Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Showbox 2 2012

1,326 views

Published on

Showbox is the new Exchange Data Center site which monitors every exchange server in the entire world. This site helps our engineers and on call incident managers stay informed about the health and status of Exchange, and get it back up and healthy as quickly and efficiently as possible.

Published in: Technology, Business
  • Be the first to comment

  • Be the first to like this

Showbox 2 2012

  1. 1. showbox toolkit 2/2012
  2. 2. showbox scenariosAlert > Assess > Act > EvaluateIncident ManagementAlert: Investigations are triggered from service thresholds, partner service teams, or from customer support. A page is sent to the On CallEngineer‟s phone and the engineer Acknowledges (acks) the alert so it doesn‟t roll over to other on call staff.Assess: The alert is read, and provides a place to start investigation. If the alert is not a problem the engineer can fix they will Lateral it tosomeone who can solve it. The scope of the issue is assessed. In the case of a data base outage, the backup copies are checkedAct: If the copies are good the service is restarted Engineer waits until the service indicates it is back online. Alerts are monitored forrelated problems and when possible the Engineer suppresses them to not wake up other Engineers unnecessarily. The Engineer goesback to bed and does further investigation the following day.Evaluate: Failover logs and debug scripts are launched for later root cause analysis. Bugs are edited and filed as appropriate.If the impact is significant an Incident Manager will be engaged.Alert: IM Requests happen when there is a significant customer impact. These requests engage the Incident Manager, and theCommunication Manager.Assess: The IM works with the On Call Engineer to assess the impact, informs the CM who publishes external posts to the public ifrequired. The IM will make the call when additional people need to be brought in, which may include partner teams, Ops, and otherengineers to diagnose and generate a recovery plan. Minutes count!Act: The plan is put into play. Once recovery is completed the service is monitoredEvaluate: Post-mortem will be done.
  3. 3. showbox scenariosAlert > Assess > Act > EvaluateCustomer Service RequestsAlert: A customer calls Frontline Engineer, who then must verify who the customer is over the phone, this is usually by finding the primarydomain or the Company name.Assess: While the customer is relating their problem CSS checks to see if there are any known issues that might be impacting the domain,these might be existing escalations, bugs, or service issues or known work around.Act: If the call has a high impact or if the Frontline CSS cannot solve the problem they escalate to the Escalation Engineer who can fix theproblem or escalate to Engineering.Evaluate: Always reviewed by the Customer Experience Team (CXP) monthly.Change Request ExternalAlert: Escalation request is received.Assess: The Escalation Engineer attempts to identify the correct recovery actionAct: A customer request such as moving a mailbox to another server is generated and goes through a triage process. The FrontlineEngineer is apprised of the status of the request and then contacts the company with the outcome.Evaluate: Always reviewed by the Customer Experience Team (CXP) monthly.Change Request InternalAlert: A change request ( a bug or upgrade) is communicated to the EngineerAssess: Engineering assesses the impact of the changeAct: The change is rolled out and monitored closely for a period of time. AnalysisEvaluate: why did we need the change right away, why wasn‟t it rolled out as a checked in build, why wasn‟t it automated.
