the method specifically includes the following steps:

s51: acquiring a PSU abnormal log;

each result code returned in the abnormal log represents one type of event, and a storage queue for storing the log is correspondingly arranged for each type of event so as to facilitate the classified storage of the log;

s52: determining the serial number of the corresponding event according to the result code and the node ID in the PSU abnormal log;

the corresponding event type can be determined according to the returned result code, the returned result can be multiple for the same type of event, but a unique serial number can be set for each event, for example, when the output current of the PSUA1 is less than 0.5A, a log is output, the corresponding serial number 0912, each time the output current of the PSUA1 is less than the log of 0.5A, whether the corresponding event already exists in the current storage queue is firstly inquired in the current storage queue by using the result code and the node ID as keys, if so, the serial number 0912 corresponding to the event existing in the queue is obtained and allocated to the newly generated event, otherwise, a new serial number is allocated to the newly generated event, and the new serial number is the maximum serial number existing currently plus 1.

S53: and storing the PSU abnormal log into a storage queue of a log table in the RMC according to the result code and the sequence number.

As mentioned in step S4 above, first, counting, for each PSU, the number of times the PSU output current is below the set current threshold, i.e., the number of times the RMC outputs the log that the PSU current is too low, may be achieved by:

firstly, a returned result code and a node ID are used as keywords to inquire a serial number in a current storage queue;

The terms "first," "second," "third," "fourth," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," as well as any variations thereof, are intended to cover non-exclusive inclusions.

The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for monitoring RACK cabinet power supply in real time is characterized by comprising the following steps:

RMC polls each PSU through PMbus to read the numerical value of PSU output current;

respectively comparing the read PSU output current value with a set current threshold value;

if the output current value of the PSU is smaller than the set current threshold value, the RMC outputs the log of the over-low PSU current;

the RMC issues a closing command to the PSU and starts timing;

reading the PSU output current value and judging the read current value, wherein the method specifically comprises the steps that if the read current value is judged to be smaller than a set current threshold value, the RMC issues a closing command to the PSU and starts timing; if the time reaches a set time threshold, the RMC issues an opening command to the PSU; reading the PSU output current value, and judging the read current value; and if the PSU cannot be recovered to be normal after the operation of restarting the PSU is continuously carried out for three times, outputting an alarm log corresponding to the PSU.

2. A method of real time monitoring RACK cabinet power supply as claimed in claim 1, the method further comprising:

PSU anomaly logs were automatically collected and saved.

3. The method for real-time monitoring of RACK power supply of claim 2, wherein the step of automatically collecting PSU anomaly log and storing it comprises the following steps:

acquiring a PSU abnormal log;

4. A method for real time monitoring RACK power supply as claimed in claim 3, wherein the step of determining the serial number of the event corresponding to the abnormal log according to the returned result code and node ID in the PSU abnormal log comprises:

5. The method for real-time monitoring of RACK power supply of claim 4, wherein the step of counting, for each PSU, the number of times the RMC outputs the log of the PSU current that is too low specifically comprises: