CN114615192A - Distributed full-link gray level publishing method and device - Google Patents
Distributed full-link gray level publishing method and device Download PDFInfo
- Publication number
- CN114615192A CN114615192A CN202210143050.XA CN202210143050A CN114615192A CN 114615192 A CN114615192 A CN 114615192A CN 202210143050 A CN202210143050 A CN 202210143050A CN 114615192 A CN114615192 A CN 114615192A
- Authority
- CN
- China
- Prior art keywords
- link
- node
- traffic
- request
- label
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000012360 testing method Methods 0.000 claims description 33
- 238000004043 dyeing Methods 0.000 claims description 21
- 238000004040 coloring Methods 0.000 claims description 4
- 230000007613 environmental effect Effects 0.000 claims description 3
- 239000000975 dye Substances 0.000 abstract description 8
- 238000010586 diagram Methods 0.000 description 9
- 238000005096 rolling process Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/50—Routing or path finding of packets in data switching networks using label swapping, e.g. multi-protocol label switch [MPLS]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/14—Routing performance; Theoretical aspects
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/31—Flow control; Congestion control by tagging of packets, e.g. using discard eligibility [DE] bits
Abstract
The invention relates to the field of distributed systems, and provides a distributed full-link gray level publishing method and a distributed full-link gray level publishing device, which comprise the following steps: respectively adding node labels to all link nodes in distributed deployment, and grouping the link nodes according to different dimensions according to the node labels; issuing an identification rule of a request flow and a routing rule of the request flow to a link node; when the link node receives the request traffic which is not dyed, the link node dyes and marks the received request traffic according to the traffic identification rule, and then other link nodes with matched dimensionality are searched according to the traffic label of the request traffic and the routing rule of the request traffic; and when the link node receives the dyed request traffic, the link node calls the matched application according to the received traffic label of the request traffic. The invention adopts the label routing mode to configure the nodes, effectively improves the configuration efficiency, matches according to the flow label of the current request flow when calling the application, and improves the fault tolerance rate of the flow routing.
Description
Technical Field
The invention relates to the technical field of distributed systems, in particular to a distributed full-link gray level publishing method and device.
Background
The gray scale publishing is an effective means for avoiding the publishing risk in the industry, and can be realized in a plurality of modes such as blue-green deployment, rolling publishing, gray scale publishing and the like. The blue-green release needs to run the applications of two versions at the same time, the applications of the new versions are deployed under the condition that the original applications are not changed, and then the flow is switched to the new versions after the new applications pass the test. However, the blue green release requires the existence of both new and old applications, which consumes twice the resources and causes a problem of high cost. And in the release process of the rolling release, the old version is upgraded into the new version in place in batches until the upgrade is completed. The rolling release does not consume extra resources, but has high risk, once the rolling upgrade is started, the traffic of a user flows to a new version, and the new version is not completely tested, so that the system is in an unstable state during the release.
The current gray release technology requires that a new version application is started first, but the flow of a user is not switched temporarily, and a tester is enabled to test the new version on line. And after the user traffic passes the test, gradually switching the user traffic to a new version. The scheme can ensure the stability of the online system under the condition of not increasing the cost. However, the existing gray scale distribution is only suitable for a single node, and when a plurality of link nodes exist in a distributed system, each node needs to be configured, so that the problem of low efficiency exists. And when a matching node cannot be found, a large number of requests are easy to fail, and further an avalanche phenomenon is caused to a container in the distributed system.
Disclosure of Invention
The invention provides a distributed full-link gray scale publishing method and a distributed full-link gray scale publishing system for overcoming the defects of low configuration efficiency and low fault tolerance rate of gray scale publishing in a distributed system in the prior art.