  4. 4. technical ability variesScenarios focus on alert and overview in the portal the portal is a great place for overviews and a starting place for deep dive SMB ADMIN LORG ADMIN CSS EE SLT EXCHANGE ENGINEER LOW OFF THE CHART Technical ability and tolerance
  5. 5. reports for everyone Chart gardens are where we are more open ended M1 focus SMB ADMIN LORG ADMIN CSS EE SLT EXCHANGE ENGINEER LOW OFF THE CHARTPortal pages OVERVIEWS ANSWER THE QUESTION OF “HOW IS MY STUFF DOING?” CHART GARDENS FOR TROUBLE SHOOTING Configurable pop-outs are a useful scenario for: • IMs • On call engineers • Product unit engineers • …anyone doing deep comparative analysis of specific issues
  6. 6. anatomy of showboxScope controlNoun: a place or object withinthe topology. Defined by thescope control. Scope controlcan be also be changed by alink within a chart or piece ofdata.NavigationSupports keyscenarios, navigates to contentContentAdjective: Adescription, state, or otherinformation about the scopedselection. Some links in contentcan change scopeActionsVerbs: Actions taken are inrelation to the scope and are areaction to the description.Optimally actions like “EscalateNow”, “Lateral”, and“Communicate Now” includethe scope as an editable textfield in the form, and includestate information if possible tomake it easier on therecipient, and to help alert mailsclick through to the correctscope
  7. 7. primary navigationhealth escalation changes opticsSecondary Nav Overview Availability Customer Service PerformanceControl Data block list + Rotator Tap Rotator Tap Rotator Tap Rotator Tap PreviewContent Availability, Keynote, OWA, OLK, LiveID, OrgId, Auto CAS, Hub, MBX, more Spike-o-meter Mobile+RIM, Mailflow, Discover Service, Auto UM, EWS Discover Xml, OABContent changes Yes No No No Yes add domain specific just narrow scope to just narrow scope to just narrow scope to add domain specificfor narrow scope metrics, or for server selection selection selection metrics, or for server change to a role change to a role specific view specific view
  8. 8. primary navigationhealth escalation changes opticsSecondary Nav Alerts Support Calls People Directory ProtocolsControl List+Preview+alert block List+Preview+alert block table tableContent Alerts Support calls Directory Protocols and PDFsContent changes for When scoped to a server or domainscope
  9. 9. primary navigationhealth escalation changes opticsSecondary Nav Overview Inventory Deployment RequestsControl Data block list + Preview Rotator Rotator List view and network in detailsContent Pivot+Network Timeline of future List of requests with request changes+Network info and network chart in detailsContent changes for Yes add domain specificscope metrics, or for server change to a role specific view
  10. 10. primary navigationhealth escalation changes opticsSecondary Nav MSR Service Triage AdHoc More…Control Link farm Link farm Query submission PivotContent Executive level reports that Useful perf counters and Place to submit a query to Everything in ESP now summarize core service reports grouped by feature black box server and review statistics areas for Engineers results
  11. 11. 2 key layouts1. OverviewsA quick scan of the content shouldanswer the question, “Is there anythingwrong?”Overviews should summarize thecontents of a primary navigationalarea, and the general state of thescoped selection.If there are three additional secondarynav tabs, each tab should berepresented in the overview. Quickinvestigation on this page (preview)should let the user drill down quickly toa specific point of interest.It contains:• Data Blocks• and a preview pane.• Possibly a map too This proposed design is being built for o365 Wave 15
