CN109391655A - Service gray scale dissemination method, device, system and storage medium - Google Patents
Service gray scale dissemination method, device, system and storage medium Download PDFInfo
- Publication number
- CN109391655A CN109391655A CN201710674368.XA CN201710674368A CN109391655A CN 109391655 A CN109391655 A CN 109391655A CN 201710674368 A CN201710674368 A CN 201710674368A CN 109391655 A CN109391655 A CN 109391655A
- Authority
- CN
- China
- Prior art keywords
- service
- gray scale
- server
- response results
- percentage
- 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.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/51—Discovery or management thereof, e.g. service location protocol [SLP] or web services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computer And Data Communications (AREA)
Abstract
Present application example provides a kind of service gray scale dissemination method, device, system and storage medium, this method comprises: S1', service gray scale distributing device setting gray scale issue the initial value of percentage;Current gray scale publication percentage is sent to second service system by S2', service gray scale distributing device;An at least second server in S3', second service system determines service request screening conditions, the service request for meeting service request screening conditions is filtered out from the service request received, and response is carried out to the service request filtered out and remaining service request is sent to first service system;An at least first server in S4', first service system responds remaining service request;S5', service gray scale distributing device obtain the first response results from second service system, if the correct and current gray scale publication percentage of the first response results increases current gray level publication percentage, and return to S2' less than 1.
Description
Technical field
This application involves server technology field more particularly to a kind of service gray scale dissemination methods, service gray scale publication dress
It sets, service gray scale delivery system and non-volatile computer readable storage medium storing program for executing.
Background technique
Present user uses various services on the internet, for example, the VIP user information of Tencent's video is inquired
Information sharing service etc. in service, wechat, these services usually require ceaselessly upgrading and update, preferably to realize its function
With meet user demand.The version publication of these services each time, can all influence several hundred million users, therefore network service now
Quotient, mostly can be using the method for gray scale publication when carrying out the publication of service new version.
Summary of the invention
The application provides a kind of service gray scale dissemination method, and this method is by service gray scale distributing device, first service system
And second service system executes;The first service system include an at least first server, described at least one first
The first service of first version is deployed on server;The second service system includes an at least second server, described
The first service of the second edition is deployed on an at least second server;The second edition is higher than first version;
The described method includes:
S1', the service gray scale distributing device setting gray scale issue the initial value of percentage;
Current gray scale publication percentage is sent to the second service system by S2', the service gray scale distributing device;
An at least second server in S3', the second service system issues percentage according to current gray scale
Than determining service request screening conditions, being filtered out from the service request received and meet the service request screening conditions
Service request carries out response to the service request filtered out and remaining service request is sent to the first service system
System;
An at least first server in S4', the first service system to remaining described service request into
Row response;
S5', the service gray scale distributing device obtain the first response results from the second service system, if described first
The correct and current gray scale publication percentage of response results then increases current gray level publication percentage less than 1, and returns to S2';Its
In, first response results are the first service of the second edition on an at least second server to the screening
The response results of service request out.
In some instances, the S3' can also include: an at least second server by the industry filtered out
Business request is sent to the first service system;And the S4' further include: an at least first server is to described
The service request filtered out is responded;Wherein, the whether correct method packet of the first response results is determined in the S5'
Include: the service gray scale distributing device obtains the second response results from the first service system, and according to first response
As a result determine whether first response results are correct with the comparing result of second response results;Second response results
For first version in an at least first server first service to the response results of the service request filtered out.
In some instances, if the S5' can also include: that the service gray scale distributing device determines first response
As a result there is any one mistake in, then corresponding second response results of the first response results of mistake are fed back into corresponding service request
Originating end.
In some instances, if the S5' can also include: that the service gray scale distributing device determines first response
As a result there is any one mistake in, then current gray scale publication percentage is set to upper gray scale publication percentage, and by the first sound
Should result the message of mistake occur be sent to monitoring system so that the monitoring system carries out positioning problems and amendment;And it is connecing
After receiving the message that the amendment that the monitoring system returns is completed, S2' is returned.
In some instances, if the S5' can also include: that the service gray scale distributing device determines first response
As a result correct and current gray scale publication percentage is 1, then indicates that the second service system disconnects and the first service system
Connection between system, so that an at least second server responds the whole service requests received.
Present application example provides a kind of service gray scale dissemination method, for passing through the of first service system and its connection
The first service of the two service systems publication second edition;The first service system includes an at least first server, described
The first service of first version is deployed in an at least first server;The second service system includes at least one second
Server is deployed with the first service of the second edition on an at least second server;The second edition is higher than first
Version;
The described method includes:
S1, setting gray scale issue the initial value of percentage;
S2, current gray scale publication percentage is sent to the second service system, so that described at least one second
Server issues percentage according to current gray scale, determines service request screening conditions, screens from the service request received
The service request for meeting the service request screening conditions out carries out response to the service request filtered out and by remaining industry
Business request be sent to the first service system so that an at least first server to remaining described service request into
Row response;
S3, the first response results are obtained from the second service system, first response results are at least one described
Response results of the first service of the second edition to the service request filtered out on second server;
If the correct and current gray scale publication percentage of S4, first response results increases current gray level hair less than 1
Cloth percentage, and return to S2.
In some instances, the service request filtered out is sent to described by an at least second server
One service system, so that an at least first server responds the service request filtered out;Described in determination
The whether correct method of first response results includes: to obtain the second response results, second sound from the first service system
Should result be first version in an at least first server sound of the first service to the service request filtered out
Answer result;Determine that first response results are according to the comparing result of first response results and second response results
It is no correct.
It in some instances, will be wrong if the S4, which can also include:, has any one mistake in first response results
Corresponding second response results of the first response results accidentally feed back to the originating end of corresponding service request.
In some instances, if the S4, which can also include:, has any one mistake in first response results, will work as
Preceding gray scale publication percentage is set to gray scale publication percentage, and the message that mistake occurs for the first response results is sent to
Monitoring system, so that the monitoring system carries out positioning problems and amendment;Receiving, the amendment that the monitoring system returns is complete
At message after, return S2.
In some instances, the method also includes: if first response results it is correct and current gray scale publication hundred
Divide than being 1, then indicates that the second service system disconnects the connection between the first service system, so that described at least one
Platform second server responds the whole service requests received.
Present application example provides a kind of service gray scale distributing device, for passing through the of first service system and its connection
The first service of the two service systems publication second edition;The first service system includes an at least first server, described
The first service of first version is deployed in an at least first server;The second service system includes at least one second
Server is deployed with the first service of the second edition on an at least second server;The second edition is higher than first
Version;
Described device includes:
Initialization module, setting gray scale issue the initial value of percentage;
Request screening module, current gray scale publication percentage be sent to the second service system so that it is described extremely
A few second server issues percentage according to current gray scale, service request screening conditions is determined, from the business received
The service request for meeting the service request screening conditions is filtered out in request, to the service request filtered out carry out response and
Remaining service request is sent to the first service system so that an at least first server to it is described remaining
Service request is responded;
As a result module is obtained, obtains the first response results from the second service system, first response results are institute
State response results of the first service to the service request filtered out of the second edition on an at least second server;
Ratio increases module, when the correct and current gray scale publication percentage of first response results is less than 1, increases
Current gray level issues percentage, and it is described by current gray scale publication percentage transmission to enable the first request screening module execute
To the processing of the second service system.
In some instances, the service request filtered out can also be sent to by an at least second server
The first service system, so that an at least first server responds the service request filtered out;Institute
Whether stating ratio to increase module may include result determination unit, correct with determination first response results;The result is true
Order member may include: acquisition subelement, obtain the second response results, second response results from the first service system
For first version in an at least first server first service to the response results of the service request filtered out;
It determines subelement, determines that first response is tied according to the comparing result of first response results and second response results
Whether fruit is correct.
