A kind of method that CMDB configuration items reconcile
Technical field
The present invention relates to IT O&Ms field, mainly a kind of method of CMDB configuration items mediation.
Background technology
Modern IT systems include a large amount of software and hardware resources, and after prolonged construction and safeguarding, IT operation maintenance personnels are general
It all can not clearly enumerate these resources and the contact between them so that in IT system change with escalation process, lacking and matching somebody with somebody
Put foundation.Therefore product technology as CMDB (configuration management database) is generated, to manage these software and hardware resources information.
But traditional CMDB product technologies, the accuracy of its data are typically responsible for by people, pass through a large amount of IT O&Ms people
The audit of member is ensured with modification.But this mode can cause substantial amounts of typing and audit work in large-scale IT environment
Amount so that CMDB construction period length, maintenance cost is high, hinders CMDB popularization.Pass through various monitoring scanning imaging systems simultaneously
It was found that resource information and manual entry resource information, when entering back into CMDB, also tend to exist conflicting with conflict, lead
Cause the accuracy of data low.
The content of the invention
It is an object of the invention to overcome the shortcomings of the prior art, and provide the side that a kind of CMDB configuration items reconcile
Method, identified by data and conclude integration technology, realized that the resource record submitted in all kinds of monitoring data sources merges, reduce rubbish number
According to.
The purpose of the present invention is completed by following technical solution.The method that this CMDB configuration items reconcile, utilize
The key feature attribute of all kinds of resources, to identify the Various types of data received, and isolate colliding data using sandbox technology, finally
Be put in storage after solving conflict by presetting rule or human intervention, step is as follows:
1), to each resource record received, its determinant attribute identifier is generated;
2) configuration item of this resource record, is determined according to determinant attribute identifier, and saves as an independent sandbox record,
Configuration item data is independently preserved by data source, avoids submitting record to obscure with other monitoring systems;
3), merge sandbox record automatically according to preset rules, realize that the sandbox record in different pieces of information source merges, generation is just
True unique configuration item record;
4) record of preset rules, is not provided with, user is correctly unique by human intervention, resolving data conflicts, generation
Configuration item records.
Beneficial effects of the present invention are:Solve the deficiency of traditional CMDB products, by preset rule, input number can be achieved
According to automatic identification with merging.This technology effectively reduces the operation of people, shorten CMDB construction period and reduce maintenance into
This, and improve the accuracy of CMDB data.
Brief description of the drawings
Fig. 1 is the configuration item mediation schematic flow sheet of the present invention;
Fig. 2 is configuration item identification process schematic diagram;
Fig. 3 is that configuration item merges schematic flow sheet.
Embodiment
Detailed introduction is done to the present invention below in conjunction with drawings and examples:
The present invention is related to 3 modules altogether:Configuration item identification module, configuration item sandbox preserving module, configuration item merging module.
1. configuration item identification module, configuration item identifier is generated according to rule;2. configuration item preserves sandbox module, configuration item is pressed into data
Source isolation is saved in each sandbox;3. configuration item merging module, all sandbox records of identifier identical are recorded with configuration item
Itself carries out comparing, and according to rule selection preferred value or by operation maintenance personnel preferred value is selected for conflict attribute, using as
Final configuration item record preserves.Fig. 1 is the operational flow diagram of above-mentioned each module;Data flow is described as indicated above;Each system
The configuration item data that system is submitted, configuration item identification is first passed around, forms unique data identifier;Secondly identifier and source are pressed
System, it is saved in independent sandbox record;Together with finally being recorded in itself with sandbox by configuration item CI, configuration item data is merged into.
The present invention can be solved effectively in IT O&Ms field, and CMDB data volumes are big and the problem of being difficult to management, and is effectively reduced
CMDB data maintenance workloads.
Embodiment:Comprise the following steps that:
1), presetting rule, the key feature attribute Keys of all kinds of IT resources of system intialization, and the merging of respective data sources
Regular MergeRules, and allow artificial adjustment as needed;
2), receive with identifying configuration item CI, calculate its identifier:
Identifier=getAttr (CI, Key1)+getAttr (CI, Key2)+...+getAttr (CI, Keyn)
3) configuration item, is pressed into data source, independently saves as sandbox record:SR;
4) configuration item, is merged
(1) Identifier identical SR, are inquired about, and configuration item CI corresponding to this Identifier is in itself;
(2) data comparison, is carried out to each attribute, to obtain entry/exit conflicts property set CollisionAitrs;
(3) it is, automatic to merge:Using corresponding MergeRules to CollistonAttrs processing, priority is selected
Highest conflict value is saved in CI as right value;
(4) it is, artificial to merge:CollistonAttrs is shown to user, selects right value to preserve in conflict value by user
To CI;
(5) CI after merging, is preserved.
Configuration item identification process in the present invention is described in detail in Fig. 2;When input is a server configuration item, first may be used
Inquire about Keys rules are " baseType, ipAddress, macAddress ";Next extracts this 3 property values
“Device,192.168.0.254,00:50:56:C0:00:01”;Last accumulation calculating can obtain Identifier " Device-
192.168.0.254-00:50:56:C0:00:01”.This server data that other data sources provide, its Keys property value must
It is so identical, therefore identical Identifier can be obtained, reach the purpose of identification configuration item.
Fig. 3 is described in detail configuration item in the present invention and merged;When input is two server configuration item sandbox records and one
It is individual audited configuration item data when, compare this 3 item data record first, there will be inconsistent data row individually extraction, shape
Into a conflict row set, such as " CPU frequency:2.6Ghz (CI-1), 2.6Ghz (SR-1.1), 2.4Ghz (SR-1.2);Disk is big
It is small:512GB (CI-1), 512GB (SR-1.1), 488.28GB (SR-1.2) ";Secondly search such configuration item conflict and merge rule
MergeRules, it may be determined that rule is " preferentially using the value consistent with CI versions ", and two conflict attributes can be obtained according to this rule
Preferred value be " CPU frequency:2.6Ghz(SR-1.1);Disk size:512GB(SR-1.1)”;Finally by all conflict attributes
Preferred value is merged into CI current versions and preserved, the final data as configuration item.
The present invention is not limited to the above-described embodiments, makees any change no matter being formed in its shape or material, every to adopt
With structure design provided by the present invention, all it is that a kind of of the present invention deforms, is considered as within the scope of the present invention.