  12. 12. 2 key layouts2. List and previewStandard UMC control with theaddition of the alert block.Use for Alerts, Support calls, Change Visual treatmentrequests UnAcked incidentList is sortable, searchable, can be Open incidentfiltered, and can add and remove Resolved incident or alertitems.Preview is highly configurable andcan display custom layouts if needede.g. inject a chart or as E14 Discoverydid, insert a table
  13. 13. showbox scope control
  14. 14. 3Showbox ! Steven McQueen region: all forest: all dag: all site: all copygroup: all server: all ! Secondary navigation 1 secondary navigation 2 secondary navigation 3 secondaryPrimary navigation item 1 navigation 4Primary navigation item 2Primary navigation item 3Primary navigation item 4 Content which is filtered by scope control
  15. 15. Using dropdowns to select scope parent: all node2: all node3: all node4: all node5: all leaf: all parent1 parent: all node2: all node3: all node4: all node5: all leaf: all all parents parent1 parent2 parent3 parent4 parent1 node2-1 node2: all node3: all node4: all node5: all leaf: all all node2 node2-1 node2-2 node2-3 node2-4 parent1 node2-1 node3: all node4: all node5: all leaf: all
  16. 16. Using type down and dot to parent: all node2: all node3: all node4: all node5: all leaf: alladvance to the next field parent1 parent: all node2: all node3: all node4: all node5: all leaf: all all parents parent1 parent2 parent3 parent4 parent:node2- parent1 . all node2: all node3: all node4: all node5: all leaf: all all node2 node2-1 node2-2 node2-3 node2-4 parent:node2-1 . node3-1 parent1 . all node2: all node3: all node4: all node5: all leaf: all all node3 node3-1 node3-2 node3-3 node3-4 parent1 node2-1 node3-1 node4: all node5: all leaf: all
  17. 17. Zooming in and out of dataContent area in UI shows the State 1appropriate content for the parent1 node2-1 node3-1 node4-2 node5: all leaf: allselection.Original state is not changed untiluser explicitly changes it. selection parent1 node2-1 node3-1 node4-2 node5: all leaf: all State 2 parent1 node2-1 node3-1 node4-2 node5: all leaf: all selection parent1 node2-1 node3-1 node4-2 node5: all leaf: all State 3 parent1 node2-1 node3-1 node4-2 node5: all leaf: all Explicit Change parent1 node2-1 node3-1 node4-2 node5: all leaf: all …change continued parent1 node2-1 node2-1 node3-1 node4-2 node5: all leaf: all all node2 node2-1 node2-2 node2-3 node2-4 State 1 parent: all node2-2 node3: all node4: all node5: all leaf: all
  18. 18. Rendering data clusters usingparentheses and simple Boolean parent1 node2-1 (node3-1 - node3-5) node4: all node5: all leaf: allqueries. parent1 node2-1 (node3-1 , node3-5) node4: all node5: all leaf: all
  19. 19. Searching for an object State 1 parent: all node2: all node3: all node4: all node5: all leaf: all query parent: all node2: all node3: all node4: all node5: all leaf: all Leaf12- parent: all node2: all node3: all node4: all node5: all leaf: all Leaf12-1 Leaf12-10 Leaf12-11 Leaf12-12 resolution parent2 node2-2 node3-6 node4-4 node5-1 leaf12-10Searching for a clustering concept State 1 parent: all node2: all node3: all node4: all node5: all leaf: all query parent: all node2: all node3: all node4: all node5: all leaf: all Concept parent: all node2: all node3: all node4: all node5: all leaf: all resolution parent1 node2-1 (node3-1 , node3-5) Concept
  20. 20. showbox data blocks