In order to solve the technical problems, the technical scheme of the invention is as follows:
a distributed full-link gray level release method comprises the following steps: respectively adding at least one node label to all link nodes in distributed deployment, and grouping each link node according to different dimensions according to the node labels; issuing an identification rule of a request flow and a routing rule of the request flow to the link node; when a link node receives an undyed request flow, the link node performs dyeing marking on the received request flow according to the flow identification rule, namely marks a corresponding flow label on the request flow, and then searches other link nodes with matched dimensions according to the flow label of the request flow and the routing rule of the request flow; and when the link node receives the request traffic marked by dyeing, the link node calls the matched application according to the received traffic label of the request traffic.
In the technical scheme, a label routing mode is adopted, and a service framework is combined to accurately guide user traffic in a distributed system to certain specific nodes or machines. The link nodes in the distributed system are added with at least one node label, and the node labels are used as loose metadata, so that the grouping of machines can be flexibly configured. For the request traffic sent by the user, the node dyes and marks the matched traffic according to the issued configuration, namely, a traffic label is marked, the traffic label can be transmitted in the whole link, and when the application calls a downstream application, the matched node and the application can be searched according to the traffic label of the current request traffic, so that the configuration efficiency and the fault tolerance rate are improved.
As a preferred scheme, when the link node performs color marking on the received request traffic according to the traffic identification rule, judging according to a user value in an HTTP request Header, and when the user value is a normal user, color marking the request traffic as a normal user label; and when the user value is the test user, the request traffic is dyed and marked as a test user label.
Preferably, the applications in the link node include a normal version application and/or a gray version application.
As a preferred scheme, when the link node calls a matched application according to the received traffic label of the request traffic, the method includes the following steps: when the link node comprises normal version application and gray version application, calling normal version application for the request traffic marked as a normal user and calling gray version application for the request traffic marked as a test user according to the traffic label of the request traffic received by the link node; when the link node only contains a normal version application, calling the normal version application for the request flow marked as a normal user, and ending the request for the request flow marked as a test user; and when the link node only contains the gray version application, calling the gray version application for the request traffic of the dye mark of the normal user and the test user.
Preferably, the dimensions of the node labels added to the link nodes include environment, machine configuration, and organizational framework departments.
Preferably, when adding a node label to the link node, the method includes one or more of the following identification operations: hard-coded designations in application code, dynamic configuration using a configuration center, configuration through environmental variables, and configuration using a container orchestration tool.
Preferably, the method further comprises the following steps: after obtaining the node label, the link node registers registration information to a registration center, wherein the registration information comprises a machine IP (Internet protocol) to which the link node belongs and the node label; and the link nodes carry out grouping with different dimensionalities according to the registration information stored in the registration center.
Further, the invention provides a flow routing device based on a distributed system, which is applied to the distributed system, wherein the distributed system comprises distributed link nodes, and normal version application and/or gray version application are/is arranged in the link nodes; the traffic routing device comprises a node identification module and a rule configuration module, wherein the node identification module is used for identifying the link nodes with node labels with different dimensions and grouping the link nodes according to different dimensions; the rule configuration module is used for configuring an identification rule of request traffic and a routing rule of the request traffic to the link node.
In the technical scheme, when the link node receives an undyed request flow, the link node performs dyeing marking on the received request flow according to the flow identification rule, and then searches other link nodes with matched dimensions according to a flow label of the request flow and a routing rule of the request flow; and when the link node receives the request traffic marked by dyeing, the link node calls the matched normal version application or gray version application according to the traffic label of the received request traffic.
As a preferred scheme, the traffic routing device further comprises a registration center, and the link node acquires the node label and registers registration information to the registration center, wherein the registration information comprises the machine IP to which the link node belongs and the node label; and the link nodes carry out grouping with different dimensionalities according to the registration information stored in the registration center.
Preferably, the node identification module comprises at least one of a code specification unit, a dynamic configuration unit, an environment variable configuration unit and a container arrangement unit, wherein the code specification unit is used for adding a node tag to a specified link node in application code through hard coding; the dynamic configuration unit is used for dynamically configuring the node labels of the link nodes; the environment variable configuration unit is used for configuring the node labels of the link nodes through environment variables; the container arranging unit is used for configuring the node labels of the link nodes through a container arranging tool.