In some instances, it can also include: feedback unit that the ratio, which increases module, in first response results
When having any one mistake, corresponding second response results of the first response results of mistake are fed back to the initiation of corresponding service request
End.
In some instances, it can also include: wrong reporting unit that the ratio, which increases module, in the first response knot
When having any one mistake in fruit, current gray scale publication percentage is set to a upper gray scale and issues percentage, and first is responded
As a result wrong message is sent to monitoring system, so that the monitoring system carries out positioning problems and amendment;It is described receiving
After the message that the amendment that monitoring system returns is completed, the first request screening module is enabled to execute described by current gray scale publication
Percentage is sent to the processing of the second service system.
In some instances, device can also include: that connection disconnects module, correct and current in first response results
Gray scale publication percentage when being 1, indicate that the second service system disconnects the connection between the first service system, with
Respond an at least second server to the whole service requests received.
Present application example provides a kind of service gray scale delivery system, comprising: first service system and second service system,
Wherein, the first service system includes an at least first server, and the is deployed in an at least first server
The first service of one version;The second service system includes an at least second server, an at least second service
The first service of the second edition is deployed on device;The second edition is higher than first version;The system also includes: for realizing
The service gray scale distributing device of any of the above-described method.
Present application example provides a kind of non-volatile computer readable storage medium storing program for executing, is stored thereon with computer program,
It is characterized in that, the step of program realizes any of the above-described method when being executed by processor.
Based on the above-mentioned technical proposal that present application example provides, although the deployment the on second server when starting
The first service of two versions, but be not all to respond all service requests by second server, but starting
One initial gray scale publication percentage is set, the service request of gray scale publication percentage is rung by second server
It answers, and this partial service request simultaneously and remaining service request are also responded by first server, then by first service
The response results of device and second server compare, and are determined whether to increase gray scale publication percentage according to comparing result.As it can be seen that
The application is not that gray scale publication is carried out by platform server, and it is unrelated with the number of units of server that gray scale issues percentage, therefore gray scale is sent out
The initial value and growth step-length of cloth percentage are not dependent on the number of units of server, can independently be arranged, even if the platform of server
Number seldom, can also realize the gray scale issuing process of small step-length, therefore can be to the hundred of impacted user according to realization demand
Divide ratio to be controlled, improves the effect of service release gray scale publication.
Detailed description of the invention
It, below will be to example or existing skill in order to illustrate more clearly of present application example or technical solution in the prior art
Attached drawing needed in art description is briefly described, it should be apparent that, the accompanying drawings in the following description is only the application
Some embodiments for those of ordinary skill in the art without any creative labor, can be with root
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is the model schematic that gray scale issuing process is serviced in one example of the application;
Fig. 2 is the system architecture figure that present application example is related to;
Fig. 3 is the flow diagram that gray scale dissemination method is serviced in one example of the application;
Fig. 4 is the flow diagram that gray scale dissemination method is serviced in one example of the application;
Fig. 5 is the structural schematic diagram that gray scale distributing device is serviced in one example of the application;
Fig. 6 is the flow diagram that gray scale dissemination method is serviced in one example of the application;
Fig. 7 is the model schematic before servicing gray scale publication in one example of the application;
Fig. 8~12 are the model schematics serviced in gray scale issuing process in one example of the application;
Figure 13 is the model schematic serviced after the completion of gray scale publication in one example of the application;
Figure 14 is the composed structure schematic diagram of computer equipment in one example of the application.
Specific embodiment
In one example of the application, as shown in Figure 1, it is assumed that a service needs to be deployed on n platform server, that is,
It says, the service that version x is run on n platform server is replaced with to the service of operation version x+1, inventor is in the clothes to new version
When business carries out gray scale publication, using following methods: being first to run a service publication service wherein for version x+1 by new version
On device, then second server is issued further according to the operating condition of this server, until n platform server is whole
Until being mounted with new version.Wherein, n is the number of units for needing to carry out the server of gray scale publication, for the integer greater than 1.
In issuing process, inventor has found following problems:
The above-mentioned number of units according to server carries out the process of gray scale publication, and the step-length of each gray scale publication is 1/n, if n
=2, that for the first time gray scale publication percentage be 50%, step-length is too big, it is seen that this gray scale published method for server number of units compared with
It can not achieve the gray scale publication of small step-length for few situation, and percentage issued due to gray scale and is determined by the number of units of server
It is fixed, so the percentage of affected user cannot be controlled arbitrarily.
In view of the above-mentioned problems, present application example provides a kind of service gray scale dissemination method, the applicable system tray of this method
Structure is as shown in Figure 2.The system architecture includes the originating end 101 of service request, first service system 102, second service system 103
With service gray scale distributing device 104;It is connected between each hardware device by internet 105 to realize information exchange, internet 105
It may include cable network and wireless network;Wherein:
The originating end 101 of above-mentioned service request can be user terminal, and the service request of initiation at this time can be referred to as
User's request, naturally it is also possible to be server, can also be what other hardware devices issued certainly, as long as issuing service request
Hardware device can be referred to as the originating end of service request.
Above-mentioned first service system 102 includes an at least first server, an at least first server top
There is the first service of first version in administration.
Above-mentioned second service system 103 includes an at least second server, an at least second server top
There is the first service of the second edition in administration, and the second edition is higher than first version.
So-called first service can be any one Internet service, for example, the VIP user information of Tencent's video is looked into
The service of inquiry, the information sharing service in wechat, video information search service etc..Due to first server and second server top
Administration is first service, and the version of the first service only disposed on second server, which is higher than, disposes in first server
The version of first service.In addition, the second clothes in first server and second service system 103 in first service system 102
The number of units of business device can be different, can also be identical.
Above-mentioned service gray scale distributing device 104, for example, service gray scale publisher server, independently of first service system 102
It is taken with second service system 103, second for passing through first service system 102 and the connection first service system 102
The first service of the publication second edition of business system 103.
Before being not carried out gray scale publication, service request is sent in second service system 103, then second service system
Service request is pass-through to first service system 102 by 103, is responded by first service system 102, to guarantee service request energy
Access correct response.As it can be seen that first service system 102 is the current system responded for service request, and the second clothes
The system that business system 103 is first service system 102 to be substituted to be responded for service request, so-called service is grey here
Degree dissemination method is exactly the process that second service system 103 is gradually replaced to first service system 102.Below to present application example
The service gray scale dissemination method of offer is introduced, and this method can be by above-mentioned service gray scale distributing device 104 (for example, service
Gray scale publisher server) it executes, with reference to Fig. 3, it can specifically include following steps:
S1, setting gray scale issue the initial value of percentage;
In practical application, the initial value of gray scale publication percentage can according to need self-setting, and not by servicing
The number of units of device determines.A smaller gray scale publication percentage can be set, such as be set as the lesser value such as 1%, 5%,
Once the first service of the second edition in this way goes wrong, the percentage of affected user can reduce.When gray scale issues percentage
When than being 0, service request is responded by the first server in first service system 102 completely.
S2, current gray scale publication percentage is sent to the second service system 103, so that described at least one the
Two servers issue percentage according to current gray scale, determine service request screening conditions, sieve from the service request received
Select the service request for meeting the service request screening conditions, to the service request filtered out carry out response and by remaining
Service request is sent to the first service system 102, so that an at least first server is to remaining described business
Request is responded;
Above-mentioned service request is sent in first service system 102 or second service system 103, by first service into
Row response, thus service request with first service be it is corresponding, because the request responded by first service can be just sent to
In first service system 102 or second service system 103, for example, the VIP user information that first service is Tencent's video inquires clothes
Business, then service request is the VIP user information inquiry request of Tencent's video, and other service requests will not be sent to the first clothes
It is engaged on device and second server.