  21. 21. data blockStackable UX Lego blocks that for v1 will be organized statically within layouts, however since thedata they get is subject to the scope the data will change appropriately.How it worksEach instance of the data block is encoded with a usage Scoped to Lots of labels for one datasuperset of data fields. When there is no data for all regionsa label the label will not be shown, and when Tenants +2% 486,012 block: Mailboxes +5% 82,763,121there is no data for the block the entire block is Active -7% 8,453,454 Tenantshidden. Sent mail +10% 50 million DomainsThe same commandlet is called from a given Mailboxeslayout all the time, but since it is scoped by the Active usersUX different combinations of data can be usage Scoped to Sent mail a tenantreturned. Mailboxes +2% 1000 Etc Active +5% 33 EtcThis control supports flagging, links, trends, and Sent mail -7% 321 Etcsimple tabular data layouts. Etc EtcLayout Etc…Data blocks should be fixed width, have a Scope dictates the query, only fields whichmaximum of four columns and those columns have data are shownshould to align with all the columns for criticaldata so they can be easily scanned APC Current 99.90% 9:00 AM Low 96.76% 8:45 AM Average 97.70% 1 hour availability Current 99.90% Low 96.76%
  22. 22. showbox link farms
  23. 23. Link FarmsCanned reportsOrganized by scenarios. This UI isalready supported by UMC. See auditing
  24. 24. showbox stacker plot
  25. 25. stacker plot heatmapHow it works• Every site is represented, and each site is represented only once Black selection• Each chart has four selectable regions, green, yellow, red, which load the corresponding list view. Red selectionPros• Outliers are bigger, and more in focus• Chart scales to very large data sets Yellow selection• With list view, very meaningful information is available• Groups of items with the same capacity are selectable Green selection• Scope control is now the way to change scope instead of drilling down a lot from the chart, making back more difficultCons• Harder to compare regions or other large groups. Eg. APC vs. NAM, Namprod01 vs. Namprod02
  26. 26. showbox page examples
  27. 27. 3Showbox ! Steven McQueen ! parent: all node2: all node3: all node4: all node5: all leaf: allhealth overview availability customer service performanceescalations alerts availability and alert volume Updated: 2/9/2012 9:00AMchanges 4 active alerts APC 99.5 LAMoptics availability 15min 1 hr NAM EUR Active monitoring 99.7% 99.5% ESC Keynote 99.5% 94.8% Alerts 2 customer latency failures 95 Outlook +2% 5 7 Mobile +5% 14 8 AM 8:05 8:10 8:15 8:20 8:25 8:30 8:35 8:40 8:45 8:50 8:55 9 AM Mailflow -7% 35 Provisioning +10% 3 TIME: 1h 2h 6h Custom service latency failures Network +2% 12 availability and alerts / 8:50am – 9:00am Live ID +5% 3 REGION TYPE TIME AVAILABILITY [SERVICE INCIDENT] Monitoring -7% 4 NAM [SERVICE INCIDENT] keynote failures 12/9 9:12 95% ACP keynote failures AD +10% 10 for connections via NAM [RESOLVED INCIDENT] Quis nostrud 12/9 9:12 99% Singapore SingTel FOPE -1% 2 OWNER Datacenter - Ack Now! Engage IM! SCOPE NAM/NAMPROD07/CH1PRO D702/CH1PRD0702CA017 IMPACT Outlook Connectivity
  28. 28. 3Showbox ! Steven McQueen ! parent: all node2: all node3: all node4: all node5: all leaf: allhealth alerts support calls people directory protocolsescalationschanges ALERT TIME OWNER [SERVICE INCIDENT] ACPoptics [SERVICE INCIDENT] keynote failures for connections via Singapore SingTel 09/27 9:12 Pending – Datacen… keynote failures for connections [SERVICE INCIDENT] This database has had only one good copy for 20 minutes 09/27 9:12 Jessed-High Availabi… via Singapore SingTel [SERVICE INCIDENT] one healthy copy for TestADReplication: One or more 09/27 9:12 Jessed-High Availabi… ! [SERVICE INCIDENT] Lorem ipsum dolor sit amet, consectetur adipisicing elit, 09/27 9:12 Jessed-High Availabi… [INVESTIGATION] Lorem ipsum dolor sit amet, consectetur adipisicing elit, 09/27 9:12 Jessed-High Availabi… owner: Datacenter - Ack Now! [INVESTIGATION] Ut enim ad minim veniam, quis nostrud exercitation ulla 09/27 9:12 Jessed-High Availabi… Engage IM! [INVESTIGATION] Quis nostrud exercitation ullamco laboris nisi