Compared with the prior art, the technical scheme of the invention has the beneficial effects that: the invention adopts a label routing mode, realizes the accurate guidance of user flow in a distributed system to certain specific nodes or machines by combining a service framework, and can effectively improve the configuration efficiency; for the request traffic sent by the user, the node performs dyeing marking on the matched traffic according to the issued configuration, and when the application calls the downstream application, the matched node and the application can be searched according to the traffic label of the current request traffic, so that the fault tolerance rate of the traffic route can be effectively improved.
Drawings
Fig. 1 is a flowchart of a traffic routing method based on a distributed system according to embodiment 1.
Fig. 2 is a schematic flow diagram of traffic routing in embodiment 1.
Fig. 3 is a flowchart of a traffic routing method based on a distributed system according to embodiment 2.
Fig. 4 is a diagram illustrating the link node registration of embodiment 2.
Fig. 5 is a schematic diagram of a link node rule configuration according to embodiment 2.
Fig. 6 is a schematic view of traffic routing in which a link node of embodiment 2 only includes a gray version application.
Fig. 7 is an architecture diagram of a traffic routing apparatus based on a distributed system according to embodiment 3.
Detailed Description
The drawings are for illustrative purposes only and are not to be construed as limiting the patent;
for the purpose of better illustrating the embodiments, certain features of the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product;
it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The technical solution of the present invention is further described below with reference to the accompanying drawings and examples.
Example 1
The present embodiment provides a traffic routing method based on a distributed system, and as shown in fig. 1, the method is a flowchart of the traffic routing method based on the distributed system according to the present embodiment.
The traffic routing method based on the distributed system provided by the embodiment comprises the following steps:
and S1, respectively adding at least one node label to all the link nodes deployed in a distributed manner, and grouping each link node according to different dimensions according to the node labels.
S2, issuing the identification rule of the request traffic and the routing rule of the request traffic to the link node.
S3, when the link node receives the request flow which is not dyed, the link node dyes and marks the received request flow according to the flow identification rule, and then searches other link nodes with matched dimensionality according to the flow label obtained by the request flow through the dyeing and marking and the routing rule of the request flow;
and when the link node receives the request traffic marked by dyeing, the link node calls the matched application according to the received traffic label of the request traffic.
In this embodiment, node tags are added to the machine or link nodes in the distributed system, and the node tags are used as loose metadata, so that the grouping of the machine or link nodes can be flexibly configured.
In the process of dyeing and marking the requested traffic, the present embodiment issues rules to the link node through the control plane, where the rules include an identification rule of the requested traffic and a routing rule of the requested traffic, and are used to describe what traffic belongs to which group. And the link node performs dyeing marking on the matched traffic according to the configured rule, namely, a traffic label is marked, and the traffic label identified by the request traffic can be transmitted in the whole link. And when the link node calls the downstream application, the matched node is found according to the current requested flow label.
In a specific implementation process, as shown in fig. 2, a schematic flow diagram of the traffic routing of this embodiment is shown. A, B, C three link nodes are deployed in the distributed system in the embodiment, each link node comprises normal version application and gray version application, wherein the normal version application is a normal version and is accessed by a normal user; the gray version is applied as a test version for the test user to access, and when the test user completes the test on line, the test version is accessed by the normal user after confirming that the function is correct. When a link node A receives a request flow sent by a user, judging whether the current request flow is dyed or not, if not, dyeing and marking the received request flow by the link node A according to a issued flow identification rule, and then searching other link nodes with matched dimensions according to a flow label obtained by dyeing and marking the request flow and an issued routing rule of the request flow; and if the request flow is marked by the dyeing, the link node A calls the matched application according to the received flow label of the request flow.