In this step, there are many modes that service request screening conditions are determined according to current gray scale publication percentage,
One way in which is illustrated for using a certain query service as first service: assuming that current gray scale publication hundred
Divide than being 1%, above-mentioned service request is user's request, and the identity or logon account for issuing the user that user requests are 5
The range of the last two digits of bit digital, identity or logon account is 00~99, for the industry of this case setting
Business request screening conditions are that the last two digits of identity are 00, that is to say, that if second service system 103 receives
User to request the identity of corresponding user or the mantissa of logon account be 00, then screened and responded, it is right
It is sent to first service system 102 in user's request that identity or logon account mantissa are 01~99, by first service system
System 102 is responded.It is similar, if gray scale publication percentage is 5%, by identity or logon account mantissa be 00~
04 user's request screens.
In this step, first version is lower than the second edition, and for, first version is early version, first version
Service it is more stable, be able to carry out correct response, the second edition is new version, cannot determine if to be able to carry out correct
Response.So remaining service request is sent to first service system 102 by second service system 103, that is to say, that remaining
Service request responded by the first service of first version, to guarantee that remaining service request can correctly be rung
It answers.As for the service request filtered out, it is equivalent to the service request of test, is responded by second service system 103, also
It is to say, is responded by the first service of the second edition, so that it is determined that whether the first service of the second edition is correctly responded.
S3, the first response results are obtained from the second service system 103, first response results are described at least one
Response results of the first service of the second edition to the service request filtered out on platform second server;
If the correct and current gray scale publication percentage of S4, first response results increases current gray level hair less than 1
Cloth percentage, and return to S2.
In this step, if the first response results are correct, that is to say, that the first service of the second edition being capable of needle
The service request filtered out is correctly responded, and gray scale publication percentage at this time has not been reached yet 100%, then increases
Gray scale issues percentage, then returnes in step S2, continues to participate in circulation.
As an example it is assumed that the initial value of gray scale publication percentage is 1%, service gray scale distributing device 104 compares this
Example is sent to second service system 103, and each second server in such second service system 103 will be according to this ratio
It determines service request screening conditions, is then screened from the service request received, filter out and meet service request screening
Then the service request of condition responds the service request filtered out, and remaining service request is sent to the first clothes
Business system 102, such first service system 102 will respond remaining service request, to guarantee remaining industry
The response that business request can be striven for.For second service system 103 to the response results of the service request screened i.e.
First response results, service gray scale distributing device 104 are confirmed, if correctly, gray scale publication percentage increased to
5%, this ratio is then sent to second service system 103, second service system 103 is sieved according to 5% setting service request
The processes such as condition are selected, and so on, until gray scale publication percentage when the first response results of confirmation are correct is 100%.
Based on above-mentioned analysis it is found that in the service gray scale dissemination method that present application example provides, gray scale issues percentage
It is not to be determined by the number of units of server, can be arbitrarily arranged, even if the number of units of server is seldom, small step-length also may be implemented
Gray scale issuing process, therefore the percentage of impacted user can be controlled.
In some instances, determine that one way in which is there are many whether correct modes of above-mentioned first response results
Comparison, in order to which the response results of service request and the first service of first version ask business with the first service of the second edition
The response results asked compare, and can make second service system 103 that the service request screened is sent to first service
System 102, so that at least a first server responds the service request filtered out.That is, being screened
Service request out was both responded by the first service of the second edition in second service system 103, also by first service system
The first service of first version in system 102 is responded.The whether correct method packet of first response results is determined in this way
It includes:
The second response results are obtained from the first service system 102, second response results are at least one described
Response results of the first service of first version to the service request filtered out in first server;According to first sound
The comparing result of result and second response results is answered to determine whether first response results are correct.
Since the first service version of first version is higher to the response accuracy of service request, so if the first response
As a result be with the comparing result of the second response results it is consistent, then illustrate the first response results be it is correct, if inconsistent, explanation
First response results are incorrect, and then realize the judgement to the first response results.
It is understood that the first response results and the second response results that compare are for the same service request
, the first response results of each service request for being screened out are consistent with the comparison of the second response results, can just think to compare
As a result it is consistent, if the first response results and the second response results of one of service request are inconsistent, is not to be considered as pair
It is more consistent than result, therefore the operation for increasing gray scale publication percentage and return step S2 cannot be executed.In addition, passing through the side of comparison
Formula can also observe the operating condition of the first service of the second edition in real time.
In some instances, if there is any one mistake in the first response results, can mistake the first response results pair
The second response results answered feed back to the originating end 101 of corresponding service request.That is, wherein at least one service request
First response results and corresponding second response results are inconsistent, at this time second corresponding to that wrong first response results
Response results are sent to the transmitting terminal of service request, to guarantee that the transmitting terminal of service request can be responded correctly.
In some instances, if there is any one mistake in the first response results, current gray scale can be issued into percentage
Than being set to upper gray scale publication percentage, and the message of the first response results mistake is sent to monitoring system, so that institute
It states monitoring system and carries out positioning problems and amendment;After receiving the message that the amendment that the monitoring system returns is completed, return
S2。
For example, current gray scale publication percentage is 5%, learns there is at least one in the first response results by confirmation
It is a inconsistent with corresponding second response results, then it is that last gray scale issues percentage by gray scale publication percentage rollback
1%, and error message is reported to monitoring system, problem, which is carried out, by searching for log or other modes by monitoring system determines
, error correction is carried out after positioning problems, disappearing for amendment completion can be returned to service gray scale distributing device 104 after the completion of amendment
Breath returns to S2 after service gray scale distributing device 104 receives the message, continues to participate in circulation.The message hair of mistake will occur
It send to monitoring system and carries out the mode of positioning problems and reparation, be a kind of mode of active, need not wait for user feedback problem
It can just go to position and solve the problems, such as, can more timely find the problem and correct problem.
In some instances, if current gray scale publication percentage is 1 and the gray scale issues first under percentage
When response results are correct, illustrate to be substituted the first service of first version with the first service of the second edition come to service request
It is responded, the first service without reusing first version respond, and can indicate that second service system disconnects at this time
With the connection between first service system so that an at least second server carries out whole service requests for receiving
The publication work of the first service of the second edition is so far realized in response.
Illustrative explanation is carried out to the service gray scale dissemination method that present application example provides below with reference to Fig. 4;
S401, setting gray scale issue the initial value of percentage;
S402, current gray scale publication percentage is sent to the second service system, so that described at least one the
Two servers issue percentage according to current gray scale, determine service request screening conditions, sieve from the service request received
Select the service request for meeting the service request screening conditions, to the service request filtered out carry out response and by remaining
Service request is sent to the first service system, so that an at least first server is to remaining described service request
It is responded;
S403, the first response results are obtained from the second service system, first response results are described at least one
Response results of the first service of the second edition to the service request filtered out on platform second server;
S404, the second response results are obtained from the first service system, second response results are described at least one
Response results of the first service of first version to the service request filtered out in platform first server;
S405, judge whether each first response results and corresponding second response results are consistent;If consistent, execute
Otherwise current gray scale publication percentage is set to a upper gray scale and issues percentage, and first response results are wrong by S406
Message accidentally is sent to monitoring system, so that the monitoring system carries out positioning problems and amendment;Receiving the monitoring system
After the message that the amendment that system returns is completed, S402 is returned;
S406, judge whether current gray scale publication percentage is 100%;If 100%, then execute S407;Otherwise increase
Gray scale before sovolin issues percentage, and returns to S402;
Connection between S407, disconnection first service system and second service system, i.e. removal first service system, so far
Gray scale issuing process terminates.
It will be appreciated that the executing subject of above-mentioned steps S401~407 is service gray scale distributing device, such as service gray scale
Publisher server.The wherein ordinal relation between step S403 and S404 not necessarily, step S403 can prior to, simultaneously or after
It is executed in step S404.