ut aliquip ex 09/27 9:12 Jessed-High Availabi… [RESOLVED INCIDENT] Quis nostrud exercitation ullamco laboris nisi ut aliqui 09/27 9:12 Jessed-High Availabi… scope: CH1PRD0702CA017 [RESOLVED INVESTIGATION] Quis nostrud exercitation ullamco 09/27 9:12 Jessed-High Availabi… More… [RESOLVED INVESTIGATION] Quis nostrud exercitation ullamco laboris nisi 09/27 9:12 Jessed-High Availabi… [RESOLVED INCIDENT] Quis nostrud exercitation ullamco laboris nisi ut aliqui 09/27 9:12 Jessed-High Availabi… impact Tenants: 363 [RESOLVED INVESTIGATION] Quis nostrud exercitation ullamco 09/27 9:12 Jessed-High Availabi… Users: 8834 [RESOLVED INVESTIGATION] Quis nostrud exercitation ullamco laboris nisi 09/27 9:12 Jessed-High Availabi… More… [RESOLVED INCIDENT] Quis nostrud exercitation ullamco laboris nisi ut aliqui 09/27 9:12 Jessed-High Availabi… RESOLVED INVESTIGATION] Quis nostrud exercitation ullamco 09/27 9:12 Jessed-High Availabi… [RESOLVED INVESTIGATION] Quis nostrud exercitation ullamco laboris nisi 09/27 9:12 Jessed-High Availabi… [RESOLVED INCIDENT] Quis nostrud exercitation ullamco laboris nisi ut aliqui 09/27 9:12 Jessed-High Availabi… [RESOLVED INVESTIGATION] Quis nostrud exercitation ullamco 09/27 9:12 Jessed-High Availabi… [RESOLVED INVESTIGATION] Quis nostrud exercitation ullamco laboris nisi 09/27 9:12 Jessed-High Availabi…
  29. 29. 3Showbox ! Steven McQueen ! parent: all node2: all node3: all node4: all node5: all leaf: allhealth monthly service review service triage ad hoc moreescalations monthly service reviewchanges Key Usage Stats Server and Hardwareoptics Total Mailbox Count and Active Mailbox Count Lorem Ipsum Dolor Sit and Consectetur Adipisicing Elit Provisioning Support Provisioning Latency and Failures, Tenant Growth by Offering, Lorem Ipsum Dolor Sit and Consectetur Adipisicing Elit and Tenant Growth by Segment Upgrades Availability & Incidents Lorem Ipsum Dolor Sit and Consectetur Adipisicing Elit Keynote Availability, SCOM Availability, and Availability Incidents Migration Lorem Ipsum Dolor Sit and Consectetur Adipisicing Elit Escalation Analysis Top Escalations by Type, Top Root Causes Site Resiliency Lorem Ipsum Dolor Sit and Consectetur Adipisicing Elit Networking, Directory & Capacity Heatmap Migrations, Connections, Load Balancer, AD Health , and Capacity Heatmap, Build Release & Operations Scorecard Build Release Scorecard , Operations Scorecard , and Data Protection
  30. 30. 3Showbox ! Steven McQueen ! parent: all node2: all node3: all node4: all node5: all leaf: allhealth overview availability customer service performanceescalations Updated: 06/25/2011 9:00AMchanges CAS CPU HUB CPU HUB IO MBX CPU MBX SPACE MBX IO AD CPU AD IO F5 CPU F5 MEM UMoptics CAS CPU failures SITE Failover Resource Value MBXs ACTVE MBXs DBs MACHINES state NAM06/SN2PRD0602 Unstable 79% 0 0 0 41/48 NAM01/SN2PRD0602 details State: Provisioned Version: R5 Build: 14.01.0225.071 More… impact Client Session Concurrency: 113,076 Deliveries/Sec: 267 related CAS CPU SN2PRD0602 SN2PRD0102 CH1PRD0106
  31. 31. 3Showbox ! Steven McQueen ! parent: all node2: all node3: all node4: all node5: all leaf: allhealth overview availability customer service performanceescalations Updated: 06/25/2011 9:00AMchanges CAS CPU HUB CPU HUB IO MBX CPU MBX SPACE MBX IO AD CPU AD IO F5 CPU F5 MEM UMoptics CAS CPU at 60% capacity SITE Failover state Resource Value MBXs ACTVE MBXs DBs MACHINES NAM06/SN2PRD0602 Unstable 75% 0 0 0 41/48 NAM01/SN2PRD0602 NAM06/CH1PRD0602 Unstable 75% 0 0 0 41/48 details NAM04/SN2PRD0402 Critical 65% 0 0 0 41/48 State: Provisioned Version: R5 NAM04/CH1PRD0402 Warning 55% 0 0 0 41/48 Build: 14.01.0225.071 NAM06/SN2PRD0604 Warning 55% 0 0 0 41/48 More… NAM02/SN2PRD0202 Warning 55% 0 0 0 41/48 impact APC01/HKNPRD0102 Warning 55% 0 0 0 41/48 Client Session Concurrency: 113,076 Deliveries/Sec: 267 EUR01/AMSPRD0302 Warning 45% 0 0 0 41/48 NAM01/SN2PRD0102 Warning 45% 0 0 0 41/48 related CAS CPU SN2PRD0602 SN2PRD0102 CH1PRD0106