Specifically, if the request traffic is sent by a normal user, the request traffic is dyed and marked as a normal user label, and at this time, the link node a routes the request traffic to a normal user packet and invokes a normal version application. And if the request flow is sent by the test user, the request flow is gray level flow, the request flow is dyed and marked as a test user label, and at the moment, the link node A routes the request flow to a gray level group and calls the gray version application.
Meanwhile, when the link node a has color-marked or identified the request traffic, for example, when the traffic label of the request traffic is a normal user, the link node a routes the request traffic to the normal version application in the link node B, and after completing the application access, the link node B routes the request traffic to the normal version application in the link node C. The load balancing algorithm needs a service framework in an application to be uniformly realized and is transparent to business logic.
The embodiment is suitable for a full-link distributed system, and the label routing mode is adopted, and the service framework is combined to realize that the user traffic in the distributed system is accurately directed to certain specific nodes or machines. The link nodes in the distributed system are added with at least one node label, and the node labels are used as loose metadata, so that the grouping of machines can be flexibly configured. For the request traffic sent by the user, the node performs dyeing marking on the matched request traffic according to the issued configuration, namely, the corresponding traffic label is marked on the request traffic, the traffic label can be transmitted in the whole link, and when the application calls the downstream application, the matched node and the application can be searched according to the traffic label of the current request traffic, so that the configuration efficiency and the fault tolerance rate are improved.
Example 2
The present embodiment provides a traffic routing method based on a distributed system, and as shown in fig. 3, the method is a flowchart of the traffic routing method based on the distributed system according to the present embodiment.
The traffic routing method based on the distributed system provided by the embodiment comprises the following steps:
and S1, respectively adding at least one node label to all the link nodes in the distributed deployment, and grouping each link node according to different dimensions according to the node labels.
In this embodiment, loose metadata is used as a node tag for grouping link nodes. When the node label is added to the link node, the method comprises one or more of the following identification operations: hard-coded in application code, dynamic configuration using a configuration center, configuration through environment variables, and configuration with container orchestration tools.
When the identification operation is adopted, the link node can realize that the link node automatically adds the node label, and after the node label is added, the configuration can be dynamically adjusted and configured through a configuration center and the like.
In the embodiment, the number of nodes in the distributed system is large, unified management is difficult to perform, and node labels such as classification and attribute of each node are only clear to a user who manages the node, so that the node can be classified quickly and effectively by adding the node labels by the link nodes.
Further, the link node acquires the node label and registers registration information to a registration center, wherein the registration information comprises a machine IP (Internet protocol) to which the link node belongs and the node label; and the link nodes carry out grouping with different dimensionalities according to the registration information stored in the registration center. Fig. 4 is a schematic diagram illustrating the link node registration according to this embodiment.
The dimensions of the node labels added to the link nodes in this embodiment include the environment, machine configuration, and organizational framework departments. When grouping link nodes, the link nodes may be grouped by, for example, environment (development, test, production), machine configuration (low-configuration, medium-configuration, high-frequency. cpu is high, large disk space, large network bandwidth), organization and architecture department (technical department, sales department, product department).
S2, issuing the identification rule of the request traffic and the routing rule of the request traffic to the link node.
In this embodiment, the identification rule of the request traffic includes determining according to a user value in an HTTP request Header, and when the user value is a normal user, dyeing and marking the request traffic as a normal user tag; and when the user value is the test user, the request traffic is dyed and marked as a test user label. Fig. 5 is a schematic diagram of a link node rule configuration according to this embodiment.
The routing rule of the request traffic comprises a traffic-link node label rule of the request traffic, and is used for routing the request traffic identified as the corresponding traffic label to a link node label packet stored and matched in the rule.
In a specific implementation process, since the link node is added with at least one node label, that is, there is a case where there are multiple node labels in one link node, for example, the link node a may belong to both "development", "low-configuration" and "large disk space" packets. Then traffic will enter link node a when the rule is configured as "XXX traffic is routed to development packets"; when the rule is configured as "XXX traffic is routed to disk space big packets," traffic will also enter link node a.
