CN105589764B - CPU abnormality eliminating methods and device - Google Patents

CPU abnormality eliminating methods and device Download PDF

Info

Publication number
CN105589764B
CN105589764B CN201510916234.5A CN201510916234A CN105589764B CN 105589764 B CN105589764 B CN 105589764B CN 201510916234 A CN201510916234 A CN 201510916234A CN 105589764 B CN105589764 B CN 105589764B
Authority
CN
China
Prior art keywords
message
cpu
caching
information
global variable
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201510916234.5A
Other languages
Chinese (zh)
Other versions
CN105589764A (en
Inventor
宋海宾
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
New H3C Technologies Co Ltd
Original Assignee
New H3C Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by New H3C Technologies Co Ltd filed Critical New H3C Technologies Co Ltd
Priority to CN201510916234.5A priority Critical patent/CN105589764B/en
Publication of CN105589764A publication Critical patent/CN105589764A/en
Application granted granted Critical
Publication of CN105589764B publication Critical patent/CN105589764B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/14—Error detection or correction of the data by redundancy in operation
    • G06F11/1402—Saving, restoring, recovering or retrying
    • G06F11/1446—Point-in-time backing up or restoration of persistent data
    • G06F11/1448—Management of the data involved in backup or backup restore

Abstract

The application proposes CPU abnormality eliminating methods and device.Method includes：CPU receives any message, which is put into caching, and the caching first address and message length of message are recorded in preset first global variable；When CPU takes out message from caching and is successfully transmitted away, CPU deletes the caching first address and message length information of the message preserved in the first global variable；When CPU exceptions, CPU is according to the caching first address and message length information of all messages stored in the first global variable, corresponding original message content is read from caching, the original message content of reading is saved in default upper memory area domain, the information of the upper memory area domain storage does not change in equipment restarting process；After equipment is restarted, CPU reads all original message contents from upper memory area domain, and all original message contents of reading are copied in disk.The application improves CPU exception positioning accuracies.