In above process, issuing the initial value of percentage by setting gray scale and increase step-length may be implemented to send out gray scale
The control of cloth percentage, rather than the step-length of gray scale publication percentage is determined by the number of units of server.Moreover, asking
Monitoring system is reported to when topic in time, so that monitoring system can carry out positioning problems and reparation in time, it is seen that can send out in time
Existing problem, is solved the problems, such as by the way of active, rather than until user feedback just solves the problems, such as.Moreover, gray scale is issued hundred
Divide the gray scale publication percentage more last than rolling back to, reduces the influence to business as far as possible.
One example of the application provides a kind of service gray scale distributing device, for passing through first service system and its connection
The first service of the second service system publication second edition;The first service system includes an at least first server, institute
State the first service that first version is deployed in an at least first server;The second service system includes at least one the
Two servers are deployed with the first service of the second edition on an at least second server;The second edition is higher than the
One version;
As shown in figure 5, described device 500 includes:
Initialization module 501, setting gray scale issue the initial value of percentage;
Screening module 502 is requested, current gray scale publication percentage is sent to the second service system, so that described
An at least second server issues percentage according to current gray scale, service request screening conditions is determined, from the industry received
Business request in filter out the service request for meeting the service request screening conditions, to the service request filtered out responded with
And remaining service request is sent to the first service system so that an at least first server to it is described remaining
Service request responded;
As a result module 503 is obtained, obtains the first response results from the second service system, first response results are
Response knot of the first service of the second edition to the service request filtered out on an at least second server
Fruit;
Ratio increases module 504, when the correct and current gray scale publication percentage of first response results is less than 1,
Increase current gray level and issue percentage, and it is described by current gray scale publication hundred to enable the first request screening module 502 execute
Divide the processing than being sent to the second service system.
In some instances, the service request filtered out can also be sent to by an at least second server
The first service system, so that an at least first server responds the service request filtered out;Institute
Whether stating ratio to increase module 504 may include result determination unit, correct with determination first response results;The result
Determination unit may include:
Subelement is obtained, obtains the second response results from the first service system, second response results are described
Response results of the first service of first version to the service request filtered out in an at least first server;
It determines subelement, determines described according to the comparing result of first response results and second response results
Whether one response results are correct.
In some instances, the ratio, which increases module 504, to include:
Feedback unit, it is when having any one mistake in first response results, the first response results of mistake are corresponding
The second response results feed back to the originating end of corresponding service request.
In some instances, the ratio, which increases module 504, to include:
Current gray scale when having any one mistake in first response results, is issued percentage by mistake reporting unit
Than being set to upper gray scale publication percentage, and the message of the first response results mistake is sent to monitoring system, so that the prison
Control system carries out positioning problems and amendment;After receiving the message that the amendment that the monitoring system returns is completed, described the is enabled
One request screening module 502 executes the processing that current gray scale publication percentage is sent to the second service system.
In some instances, device 500 can also include:
Connection disconnects module, when the correct and current gray scale publication percentage of first response results is 1, indicates institute
The connection between the disconnection of second service system and the first service system is stated, so that an at least second server docking
The whole service requests received are responded.
It will be appreciated that the service gray scale distributing device that present application example provides is the function of above-mentioned service gray scale dissemination method
Energy structure module, the part such as explanation, citing, beneficial effect in relation to content can be with reference in above-mentioned service gray scale dissemination methods
Corresponding contents, details are not described herein again.
One example of the application provides a kind of service gray scale delivery system, which includes first service system, second service
System and service gray scale distributing device.Wherein:
The first service system includes an at least first server, is deployed in an at least first server
The first service of first version;The second service system includes an at least second server, at least one second clothes
The first service of the second edition is deployed on business device;The second edition is higher than first version;The service gray scale distributing device
For the hardware device for executing above-mentioned service gray scale dissemination method.
It will be appreciated that the service gray scale delivery system that present application example provides includes service gray scale distributing device, have
The part such as explanation, citing, beneficial effect for holding inside the Pass can refer to corresponding contents above, and details are not described herein again.
One example of the application provides a kind of service gray scale dissemination method, and this method is based on above-mentioned service gray scale delivery system,
As shown in fig. 6, this method generally comprises:
S1', the service gray scale distributing device setting gray scale issue the initial value of percentage;
Current gray scale publication percentage is sent to the second service system by S2', the service gray scale distributing device;
An at least second server in S3', the second service system issues percentage according to current gray scale
Than determining service request screening conditions, being filtered out from the service request received and meet the service request screening conditions
Service request carries out response to the service request filtered out and remaining service request is sent to the first service system
System;
An at least first server in S4', the first service system to remaining described service request into
Row response;
S5', the service gray scale distributing device obtain the first response results from the second service system, if described first
The correct and current gray scale publication percentage of response results then increases current gray level publication percentage less than 1, and returns to S2';Its
In, first response results are the first service of the second edition on an at least second server to the screening
The response results of service request out.
It will be appreciated that the portions such as explanation, citing, beneficial effect in present application example in step S1'~S5' in relation to content
Dividing can be with reference to corresponding portion above, and details are not described herein again.
In some instances, the S3' can also include: an at least second server by the industry filtered out
Business request is sent to the first service system;And the S4' can also include: an at least first server pair
The service request filtered out is responded;
In this way, determining that the whether correct method of the first response results includes: in the S5'
The service gray scale distributing device obtains the second response results from the first service system, and according to described first
The comparing result of response results and second response results determines whether first response results are correct;Second response
As a result the response for the first service of first version in an at least first server to the service request filtered out
As a result.
In some instances, if the S5' can also include: that the service gray scale distributing device determines first response
As a result there is any one mistake in, then corresponding second response results of the first response results of mistake are fed back into corresponding service request
Originating end.
In some instances, if the S5' can also include: that the service gray scale distributing device determines first response
As a result there is any one mistake in, then current gray scale publication percentage is set to upper gray scale publication percentage, and by the first sound
Should result the message of mistake occur be sent to monitoring system so that the monitoring system carries out positioning problems and amendment;And it is connecing
After receiving the message that the amendment that the monitoring system returns is completed, S2' is returned.
In some instances, if the S5' can also include: that the service gray scale distributing device determines first response
As a result correct and current gray scale publication percentage is 1, then indicates that the second service system disconnects and the first service system
Connection between system, so that an at least second server responds the whole service requests received.
In addition, in some instances, can be, but not limited to only include service in first service system and second service system
Device, such as can also include load balancing fault tolerance facility, so-called load balancing fault tolerance facility may also be referred to as L5, mainly make
With being to carry out load balancing and fault-tolerant, specifically: it collects the idle state of background server group, to being dealt into background server
Request provides the access of load balancing, to avoid certain server is in overload and what certain server was in idle condition
It happens;It also provides overload protection, judges whether to continue to send out to background server according to the state that background server returns packet
Send request, with prevent background server because need to handle it is requested cause pressure excessive and then break down, be also prevented from certain
Continue to send it request and cause snowslide after server failure.As it can be seen that service request transmitting terminal and background server it
Between a load balancing fault tolerance facility is set, background server can be played a certain protective role.
Here, it is fault-tolerant that the load balancing fault tolerance facility being arranged in first service system can be referred to as the first load balancing
Device, the load balancing fault tolerance facility being arranged in second service system can be referred to as the second load balancing fault tolerance facility.Its
In, the mapped port of the first load balancing fault tolerance facility includes first port, and first port corresponds to the first of the first version
Service;The mapped port of second load balancing fault tolerance facility includes second port, and second port corresponds to the of the second edition
One service.In this way when there is service request to be sent to second service system transmission service request, since the second load balancing is fault-tolerant
Device may map to the second port of second server, therefore the business personnel sent can be requested to be sent to second service
The second port of device is responded by the first service of the second edition on second server.When there is service request to be sent to first
When service system, the service request sent can be sent to the first end of first server by the first load balancing fault tolerance facility
Mouthful, it is responded by the first service of first version in first server.