  32. 32. showbox chart garden
  33. 33. chart gardensHow it works• Pop out to stock chart configurations, from links in the page or from the chart drop down.• URL is visible and equals a parameterized link to the visible configuration of charts. This is an aid to IMs and Engineers who want to get back to this view (add it to favorites, copy it to an email) to get others quickly up to speed on the thing they are focused on.• Charts can be modified,• Allow users to add charts from the entire suite of reports in Showbox
  34. 34. 3Showbox ! Steven McQueen ! parent: all node2: all node3: all node4: all node5: all leaf: allhealth overview availability customer service performanceescalations Updated: 06/25/2011 9:00AMchanges CAS CPU HUB CPU HUB IO MBX CPU MBX SPACE MBX IO AD CPU AD IO F5 CPU F5 MEM UMoptics CAS CPU at 60% capacity SITE Failover state Resource Value MBXs ACTVE MBXs DBs MACHINES NAM06/SN2PRD0602 Unstable 75% 0 0 0 41/48 NAM01/SN2PRD0602 NAM06/CH1PRD0602 Unstable 75% 0 0 0 41/48 details NAM04/SN2PRD0402 Critical 65% 0 0 0 41/48 State: Provisioned Version: R5 NAM04/CH1PRD0402 Warning 55% 0 0 0 41/48 Build: 14.01.0225.071 NAM06/SN2PRD0604 Warning 55% 0 0 0 41/48 More… NAM02/SN2PRD0202 Warning 55% 0 0 0 41/48 impact APC01/HKNPRD0102 Warning 55% 0 0 0 41/48 Client Session Concurrency: 113,076 Deliveries/Sec: 267 EUR01/AMSPRD0302 Warning 45% 0 0 0 41/48 NAM01/SN2PRD0102 Warning 45% 0 0 0 41/48 related CAS CPU SN2PRD0602 SN2PRD0102 CH1PRD0106
  35. 35. https://pod51005.outlook.com/showbox/CAS/ChartGardenx.aspx?pwmcid=1&ReturnObjectType=1Exchange SharePoint Lync 3 region: all forest: all dag: allallsite: all allcopygroup: all all server: all region: all forest: all dag: site: copygroup: server: all i ! ! 1h 3h 8h CustomhealthCAS CPU TOP 5 overview availability customer service performanceescalations Updated: 06/25/2011 9:00AM SITE PERF MACHINES Updated: 06/25/2011 9:00AM sec NAM01/SN2PRD0602 75% 41/48 NAM01/SN2PRD0602changes CAS CPU HUB CPU HUB IO MBX CPU 300 MBX SPACE MBX IO NAM01/SN2PRD0602 AD CPU 75% AD IO 41/48 F5 CPU F5 MEM UM ARR failover stateoptics 250 NAM01/SN2PRD0602 65% 41/48 unstable NAM01/SN2PRD0602 65% 41/48 200 details NAM01/SN2PRD0602 65% 41/48 State: Provisioned 150 Version: R5 Build: 14.01.0225.071 100 More… 50 impact 0 8:00 am 8:15 8:30 CAS CPU9:00 60% capacity 8:45 at 9:15 9:30 9:45 10:00 Client Session Concurrency: 113,076 Deliveries/Sec: 267 SITE PERF MBXs ACTVE MBXs DBs MACHINES NAM01/SN2PRD0602 75% 0 0 0 41/48 CAS MEMORY TOP 5 NAM01/SN2PRD0602 NAM01/SN2PRD0602 75% 0 0 0 41/48 SITE details NAM01/SN2PRD0602 65% Updated: 06/25/2011 9:00AM 0 0 0 41/48 PERF MACHINES State: Provisioned NAM01/SN2PRD0602 65% 0 0 sec NAM01/SN2PRD0602 0 41/48 75% 41/48 Version: R5 NAM01/SN2PRD0602 Build: 14.01.0225.071 NAM01/SN2PRD0602 65% 0 0 300 NAM01/SN2PRD0602 0 41/48 75% 41/48 failover state More… NAM01/SN2PRD0602 65% 0 0 250 NAM01/SN2PRD0602 0 41/48 65% 41/48 unstable NAM01/SN2PRD0602 NAM01/SN2PRD0602 impact 65% 0 0 200 0 41/48 65% 41/48 details Client Session Concurrency: 113,076 NAM01/SN2PRD0602 65% 0 0 NAM01/SN2PRD0602 0 41/48 65% 41/48 State: Provisioned Deliveries/Sec: 267 150 Version: R5 NAM01/SN2PRD0602 65% 0 0 0 41/48 100 related Build: 14.01.0225.071 CAS CPU More… 50 SN2PRD0602 impact SN2PRD0102 0 Client Session Concurrency: 113,076 8:00 am 8:15 8:30 8:45 9:00 9:15 9:30 9:45 10:00 CH1PRD0106 Deliveries/Sec: add charts 267 share close CHPRD0102
  36. 36. chart gardens plusHow it works• More cowbell• Don‟t start here, only use this if your scenarios typically require it.