Description

CPU abnormality eliminating methods and device
Technical field
This application involves unit exception processing technology field more particularly to CPU abnormality eliminating methods and device.
Background technology
Current network security equipment needs to carry out many complicated processing, analysis, calculating etc. for passing message, because Message on network is very strange, and in design or coding slightly mistake, a lopsided message are just enough to lead to whole equipment weight It opens, CPU exceptions.Such as current ALG (Application Layer Gateway, application layer gateway) is handled, DPI (Deep Packet Inspection, deep-packet detection) etc. processing procedures, often occur it is abnormal crash, and be not easily found cause it is different Normal message, positioning are extremely difficult.
After equipment enters exception, when system recording exceptional each CPU register information and stack information just restart recovery Scene.
But by checking stack information, can only Primary Location to go out be finally which function is abnormal, specifically what is former It can not then be learnt because causing CPU extremely.It needs to recompilate version, increases Debugging message, reappeared at the scene, needed simultaneously Message is grabbed in the interface of equipment, by analyzing these messages, message is played back one by one and causes CPU different to analyze which specific message Often.As it can be seen that orientation problem is complicated, time-consuming in this way, and the frequency that many times lopsided message occurs is very low, therefore works as After CPU exceptions, if follow-up message lopsided for a long time does not occur, the problem of may result in can not position.
Invention content
The embodiment of the present application provides CPU abnormality eliminating methods and device.
What the technical solution of the application was realized in：
A kind of CPU abnormality eliminating methods, this method include：
CPU receives any message, which is put into caching, and the caching first address and message length of message are recorded Into preset first global variable；
When CPU takes out message from caching and is successfully transmitted away, what is preserved in CPU the first global variables of deletion should The caching first address and message length information of message；
When CPU exceptions, CPU is according to the caching first address and message length of all messages stored in the first global variable Information reads corresponding original message content from caching, and the original message content of reading is saved in default upper memory area In domain, the information of the upper memory area domain storage does not change in equipment restarting process；
After equipment is restarted, CPU reads all original message contents from upper memory area domain, by all original of reading Message content copies in disk.
A kind of CPU abnormality eliminating methods, this method include：
Any core of CPU receives any message, which is put into caching, by the caching first address and message of message In length records to preset the first global variable for the core；
When CPU takes out message from caching and is successfully transmitted away, what is preserved in CPU the first global variables of deletion should The caching first address and message length information of message；
When CPU exceptions, CPU is according to the caching first address for all messages stored in the first global variable of each core And message length, corresponding original message content is read from caching, all original message contents of reading is saved in default In upper memory area domain, the information of the upper memory area domain storage does not change in equipment restarting process；
After equipment is restarted, CPU reads all original message contents for each core from upper memory area domain, will read All original message contents copy in disk.
A kind of CPU exception handling devices, the device include：
Logging modle：Any message is received, which is put into caching, the caching first address and message of message is long Degree is recorded in preset first global variable；When taking out and being successfully transmitted away from caching by message, CPU deletes first The caching first address and message length information of the message preserved in global variable；
Exception processing module：When CPU exceptions, according to the caching first address of all messages stored in the first global variable With message length information, corresponding original message content is read from caching, the original message content of reading is saved in default In upper memory area domain, the information of the upper memory area domain storage does not change in equipment restarting process；
Reboot process module：After equipment is restarted, all original message contents are read from upper memory area domain, will be read All original message contents copy in disk.
A kind of CPU exception handling devices, the device include：
Logging modle：When any core receives any message, which is put into caching, by the first ground of the caching of message Location and message length are recorded in preset the first global variable for the core；When message is taken out from caching and is successfully sent out When seeing off, the caching first address and message length information of the message preserved in the first global variable are deleted；
Exception processing module：When CPU exceptions, according to all messages stored in the first global variable for each core First address and message length are cached, corresponding original message content is read from caching, by all original message contents of reading It is saved in default upper memory area domain, the information of the upper memory area domain storage does not change in equipment restarting process；
Reboot process module：After equipment is restarted, all original messages for each core are read from upper memory area domain Content copies to all original message contents of reading in disk.
As it can be seen that in the embodiment of the present application, when message (such as：Lopsided message) when not sending successfully, preserves this and do not send into The content for not sending successful message is saved in upper memory area domain, works as equipment by the content of the message of work(after CPU exceptions After restarting, the information in upper memory area domain is copied into disk, thus can read not send from disk and successfully report Text (such as：Lopsided message) content, specific aim analysis is carried out to the content of the message, so that it may improve abnormal positioning accuracy.
Description of the drawings
Fig. 1 is the CPU abnormality eliminating method flow charts that one embodiment of the application provides；
Fig. 2 is the multi-core CPU abnormality eliminating method flow chart that another embodiment of the application provides；
Fig. 3 is the composition figure for the CPU exception handling devices that one embodiment of the application provides；
Fig. 4 is the composition figure for the CPU exception handling devices that another embodiment of the application provides.
Specific implementation mode
Below in conjunction with the accompanying drawings and specific embodiment the present invention is further described in more detail.
For convenience of the embodiment of the present application is understood, following explanation is provided first：