It is that service request is responded by first service system before carrying out service gray scale publication in actual scene,
It is the second service system that is not present in entire model at this time, and when carrying out service gray scale publication, it needs in a model
Second service system is added, makes service request need first to be sent to second service system, is then pass-through to by second service system
First service system responds service request by first service system, and service gray scale distributing device is then recycled to carry out
Gray scale publication is serviced, second service system is made gradually to replace first service system, after servicing gray scale publication completion, by first
Service system is removed from model, and second service system is so far made to replace first service system completely.
Below using the VIP user information query service of Tencent's video as example, in conjunction with Fig. 7~13 pair service gray scale publication system
The process that system carries out service gray scale publication describes in detail:
The VIP user information query service of Tencent's video refers to specially inquires whether user is VIP for terminal user
The service at family, if the user of inquiry is VIP user, server can return to the grade of user, member's time started, Hui Yuanjie
Whether the beam time is the core sensitive informations associated with the member such as annual fee member, member's integral.First version is (for example, old version
Originally VIP user information query service) is the keyword using Vipid as system, is carried out to member backstage water system completely new
After transformation, the VIP user information query service of the second edition (for example, new version) uses keyword of the vuserid as system.
Referring to the following table 1, it is assumed that the VIP user information query service spp_server_old of first version is right on the server
The port answered is 13149, and port 13149 can be referred to as first port, the VIP user information query service spp_ of first version
Server_old is deployed on 13 first servers (for example, old server), in user terminal and this 13 first servers
Between be disposed with load balancing fault tolerance facility, which is denoted as the first load balancing fault tolerance facility OLD_L5,
The data group configured on OLD_L5 may map to the IP address and port 13149 of this 13 first servers, and above-mentioned 13
One server and the first load balancing fault tolerance facility OLD_L5 form first service system.User is looked by what its terminal was sent
The port 13149 that request reaches first server by the first load balancing fault tolerance facility OLD_L5 is ask, by the VIP of first version
User information query service spp_server_old responds inquiry request.
The corresponding data table related of table 1OLD_L5
Referring to the following table 2, the VIP user information query service spp_server_new of the second edition is corresponding on the server
Port is 20048, and port 20048 can be referred to as second port, the VIP user information query service spp_ of the second edition
Server_new is deployed on 4 second servers (for example, new demand servicing device), user terminal and this 4 second servers it
Between dispose load balancing fault tolerance facility, which is denoted as the second load balancing fault tolerance facility NEW_L5, NEW_
The data group configured on L5 may map to the IP address and port 13149 of 13 first servers, also may map to 4
The IP address of two servers and port 20048, therefore the inquiry request that user is sent by its terminal can be negative by second
The port 13149 that equilibrium fault tolerance facility NEW_L5 reaches first server is carried, by the VIP user information query service of first version
Spp_server_old responds inquiry request.The inquiry request that user is sent by its terminal can also be by second
Load balancing fault tolerance facility NEW_L5 reaches the port 20048 of second server, inquires clothes by the VIP user information of the second edition
Business spp_server_new responds inquiry request.
The corresponding data table related of table 2NEW_L
The model of the VIP user information query service of first version is run under current operating environment as shown in fig. 7, user is whole
The VIP user information query service (for example, operation version X) of first version in first server is accessed by OLD_L5 for end,
The number of units of server is 13 in the model, this 13 first servers do load balancing, corresponding service by OLD_L5
Port is first port, that is, port 13149.
Step S701, as shown in figure 8, a newly-built second load balancing fault tolerance facility NEW_L5, the NEW_L5 can reflect
The port penetrated includes first port i.e. port 13149 and second port i.e. port 20048.By the second load balancing fault tolerance facility
NEW_L5 replaces the first load balancing fault tolerance facility OLD_L5, and first server can be accessed by NEW_L5 in terminal at this time
First port, due to the operation of first server at this time or first version VIP user information query service, still by
The VIP user information query service of first version responds the inquiry request of user.
Step S702, as shown in figure 9, providing four second servers, and the first edition is disposed to this four second servers
This VIP user information query service, and this four second servers are all added in current model, specially the
Addition mapping data group in two load balancing fault tolerance facility NEW_L5, by the IP address and first port of this four second servers
It is added in the second load balancing fault tolerance facility NEW_L5, has 17 servers in such model altogether.Terminal can pass through second
Load balancing fault tolerance facility NEW_L5 accesses the first port of this 17 servers, by first version on this 17 servers
VIP user information query service responds service request.Due to inquiring at this time or by the VIP user information of first version
Service provides service for user, therefore will not influence the business runed on line at this time.
Step S703,13 first servers that as shown in Figure 10, the second load balancing fault tolerance facility NEW_L5 is mapped
It all removes, only provides service by 4 second servers.Due to being run on second server at this time or first version
VIP user information query service, user terminal access this 4 second servers by the second load balancing fault tolerance facility NEW_L5
First port, service request is responded by the VIP user information query service of first version, line will not be influenced at this time
The business of upper operation.
Step S704, as shown in figure 11, by 13 first servers and the first load balancing fault tolerance facility of removal
OLD_L5 is added in model again as first service system, is then used the VIP of 4 second server deployment second editions
Family information query service, the service request that user terminal issues are sent to the second port of second server by NEW_L5, but
It is that second server has the function of transparent transmission at this time, service request can be all pass-through to the first load balancing fault tolerance facility OLD_
Service request is sent to the first port of 13 first servers by the first load balancing fault tolerance facility OLD_L5, therefore by L5
Still service request is responded by the VIP user information query service of first version in first server at this time.Due to this
When actually or by first version VIP user information query service provide service, still do not influence the business run on line.
Step S705, as shown in figure 12, the addition service gray scale publisher server in above-mentioned model, and make the service gray scale
Publisher server is all connected with first service system, second service system, controls second service by service gray scale publisher server
The gray scale of system issues percentage, no matter how small step-length may be implemented.For example, control service gray scale publisher server gray scale hair
The initial value of cloth percentage is 1%, and the gray scale is then issued percentage to being sent to second service system, second service system
Issuing the last two digits that percentage determines that service request screening conditions are Vuserid according to the gray scale is 00, and then will
The service request that the last two digits of Vuserid are 01~99 is sent to the first load balancing fault tolerance facility OLD_L5, by the
The VIP user information query service of first version is responded on one server, is 00 by the last two digits of Vuserid
Service request is responded by the VIP user information query service of the second edition on second server, while also by this 1% industry
Business request is sent to the first load balancing fault tolerance facility OLD_L5, is looked by the VIP user information of first version in first server
The service of inquiry is responded.That is, each second server can respond 1% service request, also can this 1%
And the service request of residue 99% is sent to first service system, is responded by first service system.Then, gray scale is serviced
Publisher server obtains second service system from second service system and ties to the response of response results first of 1% service request
Then fruit obtains the response knot of response results i.e. second that first service system requests identical services from first service system
Then fruit compares and analyzes the first response results and the second response results, obtain the VIP user information for learning the second edition
Whether the response results of query service i.e. the first response results are correct.Judged whether to expand gray scale publication according to comparative analysis result
Percentage: if comparative analysis result is incorrect for the response results of the VIP user information query service of the second edition, occur different
Often, then the gray scale publication percentage that second service system is immediately controlled carries out being rolled back to 0%, the i.e. state of whole transparent transmissions, then
Error message is reported to monitoring system, positioning problems is carried out by monitoring system by searching for modes such as logs, is repaired, so
Adjusting gray scale publication percentage again afterwards is 1%, continues to control gray scale issuing process.If comparative analysis result is second
The response results of the VIP user information query service of version are correct, and there is no problem, then expand gray scale publication percentage, such as
It is expanded to 5%, continues comparative analysis.So go down, gray scale publication percentage is expanded to 100% always, that is,
100% service request by the VIP user information query service of first version i.e. both by being responded, also by the second edition
VIP user information query service is responded, and is all gone well if the response results of the two are analyzed by contrast, by 13 first
Server and the first load balancing fault tolerance facility OLD_L5 remove, completely by the VIP user of the second edition on second server
Information query service provides service, and the model so far obtained is as shown in figure 13.