  37. 37. https://pod51005.outlook.com/showbox/CAS/ChartGardenx.aspx?pwmcid=1&ReturnObjectType=1 region: all forest: all dag: all site: all copygroup: all server: allCHARTS 1h 3h 8h CustomDATA CAS CPU TOP 5HISTORYMORE… Updated: 06/25/2011 9:00AM SITE PERF sec NAM01/SN2PRD0602 75% NAM01/SN2PRD0602 300 NAM01/SN2PRD0602 75% failover state 250 NAM01/SN2PRD0602 65% unstable NAM01/SN2PRD0602 65% 200 details NAM01/SN2PRD0602 65% State: Provisioned 150 Version: R5 Build: 14.01.0225.071 100 More… 50 impact 0 Client Session Concurrency: 113,076 8:00 am 8:15 8:30 8:45 9:00 9:15 9:30 9:45 10:00 Deliveries/Sec: 267 CAS CPU TOP 5 Updated: 06/25/2011 9:00AM SITE PERF sec NAM01/SN2PRD0602 75% NAM01/SN2PRD0602 300 NAM01/SN2PRD0602 75% failover state 250 NAM01/SN2PRD0602 65% unstable NAM01/SN2PRD0602 65% 200 details NAM01/SN2PRD0602 65% State: Provisioned 150 Version: R5 Build: 14.01.0225.071 100 More… 50 impact 0 Client Session Concurrency: 113,076 8:00 am 8:15 8:30 8:45 9:00 9:15 9:30 9:45 10:00 Deliveries/Sec: 267 add charts share close
  38. 38. chart library Xxx – Windows Internet Explorer x CHART LIBRARY HELP ACTIVE DIRECTORYHow it works AVAILABILITY Select individual charts or groups to add to your page.• All charts that can be natively shown in showbox are CAS CAS CPU available (excludes non-UMC optics) EAS A group of charts to find Lorem ipsum dolor sit amet, HUB consectetur adipisicing elit, sed do eiusmod tempor• Major chart grouping/taxonomy needs to be MAILBOX CAS CPU rationalized across all teams in showbox MRS Connections• Charts can be added as groups or individually Request rate per protocol Memory Related counters OWA Logons Keynote Blah Blah blah save cancel
  39. 39. showbox rotator tap
  40. 40. Availability rotator model – how it worksThe rotator tap works by allowing correlation of large amounts of dataeasily. Here are the key elements that make up the rotator:Rotation – graphs will rotate as explained in the interaction model.Hero chart - compares three/four data sets: overall availability another data set(s), as 2selectable regions which correspond with the list view and mini chartsTime – the rotator provided an „in the now‟ view of dataStatic – graphs will be static unless interacted with by the user 1Pop out – Dependency on EDS to build the chart gardens – plan for Beta 3/12. Willallow adding more graphs for comparison in a new window (Early March timeframe)Scale graphs – each graph will need to be in the hero spot as well as mini graph 3 4Error handling – need to follow up with Sean and SrdjanTime selection – width of selection area (for time) will remain constant – no matterwhat the time scale is scoped to (see example). Selection will change on mini charts to 5match hero chart selection.Legend links – the legend links that are locations/regions etc. will be links to changescopeList view – title and data correspond with selection on hero chart Hero chart List view Mini charts/Data blocksMini charts - Quickly scan able for patterns or outliers, can quickly move into the herospot for deeper analysis (on click)
  41. 41. Availability interaction modelInteractions:Hero chart• Clickable time increments (shaded area)• Wish list is to have the shaded selected area expand and collapse on mouse drag (ex: think stock charting timelines)(future)• User can change the time scope of the chart – mini charts also update to this time• Click on the pop out icon to open a new window that contains the charts and allows for addition of others 3 2• Legend locations are links – allowing for quick scope changeList view• clicking selects and highlights the item, providing details for that incident, links/actions 4• User can ack an incident• List can be filteredMini charts/data blocks• Click on the desired chart and it will move (rotate) into the 5 hero spot (carousel counter-clockwise) 1other• Custom time range can go from 15 min to???• Max of “X” mini charts in the rotator, if the user wants to see more charts they must open up in a chart garden• Animation will fade out charts and data when the actual „rotation‟ of the charts occurs to minimize noise and confusion.Order of rotation commands1. Clear2. Rotate3. Render chart4. Render list view
  42. 42. Time selection on hero chartThe time selection width will remain the same regardless of what timeframe you choose. The only change will be the specific time span you will see in the list view. The selection on the hero chart will be reflected in the mini charts. 1 hour 10 min timeframe 8 hour 80 min timeframe
  43. 43. 3Showbox ! Steven McQueen ! parent: all node2: all node3: all node4: all node5: all leaf: allhealth overview availability customer service performanceescalations alerts CTPschanges 4 active alerts availability latency failures 100% -2% 112optics availability 15min 1 hr Active monitoring 99.7% 99.5% STPs Keynote 99.5% 94.8% 99.2% 6% 36 customer latency failures network Outlook +2% 5 65% 17% 154 Mobile +5% 14 Mailflow -7% 35 Provisioning +10% 3 mailflow service latency 78% -5% 1,018 failures Network +2% 12 Live ID +5% 3 Monitoring -7% 4 AD +10% 10 FOPE -1% 2
  44. 44. 3Showbox ! Steven McQueen ! region: all forest: all site: all dag: all server: allhealth overview availability customer service performanceescalations availability and alert volume Updated: 2/9/2012 9:00AM CTPschanges APC 99.5 4.2 LAMoptics NAM EUR ESC 8AM 8:15 830 845 9AM Alerts 2 STPs 95 57 7 8 AM 8:05 8:10 8:15 8:20 8:25 8:30 8:35 8:40 8:45 8:50 8:55 9 AM TIME: 1h 2h 6h 24hr 1wk 8AM 8:15 830 845 9AM mailflow availability and alerts 8:50am – 9:00am 174 REGION TYPE TIME AVAILABILITY [SERVICE INCIDENT] ACP NAM [SERVICE INCIDENT] keynote failu 12/9 9:12 95% keynote failures for NAM [RESOLVED INCIDENT] Quis nostrud 12/9 9:12 99% connections via 8AM 8:15 830 845 9AM Singapore SingTel networking People: Owner 5,236 Datacenter - Ack Now! IM: Engage IM! 8AM 8:15 830 845 9AM MORE Scope:
  45. 45. 3Showbox ! Steven McQueen ! region: all forest: all site: all dag: all server: allhealth overview availability customer service performanceescalations availability and alert volume Updated: 2/9/2012 9:00AM CTPschanges APC 99.5 4.2 LAMoptics NAM EUR ESC 8AM 8:15 830 845 9AM Alerts 2 STPs 95 57 7 8 AM 8:05 8:10 8:15 8:20 8:25 8:30 8:35 8:40 8:45 8:50 8:55 9 AM TIME: 1h 2h 6h 24hr 1wk 8AM 8:15 830 845 9AM mailflow availability and alerts 8:30am – 8:40am 174 REGION TYPE TIME AVAILABILITY [SERVICE INCIDENT] outlook NAM [SERVICE INCIDENT] outlook failu 12/9 9:12 99% failures for connections via 8AM 8:15 830 845 9AM Singapore SingTel networking People: Owner 5,236 Datacenter - Ack Now! IM: Engage IM! 8AM 8:15 830 845 9AM MORE Scope:
  46. 46. 3Showbox ! Steven McQueen ! region: all forest: all site: all dag: all server: allhealth overview availability customer service performanceescalations availability and alert volume Updated: 2/9/2012 9:00AM CTPschanges APC 99.5 4.2 LAMoptics NAM EUR ESC 8AM 8:15 830 845 9AM Alerts 2 STPs 95 57 7 8 AM 8:05 8:10 8:15 8:20 8:25 8:30 8:35 8:40 8:45 8:50 8:55 9 AM TIME: 1h 2h 6h 24hr 1wk 8AM 8:15 830 845 9AM mailflow availability and alerts 8:30am – 8:40am 174 REGION TYPE TIME AVAILABILITY [SERVICE INCIDENT] outlook NAM [SERVICE INCIDENT] outlook failu 12/9 9:12 99% failures for connections via 8AM 8:15 830 845 9AM Singapore SingTel networking People: Owner 5,236 Datacenter - Ack Now! IM: Engage IM! 8AM 8:15 830 845 9AM MORE Scope:

×