S3, when the link node receives the request flow which is not dyed, the link node dyes and marks the received request flow according to the flow identification rule, and then searches other link nodes with matched dimensionality according to the flow label obtained by the request flow through the dyeing and marking and the routing rule of the request flow; and when the link node receives the request traffic marked by dyeing, the link node calls the matched application according to the received traffic label of the request traffic.
In this embodiment, the applications in the link nodes deployed in a distributed manner include normal version applications and/or gray version applications.
When the link node calls the matched application according to the received flow label of the request flow, the method comprises the following steps:
when the link node comprises a normal version application and a gray version application, calling the normal version application for the request flow marked by the dye as a normal user label and calling the gray version application for the request flow marked by the dye as a test user label according to the flow label of the request flow received by the link node;
when the link node only contains a normal version application, calling the normal version application for the request flow marked as a normal user label, and ending the request for the request flow marked as a test user label;
and when the link node only contains the gray version application, calling the gray version application for the request flow marked by the dye as the normal user label and the test user label.
In a specific implementation process, the link node B has a normal version application and a gray version application, wherein the normal version application is a normal version application for a normal user; the gray version application is a new online test version application, and needs a tester to perform online test and make a user access after confirming that the function is correct.
When the link node B starts up, the link node B can obtain the environmental node labels and automatically classify the applications in the link node B into a certain logical grouping. In this example, there are two groups, normal and gray. Similarly, link node C also has normal version application and gray version application, which are classified into normal group and gray group, respectively.
The request traffic of both the normal user and the test user passes through the link node a. The link node A acquires the identification rule of the request traffic and the routing rule of the request traffic from the control plane. The link node A judges according to a user value in the HTTP request Header, if the user value is a normal user, the request flow is dyed and marked as a normal user label, and the request flow is routed to a normal group; if the user value is a test user, the request traffic is colored to be labeled as a test user label and then routed to the gray packet.
For normal user requested traffic, when link node a routes the traffic request to link node B in the normal group, link node B invokes its normal version application, and link node B routes the traffic request to link node C in the normal group, link node C also invokes its normal version application. The load balancing algorithm needs a service framework in an application to be uniformly realized and is transparent to business logic.
Further, when the link node C has no normal group, that is, no normal version application, but only a gray version application, according to the conventional routing flow, the link node B cannot find the link node C corresponding to the packet, and the call will fail, resulting in a failure in processing requests of a large number of on-line normal users.
In the case that the link node only includes the gray version application, the routing rule set in this embodiment calls the gray version application for the request traffic dyed as the normal user tag and the test user tag, thereby avoiding the request failure. As shown in fig. 6, it is a traffic routing diagram in which the link node of this embodiment only includes the gray version application. For the case that the link node only includes the normal version application, which is equivalent to the case that the current node does not have the application to be tested, this embodiment invokes the normal version application to the request traffic dyed as the normal user tag, and ends the request to the traffic request dyed as the test user tag.
When the link node configuration error causes the request of the normal user to fail, the normal user cannot request, and even can spread to all nodes, and finally the whole distributed system fails. In the embodiment, only a small performance difference between the normal version application and the gray version application is considered, and degradation protection is performed on the condition that a matched node cannot be found, so that a large number of request failures are avoided, and the reliability is improved.
Example 3
The present embodiment provides a traffic routing apparatus based on a distributed system, and applies the traffic routing method based on the distributed system provided in embodiment 1 or embodiment 2. Fig. 7 is an architecture diagram of a traffic routing apparatus based on a distributed system according to this embodiment.