Service request in above steps necessarily inquires the request of VIP user information, that is to say, that is and VIP
The corresponding request of user information query service.
In the examples described above, it in order to guarantee the smooth transition from Fig. 7 to Figure 12, reduces caused by issuing because of new demand servicing to user
Influence, first can map the second load balancing fault tolerance facility NEW_L5 of first port and second port using one to replace
First load balancing fault tolerance facility OLD_L5 (referring to Fig. 8) then increases second server (referring to Fig. 9) in a model, certainly
What is disposed on second server at this time is still the service (for example, service of operation version X) of first version, then will be owned
First server remove (referring to Figure 10), only second server is responded, although it is fault-tolerant so far to have replaced load balancing
Device and server, but still responded by the service of first version, therefore still will not influence the online of user at this time
Business.Then, the first server of removal and the first load balancing fault tolerance facility OLD_L5 are added in model, by the second clothes
The service of the second edition is disposed on business device, second server has the function of transparent transmission at this time, can be saturating by the service request received
It reaches in first service system (referring to Figure 11), user's is still unaffected in line service at this time.Then increase service gray scale hair
Cloth server, and on second server dispose the second edition service, then using service gray scale publisher server to second
The gray scale publication percentage of server is controlled, and is compared according to the operation result of second server and first server
Then analysis decides whether to increase gray scale publication percentage (referring to Figure 12), in the process, is likely to occur abnormal or wrong
Accidentally, (Fig. 7 to Figure 11) is not interfered in line service transient process before, can be use up by way of this smooth transition
Amount reduces influence of the service publishing process to business.
Present application example also provides a kind of non-volatile computer readable storage medium storing program for executing, is stored thereon with computer program,
The program realized when being executed by processor above-mentioned video platform word treatment method (such as: above-mentioned steps S1~S4, above-mentioned steps S401
~S407) the step of.
There are many above-mentioned storage mediums, for example, USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only
Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. are various can store journey
The medium of sequence code.
Present application example also provides a kind of computer equipment, which can be service gray scale publisher server, such as Figure 14
Shown, which includes one or more processor (CPU) 1402, communication module 1404, memory 1406, user
Interface 1410, and the communication bus 1408 for interconnecting these components, in which:
Processor 1402 can send and receive data by communication module 1404 to realize network communication and/or locally lead to
Letter.
User interface 1410 includes one or more output equipments 1412 comprising one or more speakers and/or one
A or multiple visual displays.User interface 1410 also includes one or more input equipments 1414 comprising such as, key
Disk, mouse, voice command input unit or loudspeaker, touch screen displays, touch sensitive tablet, posture capture camera or other are defeated
Enter button or control etc..
Memory 1406 can be high-speed random access memory, such as DRAM, SRAM, DDR RAM or other deposit at random
Take solid storage device;Or nonvolatile memory, such as one or more disk storage equipments, optical disc memory apparatus, sudden strain of a muscle
Deposit equipment or other non-volatile solid-state memory devices.
The executable instruction set of 1406 storage processor 1402 of memory, comprising:
Operating system 1416, including the journey for handling various basic system services and for executing hardware dependent tasks
Sequence;
Using 1418, including the various application programs issued for gray scale, this application program can be realized above-mentioned each reality
Process flow in example, for example may include some or all of instruction module or unit serviced in gray scale distributing device 500.
Processor 1402 passes through the machine-executable instruction executed in memory 1406 in each unit at least one unit, and then can
Realize the function of at least one module in above-mentioned each unit or module.
It should be noted that step and module not all in above-mentioned each process and each structure chart be all it is necessary, can
To ignore certain steps or module according to the actual needs.Each step execution sequence be not it is fixed, can according to need into
Row adjustment.The division of each module is intended merely to facilitate the division functionally that description uses, and in actual implementation, a module can
It is realized with point by multiple modules, the function of multiple modules can also be realized by the same module, these modules can be located at same
In a equipment, it can also be located in different equipment.
Hardware module in each embodiment can in hardware or hardware platform adds the mode of software to realize.Above-mentioned software
Including machine readable instructions, it is stored in non-volatile memory medium.Therefore, each embodiment can also be presented as software product.
In each example, hardware can be by special hardware or the hardware realization of execution machine readable instructions.For example, hardware can
Think the permanent circuit specially designed or logical device (such as application specific processor, such as FPGA or ASIC) for completing specifically to grasp
Make.Hardware also may include programmable logic device or circuit by software provisional configuration (as included general processor or other
Programmable processor) for executing specific operation.
In addition, each example of the application can pass through the data processor by data processing equipment such as computer execution
To realize.Obviously, data processor constitutes the application.In addition, being commonly stored data processing in one storage medium
Program is by directly reading out storage medium or the storage by program being installed or being copied to data processing equipment for program
It is executed in equipment (such as hard disk and/or memory).Therefore, such storage medium also constitutes the application, and present invention also provides one
Kind non-volatile memory medium, wherein being stored with data processor, this data processor can be used for executing in the application
State any one of method example example.
The corresponding machine readable instructions of Figure 14 module can make operating system operated on computer etc. to complete to retouch here
The some or all of operation stated.Non-volatile computer readable storage medium storing program for executing can be institute in the expansion board in insertion computer
In the memory of setting or write the memory being arranged in the expanding element being connected to a computer.Be mounted on expansion board or
CPU on expanding element etc. can be according to instruction execution part and whole practical operations.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Within mind and principle, any modification, equivalent substitution, improvement and etc. done be should be included within the scope of the present invention.
Claims (15)
1. a kind of service gray scale dissemination method, which is characterized in that by service gray scale distributing device, first service system and second
Service system executes;The first service system includes an at least first server, in an at least first server
It is deployed with the first service of first version;The second service system include an at least second server, described at least one
The first service of the second edition is deployed on second server;The second edition is higher than first version;
The described method includes:
S1', the service gray scale distributing device setting gray scale issue the initial value of percentage;
Current gray scale publication percentage is sent to the second service system by S2', the service gray scale distributing device;
An at least second server in S3', the second service system issues percentage according to current gray scale, really
Determine service request screening conditions, is filtered out from the service request received and meet the business of the service request screening conditions and ask
It asks, response is carried out to the service request filtered out and remaining service request is sent to the first service system;
An at least first server in S4', the first service system rings remaining described service request
It answers;
S5', the service gray scale distributing device obtain the first response results from the second service system, if first response
As a result correct and current gray scale publication percentage then increases current gray level publication percentage less than 1, and returns to S2';Wherein,
First response results are that the first service of the second edition on an at least second server is filtered out to described
Service request response results.
2. the method according to claim 1, wherein the S3' further include: an at least second server
The service request filtered out is sent to the first service system;And the S4' further include: described at least one
First server responds the service request filtered out;
Wherein, determine that the whether correct method of the first response results includes: in the S5'
The service gray scale distributing device obtains the second response results from the first service system, and according to first response
As a result determine whether first response results are correct with the comparing result of second response results;Second response results
For first version in an at least first server first service to the response results of the service request filtered out.