It interrupts and exception is the system event for making processing turn to, hardware or software may detect that.
Kernel is such differentiation interruption and exception：
1) interrupt be any moment all may occur unrelated asynchronous event is currently executed with processor, mainly by I/O (Input/Output, input/output) equipment, processor clock or timer generate, and processor can open or close Response to interruption.
2) abnormal generated during specific instruction executes, and the rerun routine meeting under the same conditions of same data is used It generates again abnormal.Exception is including memory accessing outside, part debugging instruction and except zero error etc..Kernel is also by system service tune With being considered as exception.
Fig. 1 is the CPU abnormality eliminating method flow charts that one embodiment of the application provides, and is as follows：
Step 101：CPU receives any message, which is put into caching, by the caching first address and message of message In length records to preset first global variable.
Step 102：CPU carries out business processing to message, and it is complete that the business state information of message is recorded preset second In office's variable, while the business statistics information of message being recorded in preset third global variable.
Business state information includes：The status information for the business that CPU executes message.Such as：It, can be with for session service (such as according to the identification information of message：Five-tuple) find corresponding session control block, it will and words control following business shape in the block State information is recorded in the second global variable：
1) ageing time：How long the conversation aging；
2) protocol type：TCP (Transmission Control Protocol, transmission control protocol), UDP (User Datagram Protocol, User Datagram Protocol), ICMP (Internet Control Message Protocol, Yin Te Network control information protocol) etc.；
3) application type：FTP (File Transfer Protocol, File Transfer Protocol), HTTP (HyperText Transfer Protocol, hypertext transfer protocol), wechat, QQ etc.；
4) ALG types：H323, FTP, SIP (Session Initiation Protocol, Session initiation Protocol), Telnet, DNS (Domain Name System, domain name system) etc.；
5) current state：Specifically,
UDP type states have：OPEN (opening), ESTABLISH (foundation), CLOSE (closing) etc.；
TCP type states have：SYN_SEND (synchronous send), SYN_ACK (confirmation of synchronization), ESTABLISH (foundation), FIN_SEND (terminating to send), FIN_ACK (terminating to confirm), CLOSE (closing)；
6) CPU usage：Current CPU usage；
7) memory usage：Current memory occupancy；
8) number is connected：The total number of current sessions；
9) TCP connection number：The number of TCP connection；
10) UDP connections number：The connection number of current UDP.
Business statistics information is divided into public statistical information control block statistical information corresponding with this message.Such as：For meeting The public statistical information of call business, session may include：
1) rate is created：The current number for establishing session each second；
2) handling capacity：The number that current each second E-Packets；
3) E-Packet number：Total number that E-Packets；
4) message number is lost：Total dropping packets number.
The corresponding control block statistical information of this message may include：The session is current：The message number of transmission, the message of reception The byte number of the byte number, reception that count, send, the message number of discarding.
Step 103：When CPU takes out message from caching and is successfully transmitted away, CPU is deleted in the first global variable The caching first address and message length information of the message preserved.
When message is not successfully sent, the caching first address and message length of the message preserved in the first global variable Information is not deleted, and continues to retain.
Step 104：When CPU exceptions, CPU read from the first global variable each message caching first address and message it is long Information is spent, original message content is found according to the information of reading in the buffer, original message content is saved in default high-end In storage region；Meanwhile CPU reads all business state informations from the second global variable, is read from third global variable All business statistics information, will be in the information preservation of reading to default upper memory area domain.
The information stored in upper memory area domain will not change in equipment restarting process.
Step 105：After equipment is restarted, CPU reads all original message contents, all business from upper memory area domain Status information and all business statistics information, all information of reading are copied in disk.
Disk is such as：Flash memory (flash) etc..
In this way, developer can read original message content, business state information and business statistics letter from disk Breath, analyzes these information of reading, to be positioned extremely to CPU.
Fig. 2 is the multi-core CPU abnormality eliminating method flow chart that another embodiment of the application provides, and is as follows：
Step 201：Any core of CPU receives any message, which is put into caching, by the first ground of the caching of message Location and message length are recorded in preset the first global variable for the core.
In practical applications, can be that all cores set a first global variable array, each variable in array corresponds to One core.I.e.：If there is n core, a first global variable array for including n variable is set.
Step 202：CPU carries out business processing to message, and preset be directed to is recorded in the business state information of message should In second global variable of core, while the preset third global variable for the core is recorded in the business statistics information of message In.
Similarly, in practical applications, can be that all cores set a second global variable array, each change in array The corresponding core of amount.I.e.：If there is n core, a second global variable array for including n variable is set.
Can be that all cores set a third global variable array, each variable in array corresponds to a core.I.e.：If having A third global variable array for including n variable is then arranged in n core.
Step 203：When CPU takes out message from caching and is successfully transmitted away, CPU is deleted in the first global variable The caching first address and message length information of the message preserved.
Step 204：When CPU exceptions, the caching that CPU reads all messages from the first global variable for each core is first Address and message length information find original message content according to the information of reading, original message content are protected in the buffer It is stored in default upper memory area domain；Meanwhile CPU reads all service condition letters from the second global variable for each core Breath, reads all business statistics information from the third global variable for each core, and the information preservation of reading is high-end to presetting In storage region.
The information stored in upper memory area domain will not change in equipment restarting process.
Step 205：After equipment is restarted, CPU reads all original message contents, all business from upper memory area domain Status information and all business statistics information, all information of reading are copied in disk.
Disk is such as：Flash memory (flash) etc..
The advantageous effects of the embodiment of the present application are as follows：
1) CPU preservations do not send successful message, such as：The content of lopsided message will not be sent into after CPU exceptions The message of work(, such as：The content of lopsided message is saved in upper memory area domain, will be in upper memory area domain after equipment is restarted The content for not sending successful message copy to disk, can thus be read from disk and not send successful message, Such as：The content of lopsided message carries out specific aim analysis, so that it may improve abnormal positioning accuracy to the content of these messages.
2) further, the business state information of CPU recorded messages and business statistics information, after CPU exceptions, by message Business state information and business statistics information preservation to default upper memory area domain, after equipment is restarted, by the business of message Status information, business statistics information copy to disk together with the content for not sending successful message, thus can be from magnetic The business state information of message, business statistics information are read in disk together with the content for not sending successful message, to reading All information are analyzed together, so that it may further increase abnormal positioning accuracy.
Fig. 3 is the composition schematic diagram for the CPU exception handling devices that one embodiment of the application provides, the device mainly includes： Logging modle, exception processing module and reboot process module, wherein：
Logging modle：Any message is received, which is put into caching, the caching first address and message of message is long Degree is recorded in preset first global variable；When taking out and being successfully transmitted away from caching by message, CPU deletes first The caching first address and message length information of the message preserved in global variable.
Exception processing module：When CPU exceptions, according to the caching first address of all messages stored in the first global variable With message length information, corresponding original message content is read from caching, the original message content of reading is saved in default In upper memory area domain, the information of the upper memory area domain storage does not change in equipment restarting process.
Reboot process module：After equipment is restarted, all original message contents are read from upper memory area domain, will be read All original message contents copy in disk.
In a kind of embodiment, logging modle is further used for after receiving any message
The business state information of message is recorded in preset second global variable, while the business statistics of message being believed Breath is recorded in preset third global variable；
And exception processing module is further used for after CPU is abnormal：
By all business state informations stored in the second global variable and all business stored in third global variable Statistical information copies in default upper memory area domain；
And reboot process module is further used for after equipment is restarted：
All business state informations and all business statistics information are read from upper memory area domain, by all of reading Information copies in disk.
In a kind of embodiment, when the business is session service, the business state information includes one of following or appoints Meaning combination：
Ageing time, protocol type, application type, ALG types, UDP type states, TCP type states, CPU usage, Memory usage, session connection number, TCP connection number, UDP connection numbers；
The business statistics information includes one of following or arbitrary combination：
Newly-built rate, handling capacity, E-Packet number, loses message number, the currently transmitted message number of the session, which works as The message number of preceding reception, the currently transmitted byte number of the session, the currently received byte number of the session, what which currently abandoned Message number.
Fig. 4 is the composition schematic diagram for the CPU exception handling devices that another embodiment of the application provides, which mainly wraps It includes：Logging modle, exception processing module and reboot process module, wherein：
Logging modle：When any core receives any message, which is put into caching, by the first ground of the caching of message Location and message length are recorded in preset the first global variable for the core；When message is taken out from caching and is successfully sent out When seeing off, the caching first address and message length information of the message preserved in the first global variable are deleted.
Exception processing module：When CPU exceptions, according to all messages stored in the first global variable for each core First address and message length are cached, corresponding original message content is read from caching, by all original message contents of reading It is saved in default upper memory area domain, the information of the upper memory area domain storage does not change in equipment restarting process.
Reboot process module：After equipment is restarted, all original messages for each core are read from upper memory area domain Content copies to all original message contents of reading in disk.
In a kind of embodiment, logging modle is further used for after any core receives any message：
The business state information of message is recorded in preset the second global variable for the core, while by message Business statistics information is recorded in the preset third global variable for the core；
And exception processing module is further used for after CPU is abnormal：
By all business state informations stored in the second global variable for each core and for the third overall situation of each core All business statistics information stored in variable copy in default upper memory area domain；
And reboot process module is further used for after equipment is restarted：
All business state informations and all business statistics information for each core are read from upper memory area domain, it will All information read copy in disk.
In a kind of embodiment, when the business is session service, the business state information includes one of following or appoints Meaning combination：
Ageing time, protocol type, application type, ALG types, UDP type states, TCP type states, CPU usage, Memory usage, session connection number, TCP connection number, UDP connection numbers；
The business statistics information includes one of following or arbitrary combination：
Newly-built rate, handling capacity, E-Packet number, loses message number, the currently transmitted message number of the session, which works as The message number of preceding reception, the currently transmitted byte number of the session, the currently received byte number of the session, what which currently abandoned Message number.
The foregoing is merely the preferred embodiments of the application, not limiting the application, all essences in the application With within principle, any modification, equivalent substitution, improvement and etc. done should be included within the scope of the application protection god.

Claims (12)

1. a kind of CPU abnormality eliminating methods, which is characterized in that this method includes：
CPU receives any message, which is put into caching, and the caching first address of message and message length information are recorded Into preset first global variable；
When CPU takes out message from caching and is successfully transmitted away, CPU deletes the message preserved in the first global variable Caching first address and message length information；
When CPU exceptions, CPU believes according to the caching first address and message length of all messages stored in the first global variable Breath, corresponding original message content is read from caching, the original message content of reading is saved in default upper memory area domain In, the information of the default upper memory area domain storage does not change in equipment restarting process；
After equipment is restarted, CPU reads all original message contents from default upper memory area domain, by all original of reading Message content copies in disk.
2. according to the method described in claim 1, it is characterized in that, the CPU further comprises after receiving any message：
CPU carries out business processing to message, the business state information of message is recorded in preset second global variable, simultaneously The business statistics information of message is recorded in preset third global variable；
And further comprise after the CPU exceptions：
All business that CPU reads all business state informations stored in the second global variable and stored in third global variable Statistical information, will be in all information preservations of reading to default upper memory area domain；
And the equipment restart after further comprise：
CPU reads all business state informations and all business statistics information from default upper memory area domain, by reading All information copy in disk.
3. according to the method described in claim 2, it is characterized in that, when the business be session service when, the service condition Information includes one of following or arbitrary combination：
Ageing time, protocol type, application type, application layer gateway ALG types, User Datagram Protocol UDP type states pass Transport control protocol discusses TCP type states, CPU usage, memory usage, session connection number, TCP connection number, UDP connection numbers；
The business statistics information includes one of following or arbitrary combination：
Newly-built rate, handling capacity, E-Packet number, loses message number, the currently transmitted message number of the session, which currently connects The message number of receipts, the currently transmitted byte number of the session, the currently received byte number of the session, the message which currently abandons Number.
4. a kind of CPU abnormality eliminating methods, which is characterized in that this method includes：
Any core of CPU receives any message, which is put into caching, by the caching first address and message length of message Information is recorded in preset the first global variable for the core；
When CPU takes out message from caching and is successfully transmitted away, CPU deletes the message preserved in the first global variable Caching first address and message length information；
When CPU exceptions, CPU is according to the caching first address and report for all messages stored in the first global variable of each core Literary length information reads corresponding original message content from caching, all original message contents of reading is saved in default In upper memory area domain, the information of the default upper memory area domain storage does not change in equipment restarting process；
After equipment is restarted, CPU reads all original message contents for each core from default upper memory area domain, will read All original message contents copy in disk.
5. according to the method described in claim 4, it is characterized in that, any core further wraps after receiving any message It includes：
CPU carries out business processing to message, and preset second overall situation for the core is recorded in the business state information of message In variable, while the business statistics information of message being recorded in the preset third global variable for the core；
And further comprise after the CPU exceptions：
CPU reads all business state informations stored in the second global variable for each core and the third overall situation for each core All business statistics information stored in variable, will be in all information preservations of reading to default upper memory area domain；
And the equipment restart after further comprise：
CPU is read from default upper memory area domain for all business state informations of each core and all business statistics letter Breath, all information of reading are copied in disk.
6. according to the method described in claim 5, it is characterized in that, when the business be session service when, the service condition Information includes one of following or arbitrary combination：
Ageing time, protocol type, application type, application layer gateway ALG types, User Datagram Protocol UDP type states pass Transport control protocol discusses TCP type states, CPU usage, memory usage, session connection number, TCP connection number, UDP connection numbers；
The business statistics information includes one of following or arbitrary combination：
Newly-built rate, handling capacity, E-Packet number, loses message number, the currently transmitted message number of the session, which currently connects The message number of receipts, the currently transmitted byte number of the session, the currently received byte number of the session, the message which currently abandons Number.
7. a kind of CPU exception handling devices, which is characterized in that the device includes：
Logging modle：Any message is received, which is put into caching, the caching first address and message length of message are believed Breath is recorded in preset first global variable；When taking out and being successfully transmitted away from caching by message, CPU deletes first The caching first address and message length information of the message preserved in global variable；
Exception processing module：When CPU exceptions, according to the caching first address and report of all messages stored in the first global variable Literary length information reads corresponding original message content from caching, the original message content of reading is saved in default high-end In storage region, the information of the upper memory area domain storage does not change in equipment restarting process；
Reboot process module：After equipment is restarted, all original message contents are read from default upper memory area domain, will be read All original message contents copy in disk.
8. device according to claim 7, which is characterized in that the logging modle receives further after any message For
The business state information of message is recorded in preset second global variable, while the business statistics information of message being remembered It records in preset third global variable；
And the exception processing module is further used for after CPU is abnormal：
By all business state informations stored in the second global variable and all business statistics stored in third global variable Information copies in default upper memory area domain；
And the reboot process module is further used for after equipment is restarted：
All business state informations and all business statistics information are read from default upper memory area domain, by all of reading Information copies in disk.
9. device according to claim 8, which is characterized in that when the business is session service, the service condition Information includes one of following or arbitrary combination：
Ageing time, protocol type, application type, application layer gateway ALG types, User Datagram Protocol UDP type states pass Transport control protocol discusses TCP type states, CPU usage, memory usage, session connection number, TCP connection number, UDP connection numbers；
The business statistics information includes one of following or arbitrary combination：
Newly-built rate, handling capacity, E-Packet number, loses message number, the currently transmitted message number of the session, which currently connects The message number of receipts, the currently transmitted byte number of the session, the currently received byte number of the session, the message which currently abandons Number.
10. a kind of CPU exception handling devices, which is characterized in that the device includes：
Logging modle：When any core receives any message, which is put into caching, by the caching first address of message and Message length information is recorded in preset the first global variable for the core；When message is taken out from caching and is successfully sent out When seeing off, the caching first address and message length information of the message preserved in the first global variable are deleted；
Exception processing module：When CPU exceptions, according to the caching of all messages stored in the first global variable for each core First address and message length information read corresponding original message content from caching, by all original message contents of reading It is saved in default upper memory area domain, the information of the default upper memory area domain storage does not change in equipment restarting process Become；
Reboot process module：After equipment is restarted, all original messages for each core are read from default upper memory area domain Content copies to all original message contents of reading in disk.
11. device according to claim 10, which is characterized in that the logging modle receives any message in any core It is further used for later：
The business state information of message is recorded in preset the second global variable for the core, while by the business of message In statistics record to the preset third global variable for the core；
And the exception processing module is further used for after CPU is abnormal：
By all business state informations stored in the second global variable for each core and for the third global variable of each core All business statistics information of middle storage copy in default upper memory area domain；
And the reboot process module is further used for after equipment is restarted：
All business state informations and all business statistics information for each core are read from default upper memory area domain, it will All information read copy in disk.
12. according to the devices described in claim 11, which is characterized in that when the business is session service, the business shape State information includes one of following or arbitrary combination：
Ageing time, protocol type, application type, application layer gateway ALG types, User Datagram Protocol UDP type states pass Transport control protocol discusses TCP type states, CPU usage, memory usage, session connection number, TCP connection number, UDP connection numbers；
The business statistics information includes one of following or arbitrary combination：
Newly-built rate, handling capacity, E-Packet number, loses message number, the currently transmitted message number of the session, which currently connects The message number of receipts, the currently transmitted byte number of the session, the currently received byte number of the session, the message which currently abandons Number.
CN201510916234.5A 2015-12-10 2015-12-10 CPU abnormality eliminating methods and device Active CN105589764B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510916234.5A CN105589764B (en) 2015-12-10 2015-12-10 CPU abnormality eliminating methods and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510916234.5A CN105589764B (en) 2015-12-10 2015-12-10 CPU abnormality eliminating methods and device

Publications (2)

Publication Number Publication Date
CN105589764A CN105589764A (en) 2016-05-18
CN105589764B true CN105589764B (en) 2018-10-09

Family

ID=55929363

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510916234.5A Active CN105589764B (en) 2015-12-10 2015-12-10 CPU abnormality eliminating methods and device

Country Status (1)

Country Link
CN (1) CN105589764B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108415729A (en) * 2017-12-29 2018-08-17 北京智芯微电子科技有限公司 A kind of processing method and processing device of cpu instruction exception
CN110109772A (en) * 2018-02-01 2019-08-09 中兴通讯股份有限公司 A kind of method for restarting of CPU, communication equipment and readable storage medium storing program for executing
CN113055196B (en) * 2019-12-28 2022-08-26 华为技术有限公司 Message processing method, network device and computer readable storage medium
CN111159075B (en) * 2019-12-31 2021-11-05 成都海光微电子技术有限公司 Data transmission method and data transmission device
CN112799884B (en) * 2021-04-14 2021-08-03 北京八角科技有限责任公司 Automatic recovery method, device and storage medium after abnormal exit of network data processing system

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7370038B2 (en) * 2005-03-31 2008-05-06 Inventec Corporation Stereographic network monitoring system and method thereof
CN101150525B (en) * 2007-11-20 2010-07-21 杭州华三通信技术有限公司 Release method, system and logic module for buffered address
CN101277224B (en) * 2008-05-26 2011-01-05 杭州华三通信技术有限公司 Method and apparatus for locating exception message access point
CN104394099B (en) * 2014-12-08 2017-09-19 福建星网锐捷网络有限公司 A kind of message transmitting method and device

Also Published As

Publication number Publication date
CN105589764A (en) 2016-05-18

Similar Documents

Publication Publication Date Title
CN105589764B (en) CPU abnormality eliminating methods and device
US9954873B2 (en) Mobile device-based intrusion prevention system
JP4537579B2 (en) Bidirectional process-to-process byte stream protocol
US7827295B2 (en) Protocol stack
US20070208838A1 (en) Method and system for mirroring dropped packets
US10735282B1 (en) Apparatus, system, and method for debugging network devices based on the contents of dropped packets
US9015822B2 (en) Automatic invocation of DTN bundle protocol
KR20060094861A (en) Windows remote debugger service
US8356332B2 (en) Extensible protocol validation
US10015205B1 (en) Techniques for traffic capture and reconstruction
CN110830330B (en) Firewall testing method, device and system
US20150071085A1 (en) Network gateway for real-time inspection of data frames and identification of abnormal network behavior
CN107800663B (en) Method and device for detecting flow offline file
Dawson et al. Experiments on six commercial TCP implementations using a software fault injection tool
US8490173B2 (en) Unauthorized communication detection method
Masumi et al. Towards efficient labeling of network incident datasets using tcpreplay and snort
US11252184B2 (en) Anti-attack data transmission method and device
CN113098727A (en) Data packet detection processing method and device
Orgiyan et al. Tapping TCP streams
Huang et al. Stateful traffic replay for web application proxies
CN114051704A (en) Attack mitigation in packet-switched networks
CN115190077B (en) Control method, control device and computing equipment
KR102156600B1 (en) System and method for creating association between packets collected in network and processes in endpoint computing device
Hall et al. Non-intrusive estimation of web server delays
CN114328190B (en) Method, system and server for automatically splitting IPS (in-plane switching) event

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
CB02 Change of applicant information
CB02 Change of applicant information

Address after: 310052 Binjiang District Changhe Road, Zhejiang, China, No. 466, No.

Applicant after: Xinhua three Technology Co., Ltd.

Address before: 310052 Binjiang District Changhe Road, Zhejiang, China, No. 466, No.

Applicant before: Huasan Communication Technology Co., Ltd.

GR01 Patent grant
GR01 Patent grant