The traffic routing device based on the distributed system provided by the embodiment is applied to the distributed system, the distributed system comprises link nodes deployed in a distributed manner, and normal version application and/or gray version application are/is arranged in the link nodes; the traffic routing device comprises a node identification module and a rule configuration module, wherein the node identification module is used for identifying the link nodes with node labels with different dimensions and grouping the link nodes according to different dimensions; the rule configuration module is used for configuring an identification rule of request traffic and a routing rule of the request traffic to the link node.
In this embodiment, the traffic routing apparatus further includes a registration center, and the link node acquires the node label and registers registration information to the registration center, where the registration information includes a machine IP to which the link node belongs and the node label; and the link nodes carry out grouping with different dimensionalities according to the registration information stored in the registration center.
Further, the node identification module in this embodiment includes at least one of a code specification unit, a dynamic configuration unit, an environment variable configuration unit, and a container arrangement unit, where the code specification unit is configured to specify a link node to add a node tag in the application code by hard coding; the dynamic configuration unit is used for dynamically configuring the node labels of the link nodes; the environment variable configuration unit is used for configuring the node labels of the link nodes through environment variables; the container arranging unit is used for configuring the node labels of the link nodes through a container arranging tool.
In a specific implementation process, when a link node is started, a node identification module in the traffic routing device identifies node labels with different dimensions to the link node, the link node acquires the node labels and then registers registration information to a registration center, and meanwhile, the node identification module groups the link node according to the node labels of the link node and according to the different dimensions. Meanwhile, the rule configuration module configures an identification rule of the request traffic and a routing rule of the request traffic to the link node.
The code assigning unit in the node identification module assigns a link node to add a node tag to the link node in the application code through hard coding; the dynamic configuration unit dynamically configures the node labels of the link nodes; the environment variable configuration unit configures the node labels of the link nodes through environment variables; and the container arranging unit configures the node labels of the link nodes through a container arranging tool.
After all link nodes complete node label identification and rule configuration, the link nodes judge the currently received request flow: when the link node receives the request traffic which is not dyed, the link node carries out dyeing marking on the received request traffic according to the traffic identification rule, and then searches other link nodes with matched dimensionality according to traffic labels obtained by the request traffic through the dyeing marking and the routing rule of the request traffic; and when the link node receives the request traffic marked by the coloring, the link node calls the matched normal version application or gray version application according to the traffic label of the received request traffic.
The same or similar reference numerals correspond to the same or similar parts;
the terms describing positional relationships in the drawings are for illustrative purposes only and are not to be construed as limiting the patent;
it should be understood that the above-described embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the claims of the present invention.
Claims (10)
1. A traffic routing method based on a distributed system is characterized by comprising the following steps:
respectively adding at least one node label to all link nodes in distributed deployment, and grouping each link node according to different dimensions according to the node labels;
issuing an identification rule of a request flow and a routing rule of the request flow to the link node;
when a link node receives uncolored request traffic, the link node performs coloring marking on the received request traffic according to the traffic marking rule, and then searches other link nodes with matched dimensionality according to a traffic label obtained by the request traffic through the coloring marking and a routing rule of the request traffic;
and when the link node receives the request traffic marked by dyeing, the link node calls the matched application according to the received traffic label of the request traffic.
2. The distributed full-link gray scale publishing method according to claim 1, wherein when the link node performs color marking on the received request traffic according to the traffic identification rule, the link node judges according to a user value in an HTTP request Header, and when the user value is a normal user, the link node performs color marking on the request traffic as a normal user tag; and when the user value is the test user, the request traffic is dyed and marked as a test user label.
3. The distributed full link gray scale publishing method according to claim 2, wherein the applications in the link nodes comprise normal version applications and/or gray version applications.
4. The distributed full-link gray scale publishing method according to claim 3, wherein when the link node calls the matched application according to the received flow label of the requested flow, the method comprises the following steps:
when the link node comprises a normal version application and a gray version application, calling the normal version application for the request flow marked as a normal user according to the flow label of the request flow received by the link node, and calling the gray version application for the request flow marked as a test user;
when the link node only contains the normal version application, calling the normal version application for the request flow marked as the normal user by the dye, and finishing the request for the request flow marked as the test user by the dye;
and when the link node only contains the gray version application, calling the gray version application for the request traffic of the dye mark of the normal user and the test user.
5. The distributed full link gray scale publishing method of claim 1, wherein the dimensions of the node labels added to the link nodes include environment, machine configuration, and organizational framework departments.
6. The distributed full-link gray scale publishing method according to claim 1, wherein adding a node label to the link node comprises one or more of the following identification operations: hard-coded designations in application code, dynamic configuration using a configuration center, configuration through environmental variables, and configuration using a container orchestration tool.
7. The distributed full-link gray scale release method according to any one of claims 1 to 6, further comprising the steps of: after obtaining the node label, the link node registers registration information to a registration center, wherein the registration information comprises a machine IP (Internet protocol) to which the link node belongs and the node label; and the link nodes carry out grouping with different dimensionalities according to the registration information stored in the registration center.
8. The traffic routing device based on the distributed system is characterized by being applied to the distributed system, wherein the distributed system comprises link nodes deployed in a distributed manner, and normal version application and/or gray version application are/is arranged in the link nodes; the traffic routing device comprises a node identification module and a rule configuration module, wherein the node identification module is used for identifying the link nodes with node labels with different dimensions and grouping the link nodes according to different dimensions; the rule configuration module is used for configuring an identification rule of request traffic and a routing rule of the request traffic to the link node;
when the link node receives the request traffic which is not dyed, the link node carries out dyeing marking on the received request traffic according to the traffic identification rule, and then searches other link nodes with matched dimensionality according to traffic labels obtained by the request traffic through the dyeing marking and the routing rule of the request traffic; and when the link node receives the request traffic marked by the coloring, the link node calls the matched normal version application or gray version application according to the traffic label of the received request traffic.
9. The distributed full-link gray scale publishing system according to claim 8, wherein the traffic routing apparatus further comprises a registry, and the link nodes acquire the node labels and register registration information to the registry, wherein the registration information comprises machine IPs and node labels to which the link nodes belong; and the link nodes carry out grouping with different dimensionalities according to the registration information stored in the registration center.
10. The distributed full-link gray scale publishing system according to claim 8, wherein the node identification module comprises at least one of a code specification unit, a dynamic configuration unit, an environment variable configuration unit, and a container arrangement unit, wherein the code specification unit is configured to add a node tag to the application code by hard-coding a specified link node; the dynamic configuration unit is used for dynamically configuring the node labels of the link nodes; the environment variable configuration unit is used for configuring the node labels of the link nodes through environment variables; the container arranging unit is used for configuring the node labels of the link nodes through a container arranging tool.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210143050.XA CN114615192A (en) | 2022-02-16 | 2022-02-16 | Distributed full-link gray level publishing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210143050.XA CN114615192A (en) | 2022-02-16 | 2022-02-16 | Distributed full-link gray level publishing method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114615192A true CN114615192A (en) | 2022-06-10 |
Family
ID=81859836
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210143050.XA Pending CN114615192A (en) | 2022-02-16 | 2022-02-16 | Distributed full-link gray level publishing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114615192A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115237818A (en) * | 2022-09-26 | 2022-10-25 | 浩鲸云计算科技股份有限公司 | Method and system for realizing multi-environment multiplexing based on full link identification |
| CN115658348A (en) * | 2022-11-07 | 2023-01-31 | 北京瑞莱智慧科技有限公司 | Micro-service calling method, related device and storage medium |
| CN117499300A (en) * | 2023-12-27 | 2024-02-02 | 苏州元脑智能科技有限公司 | Routing link selection method, device, electronic equipment and medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6215771B1 (en) * | 1995-04-01 | 2001-04-10 | Nortel Networks Limited | Traffic routing in a telecommunications network |
| CN111786885A (en) * | 2020-06-23 | 2020-10-16 | 中国工商银行股份有限公司 | Distributed full-link gray level routing method and device |
| CN112995310A (en) * | 2021-02-10 | 2021-06-18 | 中国工商银行股份有限公司 | Multi-hierarchy calling gray level publishing method and device based on custom routing |
| CN113612686A (en) * | 2021-06-29 | 2021-11-05 | 中国人民财产保险股份有限公司 | Traffic scheduling method and device and electronic equipment |
-
2022
- 2022-02-16 CN CN202210143050.XA patent/CN114615192A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6215771B1 (en) * | 1995-04-01 | 2001-04-10 | Nortel Networks Limited | Traffic routing in a telecommunications network |
| CN111786885A (en) * | 2020-06-23 | 2020-10-16 | 中国工商银行股份有限公司 | Distributed full-link gray level routing method and device |
| CN112995310A (en) * | 2021-02-10 | 2021-06-18 | 中国工商银行股份有限公司 | Multi-hierarchy calling gray level publishing method and device based on custom routing |
| CN113612686A (en) * | 2021-06-29 | 2021-11-05 | 中国人民财产保险股份有限公司 | Traffic scheduling method and device and electronic equipment |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115237818A (en) * | 2022-09-26 | 2022-10-25 | 浩鲸云计算科技股份有限公司 | Method and system for realizing multi-environment multiplexing based on full link identification |
| CN115658348A (en) * | 2022-11-07 | 2023-01-31 | 北京瑞莱智慧科技有限公司 | Micro-service calling method, related device and storage medium |
| CN117499300A (en) * | 2023-12-27 | 2024-02-02 | 苏州元脑智能科技有限公司 | Routing link selection method, device, electronic equipment and medium |
| CN117499300B (en) * | 2023-12-27 | 2024-03-15 | 苏州元脑智能科技有限公司 | Routing link selection method, device, electronic equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114615192A (en) | Distributed full-link gray level publishing method and device | |
| US20180026887A1 (en) | Link selection for communication with a service function cluster | |
| US8195789B2 (en) | System, apparatus and method for characterizing messages to discover dependencies of services in service-oriented architectures | |
| CN108494574B (en) | Network function parallel processing infrastructure in NFV | |
| CN106130825A (en) | A kind of data transmission quality detection method and device | |
| CN108737168A (en) | A kind of micro services framework applications method for auto constructing based on container | |
| CN106685903A (en) | Data transmission method based on SDN, SDN controller and SDN system | |
| CN106127269A (en) | A kind of optical cable electronic tag control system | |
| JP2018201129A (en) | System management device, system management method, and program | |
| US11929879B2 (en) | NS instantiation method and NFVO | |
| CN108647043A (en) | A kind of realization method and system of order line input | |
| CN113703915A (en) | Access relation visualization method and device, electronic equipment and storage medium | |
| US20230344896A1 (en) | System and method for analyzing network objects in a cloud environment | |
| CN113849361B (en) | Method, device, equipment and storage medium for testing service node | |
| CN111404757A (en) | Cloud-based cross-network application integration system | |
| US11012337B1 (en) | Using machine learning to determine whether network components are being used as intended | |
| JP2006025434A (en) | System and method for high capacity fault correlation | |
| US10374941B2 (en) | Determining aggregation information | |
| CN114726785A (en) | Directional routing debugging mechanism based on request label expression | |
| CN107748773A (en) | Data search method, device and storage medium based on Enterprise Resources Plan | |
| CN106302167A (en) | route control method and device | |
| CN110266596B (en) | Message processing method, device, equipment and computer readable storage medium | |
| WO2022234870A1 (en) | Edge system resource monitoring for providing ai microservice, and configuration method therefor | |
| CN116708574A (en) | Method and system for multi-level dyeing of data packet | |
| US20240106713A1 (en) | Network inventory management 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 |