3. according to the method described in claim 2, it is characterized in that, the S5' further include:
If the service gray scale distributing device determines in first response results there is any one mistake, by the first sound of mistake
Corresponding second response results of result are answered to feed back to the originating end of corresponding service request.
4. according to the method described in claim 3, it is characterized in that, the S5' further include:
If the service gray scale distributing device determines in first response results there is any one mistake, current gray scale is sent out
Cloth percentage is set to gray scale publication percentage, and the message that mistake occurs for the first response results is sent to monitoring system,
So that the monitoring system carries out positioning problems and amendment;And in the message for receiving the amendment completion that the monitoring system returns
Afterwards, S2' is returned.
5. method according to any one of claims 1 to 4, which is characterized in that the S5' further include:
If the service gray scale distributing device determines first response results, correct and current gray scale publication percentage is 1,
Then indicate that the second service system disconnects the connection between the first service system, so that at least one second clothes
Business device responds the whole service requests received.
6. a kind of service gray scale dissemination method, which is characterized in that for the second service by first service system and its connection
The first service of the system publication second edition;The first service system include an at least first server, described at least one
The first service of first version is deployed in platform first server;The second service system includes an at least second service
Device is deployed with the first service of the second edition on an at least second server;The second edition is higher than first version;
The described method includes:
S1, setting gray scale issue the initial value of percentage;
S2, current gray scale publication percentage is sent to the second service system, so that an at least second service
Device issues percentage according to current gray scale, determines service request screening conditions, filters out from the service request received full
The service request of the foot service request screening conditions, carries out response to the service request filtered out and asks remaining business
It asks and is sent to the first service system, so that an at least first server rings remaining described service request
It answers;
S3, the first response results are obtained from the second service system, first response results are at least one second described
Response results of the first service of the second edition to the service request filtered out on server;
If the correct and current gray scale publication percentage of S4, first response results increases current gray level publication hundred less than 1
Divide ratio, and returns to S2.
7. according to the method described in claim 6, it is characterized in that, the service request filtered out is by described at least one
Two servers are sent to the first service system, so that an at least first server asks the business filtered out
It asks and is responded;
Determine that the whether correct method of first response results includes:
The second response results are obtained from the first service system, second response results are an at least first service
Response results of the first service of first version to the service request filtered out on device;
Whether first response results are determined according to the comparing result of first response results and second response results
Correctly.
8. the method according to the description of claim 7 is characterized in that the S4 further include:
If having any one mistake in first response results, by corresponding second response results of the first response results of mistake
Feed back to the originating end of corresponding service request.
9. according to the method described in claim 8, it is characterized in that, the S4 includes:
If having any one mistake in first response results, current gray scale publication percentage is set to a upper gray scale and is issued
Percentage, and the message that mistake occurs for the first response results is sent to monitoring system, so that the monitoring system carries out problem
Positioning and amendment;
After receiving the message that the amendment that the monitoring system returns is completed, S2 is returned.
10. according to any method of claim 6~9, which is characterized in that further include:
If the correct and current gray scale publication percentage of first response results is 1, indicate that the second service system is disconnected
The connection between the first service system is opened, so that an at least second server asks the whole business received
It asks and is responded.
11. a kind of service gray scale distributing device, which is characterized in that for the second service by first service system and its connection
The first service of the system publication second edition;The first service system include an at least first server, described at least one
The first service of first version is deployed in platform first server;The second service system includes an at least second service
Device is deployed with the first service of the second edition on an at least second server;The second edition is higher than first version;
Described device includes:
Initialization module, setting gray scale issue the initial value of percentage;
Screening module is requested, current gray scale publication percentage is sent to the second service system, so that described at least one
Platform second server issues percentage according to current gray scale, service request screening conditions is determined, from the service request received
In filter out the service request for meeting the service request screening conditions, to the service request filtered out carry out response and by its
Remaining service request is sent to the first service system, so that an at least first server is to remaining described business
Request is responded;
As a result obtain module, obtain the first response results from the second service system, first response results be it is described extremely
Response results of the first service of the second edition to the service request filtered out on a few second server;
Ratio increases module, when the correct and current gray scale publication percentage of first response results is less than 1, increases current
Gray scale issues percentage, and enables the first request screening module execute and described current gray scale publication percentage is sent to institute
State the processing of second service system.
12. device according to claim 11, which is characterized in that the service request filtered out is by described at least one
Second server is sent to the first service system, so that an at least first server is to the business filtered out
Request is responded;
Whether it includes result determination unit that the ratio, which increases module, correct with determination first response results;
The result determination unit includes:
Obtain subelement, obtain the second response results from the first service system, second response results be it is described at least
Response results of the first service of first version to the service request filtered out in one first server;
It determines subelement, first sound is determined according to the comparing result of first response results and second response results
Answer result whether correct.
13. device according to claim 12, which is characterized in that the ratio increases module further include:
Feedback unit, when having any one mistake in first response results, by the first response results of mistake corresponding the
Two response results feed back to the originating end of corresponding service request.
14. a kind of service gray scale delivery system characterized by comprising first service system and second service system, wherein
The first service system includes an at least first server, is deployed with first version in an at least first server
First service;The second service system includes an at least second server, an at least second server top
There is the first service of the second edition in administration;The second edition is higher than first version;The system also includes: for realizing such as right
It is required that the service gray scale distributing device of any one of 6~10 methods.
15. a kind of non-volatile computer readable storage medium storing program for executing, which is characterized in that be stored thereon with computer program, feature
The step of being, the method as any such as claim 1~10 realized when which is executed by processor.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710674368.XA CN109391655B (en) | 2017-08-09 | 2017-08-09 | Service gray level publishing method, device and system and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710674368.XA CN109391655B (en) | 2017-08-09 | 2017-08-09 | Service gray level publishing method, device and system and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109391655A true CN109391655A (en) | 2019-02-26 |
CN109391655B CN109391655B (en) | 2021-09-24 |
Family
ID=65414687
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710674368.XA Active CN109391655B (en) | 2017-08-09 | 2017-08-09 | Service gray level publishing method, device and system and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109391655B (en) |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110120971A (en) * | 2019-04-17 | 2019-08-13 | 北京奇艺世纪科技有限公司 | A kind of gray scale dissemination method, device and electronic equipment |
CN110515675A (en) * | 2019-08-30 | 2019-11-29 | 中国工商银行股份有限公司 | Using online method, apparatus, system and readable storage medium storing program for executing |
CN110647335A (en) * | 2019-07-26 | 2020-01-03 | 平安科技(深圳)有限公司 | Software gray level release method and device, electronic equipment and storage medium |
CN110737548A (en) * | 2019-10-28 | 2020-01-31 | 聚好看科技股份有限公司 | Data request method and server |
CN111078468A (en) * | 2019-11-21 | 2020-04-28 | 贝壳技术有限公司 | Service rollback method and device under micro-service architecture |
CN111399875A (en) * | 2020-03-06 | 2020-07-10 | 咪咕文化科技有限公司 | Gray scale upgrade control method and device, electronic equipment and storage medium |
CN111488170A (en) * | 2020-04-07 | 2020-08-04 | 支付宝(杭州)信息技术有限公司 | Method, device and equipment for updating business processing model |
CN112181443A (en) * | 2019-07-01 | 2021-01-05 | 中国移动通信集团浙江有限公司 | Automatic service deployment method and device and electronic equipment |
CN112202929A (en) * | 2020-12-01 | 2021-01-08 | 湖南新云网科技有限公司 | Service access method, device and equipment in micro-service architecture |
CN112256307A (en) * | 2020-11-09 | 2021-01-22 | 北京沃东天骏信息技术有限公司 | Gray scale distribution method and device, electronic equipment and computer readable medium |
CN112685287A (en) * | 2020-11-30 | 2021-04-20 | 青岛海尔科技有限公司 | Product data testing method and device, storage medium and electronic device |
CN113051182A (en) * | 2021-04-30 | 2021-06-29 | 中国工商银行股份有限公司 | Gray scale publishing method, device and equipment based on distributed transaction |
CN113542387A (en) * | 2021-07-09 | 2021-10-22 | 平安银行股份有限公司 | System publishing method, device, electronic equipment and storage medium |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20020098468A1 (en) * | 2001-01-23 | 2002-07-25 | Avatar Technology, Inc. | Method for constructing and teaching a curriculum |
CN103176790A (en) * | 2011-12-26 | 2013-06-26 | 阿里巴巴集团控股有限公司 | Application releasing method and application releasing system |
CN103577660A (en) * | 2012-07-19 | 2014-02-12 | 腾讯科技(深圳)有限公司 | Gray scale experiment system and method |
CN104102576A (en) * | 2013-04-12 | 2014-10-15 | 阿里巴巴集团控股有限公司 | Multi-version test method and device |
CN104780057A (en) * | 2014-01-13 | 2015-07-15 | 中兴通讯股份有限公司 | Version upgrade processing method and device |
US20160103758A1 (en) * | 2014-10-08 | 2016-04-14 | Yahoo! Inc. | Online product testing using bucket tests |
CN106066791A (en) * | 2016-05-25 | 2016-11-02 | 合网络技术(北京)有限公司 | The gray scale dissemination method of application and device |
-
2017
- 2017-08-09 CN CN201710674368.XA patent/CN109391655B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20020098468A1 (en) * | 2001-01-23 | 2002-07-25 | Avatar Technology, Inc. | Method for constructing and teaching a curriculum |
CN103176790A (en) * | 2011-12-26 | 2013-06-26 | 阿里巴巴集团控股有限公司 | Application releasing method and application releasing system |
CN103577660A (en) * | 2012-07-19 | 2014-02-12 | 腾讯科技(深圳)有限公司 | Gray scale experiment system and method |
CN104102576A (en) * | 2013-04-12 | 2014-10-15 | 阿里巴巴集团控股有限公司 | Multi-version test method and device |
CN104780057A (en) * | 2014-01-13 | 2015-07-15 | 中兴通讯股份有限公司 | Version upgrade processing method and device |
US20160103758A1 (en) * | 2014-10-08 | 2016-04-14 | Yahoo! Inc. | Online product testing using bucket tests |
CN106066791A (en) * | 2016-05-25 | 2016-11-02 | 合网络技术(北京)有限公司 | The gray scale dissemination method of application and device |
Non-Patent Citations (2)
Title |
---|
Y. OJIMA等: "Determining optimal filters for binarization of degraded grayscale characters using genetic algorithms", 《EIGHTH INTERNATIONAL CONFERENCE ON DOCUMENT ANALYSIS AND RECOGNITION (ICDAR"05)》 * |
屈文建等: "网络学术论坛信息可信度的灰度分析", 《图书情报知识》 * |
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110120971A (en) * | 2019-04-17 | 2019-08-13 | 北京奇艺世纪科技有限公司 | A kind of gray scale dissemination method, device and electronic equipment |
CN112181443A (en) * | 2019-07-01 | 2021-01-05 | 中国移动通信集团浙江有限公司 | Automatic service deployment method and device and electronic equipment |
CN110647335A (en) * | 2019-07-26 | 2020-01-03 | 平安科技(深圳)有限公司 | Software gray level release method and device, electronic equipment and storage medium |
CN110515675A (en) * | 2019-08-30 | 2019-11-29 | 中国工商银行股份有限公司 | Using online method, apparatus, system and readable storage medium storing program for executing |
CN110737548A (en) * | 2019-10-28 | 2020-01-31 | 聚好看科技股份有限公司 | Data request method and server |
CN110737548B (en) * | 2019-10-28 | 2022-06-10 | 聚好看科技股份有限公司 | Data request method and server |
CN111078468A (en) * | 2019-11-21 | 2020-04-28 | 贝壳技术有限公司 | Service rollback method and device under micro-service architecture |
CN111399875B (en) * | 2020-03-06 | 2023-09-05 | 咪咕文化科技有限公司 | Gray scale upgrading control method and device, electronic equipment and storage medium |
CN111399875A (en) * | 2020-03-06 | 2020-07-10 | 咪咕文化科技有限公司 | Gray scale upgrade control method and device, electronic equipment and storage medium |
CN111488170A (en) * | 2020-04-07 | 2020-08-04 | 支付宝(杭州)信息技术有限公司 | Method, device and equipment for updating business processing model |
CN112256307A (en) * | 2020-11-09 | 2021-01-22 | 北京沃东天骏信息技术有限公司 | Gray scale distribution method and device, electronic equipment and computer readable medium |
CN112685287A (en) * | 2020-11-30 | 2021-04-20 | 青岛海尔科技有限公司 | Product data testing method and device, storage medium and electronic device |
CN112685287B (en) * | 2020-11-30 | 2023-11-28 | 青岛海尔科技有限公司 | Product data testing method and device, storage medium and electronic device |
CN112202929A (en) * | 2020-12-01 | 2021-01-08 | 湖南新云网科技有限公司 | Service access method, device and equipment in micro-service architecture |
CN113051182A (en) * | 2021-04-30 | 2021-06-29 | 中国工商银行股份有限公司 | Gray scale publishing method, device and equipment based on distributed transaction |
CN113051182B (en) * | 2021-04-30 | 2024-02-09 | 中国工商银行股份有限公司 | Gray level publishing method, device and equipment based on distributed transaction |
CN113542387B (en) * | 2021-07-09 | 2023-07-04 | 平安银行股份有限公司 | System release method and device, electronic equipment and storage medium |
CN113542387A (en) * | 2021-07-09 | 2021-10-22 | 平安银行股份有限公司 | System publishing method, device, electronic equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN109391655B (en) | 2021-09-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109391655A (en) | Service gray scale dissemination method, device, system and storage medium | |
US11586673B2 (en) | Data writing and reading method and apparatus, and cloud storage system | |
CN109842651B (en) | Uninterrupted service load balancing method and system | |
CN112491606B (en) | Method for automatically deploying high-availability cluster of service system based on infrastructure | |
CN108881512B (en) | CTDB virtual IP balance distribution method, device, equipment and medium | |
CN109173270B (en) | Game service system and implementation method | |
CN103607424B (en) | Server connection method and server system | |
CN105933408A (en) | Implementation method and device of Redis universal middleware | |
CN108491163B (en) | Big data processing method and device and storage medium | |
CN113347037B (en) | Data center access method and device | |
CN107612950A (en) | A kind of method, apparatus, system, electronic equipment that service is provided | |
CN111240892A (en) | Data backup method and device | |
CN115454333A (en) | Docking method and device for cloud computing platform and storage system | |
CN103905512B (en) | A kind of data processing method and equipment | |
CN107943615B (en) | Data processing method and system based on distributed cluster | |
CN109445988A (en) | Isomery disaster recovery method, device, system, server and disaster recovery platform | |
CN109002373A (en) | Database connects abnormality eliminating method and device | |
CN110413523B (en) | Drainage testing method, drainage testing middle table and computer readable storage medium | |
CN107291575B (en) | Processing method and equipment for data center fault | |
US7962799B2 (en) | System and method for synchronizing test runs on separate systems | |
CN110502238A (en) | A kind of method and device of front and back end joint debugging | |
CN107566475B (en) | Session failover method and device | |
CN114257505B (en) | Server node configuration method, device, equipment and storage medium | |
CN102156685A (en) | Method and device for automatically configuring distributed network system script | |
CN110532135B (en) | Service migration method, system, equipment and storage medium of cluster system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |