<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Sayed Saeed on Medium]]></title>
        <description><![CDATA[Stories by Sayed Saeed on Medium]]></description>
        <link>https://medium.com/@e.sayedsaeed5?source=rss-d584a4de0b76------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*BML6pXUwZ9uar0TdkzGjZg.jpeg</url>
            <title>Stories by Sayed Saeed on Medium</title>
            <link>https://medium.com/@e.sayedsaeed5?source=rss-d584a4de0b76------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 07:37:23 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@e.sayedsaeed5/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[The Hidden Architecture of IBM Maximo: Mastering GL Accounts and ERP Integration]]></title>
            <link>https://medium.com/@e.sayedsaeed5/the-hidden-architecture-of-ibm-maximo-mastering-gl-accounts-and-erp-integration-85f8386a0151?source=rss-d584a4de0b76------2</link>
            <guid isPermaLink="false">https://medium.com/p/85f8386a0151</guid>
            <category><![CDATA[maximo]]></category>
            <dc:creator><![CDATA[Sayed Saeed]]></dc:creator>
            <pubDate>Wed, 28 Oct 2026 07:49:33 GMT</pubDate>
            <atom:updated>2026-10-28T07:49:33.769Z</atom:updated>
            <content:encoded><![CDATA[<p>When enterprise asset management (EAM) deployments stumble, it is rarely because a technician couldn’t close a work order. It happens because the operational system fails to speak the exact financial language of the corporate ERP. In IBM Maximo, the bridge between maintenance operations on the shop floor and the corporate ledger in SAP or Oracle is the General Ledger (GL) architecture.</p><p>If this architecture is misconfigured, it results in unbalanced ledgers, failed integration messages, and manual reconciliation nightmares for the finance team. Here is a deep dive into how Maximo structures financial data, the hidden dangers of configuration, and how to guarantee seamless ERP integration.</p><p><strong>The Anatomy of a Relational GL Account</strong> In many legacy systems, an account code is just a monolithic string of text. Maximo takes a structured, relational approach. A GL account code is an aggregation of individual segments — called components — stitched together by delimiters.</p><p>An account like 6100-350-SAF contains three distinct data elements:</p><ol><li><strong>Cost Center:</strong> 6100 (4 characters)</li><li><strong>Activity:</strong> 350 (3 characters)</li><li><strong>Resource:</strong> SAF (3 characters)</li></ol><p>This requires a strict two-tier design. First, administrators use the Database Configuration application to define the system-level blueprint: the number of components, their exact character lengths, and the delimiters. Second, individual Organizations (financial entities) use the Chart of Accounts application to populate their specific valid combinations based on that blueprint.</p><p><strong>Dynamic Generation and the Problem with Placeholders</strong> Maximo allows for highly dynamic account generation. Users rarely memorize a full 20-segment GL string; instead, Maximo builds the string dynamically based on the associated Work Order, Asset, and Location.</p><p>This introduces the concept of <strong>Placeholders</strong>. If a component is marked as <em>Required</em> in the database configuration, Maximo must have a value for it to process a transaction. If it lacks a value during the dynamic build, it inserts placeholders (typically question marks). For example, if Maximo knows the Cost Center and Resource but not the Activity, it generates a partially defined account: 6100-???-SAF.</p><p>A financial transaction — like issuing a part from a storeroom — cannot be completed if the resulting GL account contains placeholders. The missing component must be provided to make it fully defined.</p><p><strong>The Danger Zone: Modifying Component Lengths</strong> One of the most destructive mistakes an administrator can make is altering the character length of a GL component after data has been established. Maximo stores both the individual components and the concatenated string in the database.</p><p>If you configure Component 1 to be 4 characters and Component 2 to be 3 characters, the system expects a format like [4 chars]-[3 chars] (e.g., 6100-350). If you later modify the database configuration so Component 1 is 3 characters and Component 2 is 4 characters, you instantly corrupt the existing Chart of Accounts. The system will look at the existing 6100 string, see a 4-character value where a 3-character value is now mandated, and invalidate the component.</p><p><strong>ERP Integration: The Ultimate Source of Truth</strong> Maximo is an operational system that feeds an accounting system. It should never be the system of record for GL accounts. The corporate ERP (SAP, Oracle, Microsoft Dynamics) dictates the accounting structure.</p><p>For a flawless integration via the Maximo Integration Framework (MIF):</p><ul><li><strong>Mirror the Blueprint:</strong> The Maximo System-Level format (number of components, lengths, delimiters) must be configured to perfectly mimic the ERP’s GL string structure.</li><li><strong>Inbound Syncing:</strong> The actual valid accounts should be synced automatically from the ERP to Maximo. Users must be locked out of manually creating GL accounts in Maximo.</li><li><strong>Outbound Precision:</strong> When a Maximo user completes a Work Order, Maximo generates the fully defined GL string and sends it via an outbound financial transaction message (GLTRANS or INVISSUE). If Maximo allows a transaction to process with a partially defined account (e.g., containing ???), the ERP middleware will instantly reject the message because ??? is not a valid cost center.</li></ul><p>Successful Maximo architecture requires acknowledging that while the system must provide flexibility for the maintenance teams, it must enforce absolute, uncompromising rigidity for the finance teams. By protecting the Data Dictionary and treating the ERP as the ultimate financial authority, you ensure your asset management data translates perfectly into corporate financial reality.</p><p>#IBMMaximo #EnterpriseArchitecture #EAM #ERPIntegration #DataGovernance #SAP #Oracle #AssetManagement #SystemsEngineering</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=85f8386a0151" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[IBM Maximo Multisite Architecture]]></title>
            <link>https://medium.com/@e.sayedsaeed5/ibm-maximo-multisite-architecture-1fed63d0ff3a?source=rss-d584a4de0b76------2</link>
            <guid isPermaLink="false">https://medium.com/p/1fed63d0ff3a</guid>
            <category><![CDATA[aem]]></category>
            <category><![CDATA[ma]]></category>
            <dc:creator><![CDATA[Sayed Saeed]]></dc:creator>
            <pubDate>Wed, 21 Oct 2026 10:38:59 GMT</pubDate>
            <atom:updated>2026-10-21T10:38:59.502Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Maximo Manage’s multisite architecture</strong> allows an enterprise to run a single software instance and database while logically separating financial entities, operational facilities, and shared master data. Instead of deploying separate databases for every plant or subsidiary — which causes data duplication and administrative overhead — the system uses a four-tier hierarchy to control what data is shared globally and what remains isolated locally.</p><h3>1. The Architectural Hierarchy</h3><p>The foundation of a multisite setup is understanding where data lives. Maximo organizes data into four hierarchical levels, moving from broad enterprise-wide parameters down to day-to-day plant operations.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lFEZLWUT57TaOAiHyMJopw.png" /></figure><blockquote><strong><em>Key Architectural Rule:</em></strong><em> An Organization can only belong to </em><strong><em>one</em></strong><em> Item Set and </em><strong><em>one</em></strong><em> Company Set, though multiple Organizations can share the exact same sets. Every Organization must contain at least one Site to perform operational work.</em></blockquote><h3>2. Virtual vs. Physical Reality</h3><p>A common misconception is that an <strong>Organization</strong> must be a corporate headquarters and a <strong>Site</strong> must be a physical building with a street address. In Maximo, both are <strong>virtual logical entities</strong> designed to match your accounting and operational workflows:</p><ul><li><strong>Multiple Sites inside One Building:</strong> A single manufacturing plant might have two separate subdivisions (e.g., R&amp;D and Commercial Production) under the same roof. If they refuse to share work orders or storeroom inventory, you configure them as <strong>two separate Sites</strong>.</li><li><strong>Multiple Buildings inside One Site:</strong> If a company has three separate warehouses in Ohio that share the exact same maintenance crews, work order pool, and inventory balances, you can group all three physical locations into <strong>one single Site</strong>.</li><li><strong>Departmental Organizations:</strong> A company can create two separate Organizations within the same country — one for <strong>IT Asset Management</strong> and another for <strong>Facilities Maintenance</strong> — simply because they use completely different General Ledger account structures and cost centers.</li></ul><h3>3. Configuration &amp; Business Rules: Organization vs. Site</h3><p>The <strong>Organizations application</strong> is the primary control center where administrators define how rules cascade down the hierarchy. Notably, certain functional areas — like Work Orders and Inventory — have settings at <em>both</em> the Organization and Site levels, serving different purposes:</p><h3>Work Orders</h3><ul><li><strong>Organization-Level Rules:</strong> Govern high-level financial and compliance policies, such as editing rules (which statuses allow modifications) and mandatory prompts for failure and asset downtime reporting.</li><li><strong>Site-Level Rules:</strong> Govern local execution, such as task numbering conventions and site-specific work order autonumbering prefixes.</li></ul><h3>Inventory</h3><ul><li><strong>Organization-Level Rules:</strong> Define accounting guardrails, such as whether negative inventory balances are permitted, reorder rule defaults, and ABC breakpoint values.</li><li><strong>Site-Level Rules:</strong> Define local storeroom valuation, such as how issue costs are calculated (e.g., LIFO, FIFO, standard, or average cost) at that specific facility.</li></ul><h3>Procurement, Labor, and Finance (Organization Level Only)</h3><p>Because an Organization is a <strong>financial boundary</strong>, anything touching external money or legal compliance is configured strictly at the <strong>Organization level</strong>:</p><ul><li><strong>Procurement &amp; Contracts:</strong> Rules for converting Purchase Requisitions (PRs) into Purchase Orders (POs), and associating legal terms and conditions with contract types.</li><li><strong>Taxes &amp; Currency:</strong> Calculating multiple tax tiers and managing exchange rates against the Organization’s single base currency.</li><li><strong>Labor &amp; Workflows:</strong> Rules for internal labor costs, outside contractor labor (PO labor actuals), and automated workflow routing.</li></ul><h3>Planned Maintenance &amp; SLAs (Site Level Only)</h3><p>Because day-to-day scheduling happens on the shop floor, operational scheduling is configured at the <strong>Site level</strong>:</p><ul><li><strong>PM Options:</strong> How far in advance to generate preventive maintenance work orders and whether to schedule by priority or frequency.</li><li><strong>SLA Options:</strong> Whether a single work record can match with one or multiple Service Level Agreements.</li></ul><h3>Autonumbering Across Levels</h3><p>Autonumbering prevents record ID collisions and is configured at three distinct tiers depending on how portable the record needs to be:</p><ol><li><strong>System-Level Autonumbers:</strong> Ensure IDs are globally unique across the entire database.</li><li><strong>Organization-Level Autonumbers:</strong> Applied to records that might move between sites within the same financial entity (facilitating cross-site asset transfers without ID conflicts).</li><li><strong>Site-Level Autonumbers:</strong> Applied to localized records like Work Orders, allowing Plant A and Plant B to use custom prefixes (e.g., BOS-1001 vs. DEN-1001).</li></ol><h3>4. Security and Access Control</h3><p>Security sits alongside the hierarchy to control visibility without requiring separate software installations:</p><ul><li><strong>Users &amp; Person Records:</strong> Every User record defines interface behavior and must link to a <strong>Person record</strong> (maintained in the Users application).</li><li><strong>Security Groups:</strong> Every user must belong to at least one Security Group to access applications.</li><li><strong>Selective Site Access:</strong> Security Groups control which specific Sites a user can see. A maintenance technician in Denver can be restricted solely to the Denver Site’s work orders, while a corporate reliability engineer can be granted access to view assets across all Sites.</li><li><strong>System-Level Exemption:</strong> Users do <em>not</em> need access to any specific Site to view or manage data in strictly System-level applications (such as Database Configuration or Currency Codes).</li></ul><h3>5. Decision Framework: How to Design Your Structure</h3><p>When planning your hierarchy, you always work <strong>top-down</strong>: decide on your Organizations first (based on finance), and then decide on your Sites (based on operations).</p><ul><li><strong>Create Multiple Organizations when:</strong> You have multiple legal entities, separate General Ledgers, different base currencies, or completely independent business lines (e.g., a Utilities division vs. a Fleet Management division).</li><li><strong>Create a Single Organization with Multiple Sites when:</strong> You operate across multiple facilities — even in different countries using different vendor currencies (e.g., Peru, Chile, and Argentina) — <strong>provided</strong> they all roll up into a single General Ledger using one base currency, and each facility wants to keep its work orders, PMs, and storerooms separate.</li></ul><h3>6. Walkthrough: The Utility Company Example</h3><p>To see how this works in practice, look at the utility company described at the end of the text:</p><pre>[ SYSTEM: Utility Company Database ]<br>                                       │<br>                  ┌────────────────────┴────────────────────┐<br>         [ Item Set: Utility Parts ]               [ Company Set: Approved Vendors ]<br>                  └────────────────────┬────────────────────┘<br>                                       │<br>       ┌───────────────────────────────┼───────────────────────────────┐<br>       ▼                               ▼                               ▼<br>[ ORG 1: Power Generation ]     [ ORG 2: Water Treatment ]     [ ORG 3: Water Distribution ]<br> (Own GL, Tax &amp; PO Rules)        (Own GL, Tax &amp; PO Rules)       (Own GL, Tax &amp; PO Rules)<br>       │                               │                               │<br>   ┌───┴───┬───────┐               ┌───┴───┬───────┐               ┌───┴───┐<br>   ▼       ▼       ▼               ▼       ▼       ▼               ▼       ▼<br> Site A  Site B  Site C          Site D  Site E  Site F          Site G  Site H<br>(Plant1)(Plant2)(Plant3)        (Plant1)(Plant2)(Plant3)        (Sys 1) (Sys 2)</pre><ul><li><strong>System Level:</strong> The entire utility company runs on a single Maximo database accessed via web browsers.</li><li><strong>Set Level:</strong> Even though Power Generation and Water Treatment are separate divisions, they share an <strong>Item Set</strong> and <strong>Company Set</strong>. If both divisions buy the same industrial valves from the same supplier, that part number and vendor record are created only once.</li><li><strong>Organization Level:</strong> The company splits its operations into three Organizations (<strong>Power Plants</strong>, <strong>Water Treatment</strong>, and <strong>Water Distribution</strong>). This allows each division to maintain its own General Ledger accounting codes, budget calendars, and purchasing approval rules.</li><li><strong>Site Level:</strong> Within the Water Treatment Organization, the three individual treatment plants are set up as three separate <strong>Sites</strong>. Plant 1 cannot accidentally close Plant 2’s work orders or issue parts from Plant 2’s storeroom without explicit cross-site security permissions, keeping day-to-day maintenance clean and independent.</li></ul><p><strong><em>If you found this helpful, clap 👏 (Medium’s version of a high-five) and follow for more on digital transformation in facility management.</em></strong></p><p>— — — — — — — — — — — — — — — — — — — — — — — — — — — — — —</p><p><strong>Contact me on LinkedIn:</strong></p><p><a href="https://www.linkedin.com/in/sayed-saeed-ab3286214/">https://www.linkedin.com/in/sayed-saeed-ab3286214/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1fed63d0ff3a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Mastering IBM Maximo MAS: Why Out-of-the-Box Configuration Should Always Be Your First Move]]></title>
            <link>https://medium.com/@e.sayedsaeed5/mastering-ibm-maximo-mas-why-out-of-the-box-configuration-should-always-be-your-first-move-5f3141fe737e?source=rss-d584a4de0b76------2</link>
            <guid isPermaLink="false">https://medium.com/p/5f3141fe737e</guid>
            <category><![CDATA[mass]]></category>
            <dc:creator><![CDATA[Sayed Saeed]]></dc:creator>
            <pubDate>Wed, 07 Oct 2026 12:46:20 GMT</pubDate>
            <atom:updated>2026-10-07T12:46:20.814Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Striking the perfect balance between core asset management concepts and advanced extensibility in modern EAM architectures.</em></p><p>When architecting a solution in the IBM Maximo Application Suite (MAS), the urge to immediately jump into customization can be incredibly tempting. A stakeholder requests a highly specific business rule, and your mind immediately goes to Automation Scripting, writing Jython or JavaScript logic to force the system to behave exactly as requested.</p><p>While scripting is an undeniably powerful tool in the Maximo arsenal, jumping straight into code is a trap. In modern Enterprise Asset Management (EAM), the golden rule of system longevity, upgradeability, and performance is simple: <strong>Out-of-the-box (OOB) configuration must always be prioritized over custom scripting.</strong></p><p>Here is a breakdown of why this approach is critical for successful Maximo deployments, and how to handle complex requirements when OOB features truly aren’t enough.</p><h3>The True Cost of the “Quick Script”</h3><p>It usually starts innocently enough. You need an attribute to become mandatory based on the value of another field, or you want to auto-populate a record on creation. Writing an Object Launch Point script seems like a five-minute task.</p><p>However, heavy reliance on custom automation scripts and custom applications creates technical debt. Every script you write becomes something you must maintain, test during MAS upgrades, and troubleshoot when unexpected interactions occur. By skipping standard features, you pull focus away from core asset management concepts and risk creating a bloated, heavily customized environment that is difficult for other administrators to manage.</p><h3>Maximizing Core Configurations First</h3><p>Before writing a single line of code, exhaust Maximo’s native configuration toolkit. The platform is designed to handle immense complexity without needing custom scripts:</p><ul><li><strong>Database Configuration &amp; Domains:</strong> Need new data structures or dropdown lists? Standard business objects, attributes, and domains should always be your first stop.</li><li><strong>Conditional Expression Manager:</strong> This is arguably the most underutilized tool by eager developers. You can use conditional expressions tied to Security Groups or Application Designer UI controls to dynamically hide fields, make them read-only, or require input based on specific data states.</li><li><strong>Crossover Domains:</strong> Instead of scripting data retrieval from a parent record (like an Asset) down to a child record (like a Work Order), use Crossover Domains to seamlessly push that data out-of-the-box when a field is populated.</li></ul><p><em>By keeping the configuration within the standard UI tools, you ensure that the system remains easily upgradeable and strictly aligned with standard EAM workflows.</em></p><h3>When Extensibility is Genuinely Needed: The Standalone Approach</h3><p>Of course, not every enterprise requirement can be solved with a Conditional Expression. What happens when a client wants to integrate next-generation features, like 3D BIM Digital Twins, voice-command capabilities, or advanced AI agents for predictive maintenance?</p><p>The answer isn’t to build massive, heavy customizations inside Maximo itself. Instead, the modern best practice is to build a <strong>standalone platform architecture</strong>.</p><p>By developing a decoupled integration engine — for instance, a custom Node.js and Express backend — you can interface with Maximo externally via bi-directional OSLC REST APIs. This keeps Maximo clean and focused on what it does best: core asset and work management.</p><p>Furthermore, if you are integrating artificial intelligence into these standalone platforms to assist facility managers, prioritize direct connections to external API models. Avoid the overhead of managing local, heavy models like Ollama within your infrastructure. A lightweight integration layer connecting external AI APIs to Maximo’s OSLC endpoints provides a much more scalable, responsive, and manageable Command and Control architecture.</p><h3>The Bottom Line</h3><p>A great Solutions Architect doesn’t just know how to write complex code; they know when <em>not</em> to. By thoroughly exhausting out-of-the-box configurations before reaching for Automation Scripts, and by pushing complex integrations to standalone microservices, you ensure your Maximo environments remain resilient, upgrade-ready, and true to the fundamentals of asset management.</p><p><strong><em>If you found this helpful, clap 👏 (Medium’s version of a high-five) and follow for more on digital transformation in facility management.</em></strong></p><p>— — — — — — — — — — — — — — — — — — — — — — — — — — — — — —</p><p><strong>Contact me on LinkedIn:</strong></p><p><a href="https://www.linkedin.com/in/sayed-saeed-ab3286214/">https://www.linkedin.com/in/sayed-saeed-ab3286214/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5f3141fe737e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Mastering Security Groups and User APIs in Maximo Application Suite 9: What’s New and How to…]]></title>
            <link>https://medium.com/@e.sayedsaeed5/mastering-security-groups-and-user-apis-in-maximo-application-suite-9-whats-new-and-how-to-355081d24c96?source=rss-d584a4de0b76------2</link>
            <guid isPermaLink="false">https://medium.com/p/355081d24c96</guid>
            <category><![CDATA[maximo]]></category>
            <category><![CDATA[mass]]></category>
            <dc:creator><![CDATA[Sayed Saeed]]></dc:creator>
            <pubDate>Mon, 05 Oct 2026 08:55:08 GMT</pubDate>
            <atom:updated>2026-10-05T08:55:08.140Z</atom:updated>
            <content:encoded><![CDATA[<h3>Mastering Security Groups and User APIs in Maximo Application Suite 9: What’s New and How to Automate It</h3><p>If you’ve spent any time administering users in IBM Maximo, you know the drill: managing access meant navigating the isolated island of the Maximo Manage application. But with the release of Maximo Application Suite (MAS) 9.1 and 9.2, the paradigm has completely shifted.</p><p><strong>User identity, site access, and API security have been centralized at the Suite level. </strong>More importantly, the underlying REST APIs for provisioning users have been completely overhauled.</p><p>If you are preparing to migrate to MAS 9 or looking to update your automated provisioning scripts, here is a deep dive into the main differences and how to manipulate users and groups in this new environment.</p><h3>The Big Shift: Centralized, Suite-Level Security</h3><p>Before MAS 9.1, security groups were heavily siloed within individual applications. Now, security groups act as the absolute source of truth across the entire suite. You cannot access a suite application, view a dashboard, or make an API call until you are explicitly authorized via a Suite-level security group.</p><p>MAS 9.2 introduced two massive architectural changes to these groups:</p><h3>1. Global Site Assignments</h3><p>In the past, binding a user to a specific facility (a “Site” in Maximo) happened deep inside Maximo Manage. In MAS 9.2, a <strong>Sites tab</strong> now lives directly on the Suite security group.</p><ul><li><strong>The impact:</strong> If a user is assigned to the “Boston Technicians” group, their data visibility is automatically filtered across <em>all</em> MAS applications. They only see work orders, assets, and inventory belonging to Boston. This creates a provable audit trail for compliance and dramatically reduces UI clutter for end-users.</li></ul><h3>2. Object Structure Security (The API Lock-Down)</h3><p>This is the biggest “gotcha” for integration developers. In MAS 9.2 (when mxe.int.enableosauth is enabled), <strong>UI access no longer guarantees API access</strong>.</p><ul><li><strong>The impact:</strong> Even if a user has full rights to update Work Orders in their browser, their REST API calls will fail with a 401/403 error unless their security group explicitly includes access to the MXWO Object Structure on the new <strong>Object structures tab</strong>. This allows you to grant read-write access to human users in the UI, while restricting external integrations to read-only API access using the exact same underlying data structures.</li></ul><h3>The New User APIs: A Tale of Two Object Structures</h3><p>Prior to MAS 9.1, user provisioning via REST could be disjointed. Now, IBM has split user management into two highly specific, purpose-built object structures. Knowing when to use which is the key to successfully automating your user lifecycle.</p><ul><li><strong>MASPERUSER (The Builder):</strong> Built on the PERSON root object. Use this to create users, update biographical data, configure SAML/LDAP Identity Providers (IDPs), and assign security groups.</li><li><strong>MASAPIUSER (The Switchboard):</strong> Built on the MAXUSER root object. Use this strictly for lifecycle triggers: unlocking locked accounts, changing active/inactive status, or executing GDPR-compliant deletions.</li></ul><blockquote><strong><em>Crucial Rule:</em></strong><em> Every single user you create via the API </em>must<em> be assigned to a global system group (typically </em><em>MAXEVERYONE for production or </em><em>EVERYONE for demo environments). If your API payload omits this, the creation will fail</em></blockquote><h3>Hands-On: Manipulating Users with the New APIs</h3><p>Let’s look at how to actually manipulate these new structures using standard REST calls.</p><p><em>(Note: All API calls require a valid </em><em>apikey or </em><em>csrftoken in the headers, and a </em><em>GET call is usually required first to retrieve the user&#39;s encoded </em><em>resourceid before updating).</em></p><h3>Example 1: Assigning Security Groups to a User (MASPERUSER)</h3><p>To add a user to a new security group, you must use the PATCH method. However, because you are updating an array of groups, you <strong>must</strong> use the patchtype: MERGE header. Without it, Maximo will overwrite the entire array and delete any existing group assignments!</p><p><strong>Request:</strong></p><pre>POST /maximo/api/os/masperuser/&lt;resource_id&gt;?lean=1<br>Content-Type: application/json<br>apikey: YOUR_API_KEY<br>x-method-override: PATCH<br>patchtype: MERGE</pre><p><strong>Payload:</strong></p><pre>{<br>  &quot;personid&quot;: &quot;JSMITH&quot;,<br>  &quot;maxuser&quot;: {<br>    &quot;userid&quot;: &quot;JSMITH&quot;,<br>    &quot;groupuser&quot;: [<br>      {<br>        &quot;groupname&quot;: &quot;MAXEVERYONE&quot;<br>      },<br>      {<br>        &quot;groupname&quot;: &quot;BOSTON_MAINTENANCE&quot;<br>      }<br>    ]<br>  }<br>}</pre><h3>Example 2: Instantly Unlocking a User Account (MASAPIUSER)</h3><p>When a user locks themselves out, you don’t update their profile data; you trigger a system method. This is where MASAPIUSER shines. You simply append the wsmethod:unlockUser action to the URL.</p><p><strong>Request:</strong></p><pre>POST /maximo/api/os/masapiuser/&lt;resource_id&gt;?action=wsmethod:unlockUser&amp;lean=1<br>Content-Type: application/json<br>apikey: YOUR_API_KEY<br>x-method-override: PATCH</pre><p><em>(No JSON body is required for this specific action trigger).</em></p><h3>Example 3: Changing a User’s Status with “SYNC”</h3><p>Normally, you need to query the database to get a resource_id before you can PATCH a record. But for rapid status changes (like offboarding an employee), MAS 9 allows a shortcut using the SYNC override. This lets you push a change directly using just their userid.</p><p><strong>Request:</strong></p><pre>POST /maximo/api/os/masapiuser?lean=1<br>Content-Type: application/json<br>apikey: YOUR_API_KEY<br>x-method-override: SYNC<br>patchtype: MERGE</pre><p><strong>Payload:</strong></p><pre>{<br>  &quot;userid&quot;: &quot;JSMITH&quot;,<br>  &quot;_action&quot;: &quot;AddChange&quot;,<br>  &quot;userstatuschange&quot;: [<br>    {<br>      &quot;status&quot;: &quot;INACTIVE&quot;,<br>      &quot;memo&quot;: &quot;Employee departed company&quot;,<br>      &quot;changedate&quot;: &quot;2026-10-05T12:00:00-05:00&quot;<br>    }<br>  ]<br>}</pre><h3>The Bottom Line</h3><p><strong>The transition to MAS 9.1 and 9.2 forces administrators to think at the “Suite” level rather than the application level.</strong> By centralizing Site and Object Structure access into unified Security Groups, IBM has made auditing and compliance vastly easier.</p><p>Meanwhile, the separation of user data (MASPERUSER) and user lifecycle state (MASAPIUSER) provides a much cleaner, safer architecture for writing Python or Node.js provisioning scripts. Embrace the MERGE header, remember your MAXEVERYONE group, and your automated integrations will run smoother than ever.</p><p><strong><em>If you found this helpful, clap 👏 (Medium’s version of a high-five) and follow for more on digital transformation in facility management.</em></strong></p><p>— — — — — — — — — — — — — — — — — — — — — — — — — — — — — —</p><p><strong>Contact me on LinkedIn:</strong></p><p><a href="https://www.linkedin.com/in/sayed-saeed-ab3286214/">https://www.linkedin.com/in/sayed-saeed-ab3286214/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=355081d24c96" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Deploying Maximo Application Suite on Single-Node OpenShift & Untangling JMS Queues]]></title>
            <link>https://medium.com/@e.sayedsaeed5/deploying-maximo-application-suite-on-single-node-openshift-untangling-jms-queues-f1e71ac27607?source=rss-d584a4de0b76------2</link>
            <guid isPermaLink="false">https://medium.com/p/f1e71ac27607</guid>
            <category><![CDATA[ma]]></category>
            <category><![CDATA[openshift]]></category>
            <dc:creator><![CDATA[Sayed Saeed]]></dc:creator>
            <pubDate>Fri, 02 Oct 2026 12:46:01 GMT</pubDate>
            <atom:updated>2026-10-02T12:46:01.502Z</atom:updated>
            <content:encoded><![CDATA[<p>Deploying IBM Maximo Application Suite (MAS) often conjures images of massive, multi-node Red Hat OpenShift clusters. But what if you are building a proof of concept, setting up a disconnected edge site, or supporting a localized environment with fewer than 70 concurrent users?</p><p>Enter the <strong>Single-Node OpenShift (SNO) cluster</strong>.</p><p>An SNO cluster combines both control and worker plane node functions into one machine. While it sacrifices high availability, it drastically reduces your infrastructure footprint. In this guide, we will walk through the exact technical steps to prepare your SNO environment, fix a notorious image registry storage bug, and — as a bonus for Maximo Manage administrators — configure deep JMS message traceability directly in the Application Designer.</p><h3>Part 1: Deploying MAS on Single-Node OpenShift</h3><p>Before starting, ensure your host machine (Bare Metal or VMware vSphere) meets the minimum specifications: <strong>16 vCPUs, 64 GB RAM, and at least 300 GB of primary/ephemeral storage.</strong></p><p>You will also need your Red Hat pull secret (from access.redhat.com), your IBM entitlement key, and your MAS license.dat file.</p><h3>1. Preparing the Docker Container for the MAS CLI</h3><p>To install MAS, we first need to localize our credentials inside a Docker (or Podman) container running the MAS command-line interface.</p><p>First, set up and launch the container:</p><pre>mkdir ~/sno<br>cd ~/sno<br>docker pull quay.io/ibmmas/cli<br>docker run -dit --name sno quay.io/ibmmas/cli:latest bash</pre><p>Next, create the configuration directory inside the container:</p><pre>docker exec -it sno bash<br>mkdir masconfig<br>exit</pre><p>Finally, copy your pull secret and license file into the container:</p><pre>docker cp pull-secret sno:/mascli/masconfig/pull-secret<br>docker cp license.dat sno:/mascli/masconfig/license.dat</pre><h3>2. Configuring Local Volume Manager (LVM) Storage</h3><p>Because SNO lacks centralized, distributed storage, you must configure the node’s local storage to handle PVCs.</p><ul><li>Log in to the OpenShift web console.</li><li>Navigate to <strong>Operators &gt; OperatorHub</strong> and install the <strong>LVM Storage</strong> operator (provided by Red Hat) using default parameters.</li><li>Once installed, click <strong>Create LVMCluster</strong> and accept the defaults. Wait for it to become ready.</li></ul><p><strong>Now, we must force the OpenShift image registry to use this local storage.</strong></p><ul><li>Search for config under <strong>Resources</strong> and select Config imageregistry.operator.openshift.io/v1.</li><li>Edit the YAML with the following three changes:</li></ul><p><strong>Change the rollout strategy:</strong></p><pre>rolloutStrategy: Recreate</pre><p><strong>Assign the storage claim:</strong></p><pre>storage:<br>  pvc:<br>    claim: &#39;&#39;</pre><p><strong>Set the management state:</strong></p><pre>managementState: Managed</pre><h3>3. Troubleshooting the Pending Image Registry PVC</h3><p>A common issue during SNO deployments is the image-registry-storage Persistent Volume Claim (PVC) hanging in a pending state. If this happens, your registry cannot build or push images.</p><p>Navigate to <strong>Storage &gt; PersistentVolumeClaims</strong> and find image-registry-storage. If it is pending, download its YAML and make the following edits:</p><ul><li><strong>Delete</strong> the metadata fields: uid, resourceVersion, and creationTimestamp.</li><li><strong>Delete</strong> the entire manageFields and status sections.</li><li><strong>Modify</strong> the accessModes from ReadWriteMany to ReadWriteOnce.</li></ul><p><strong>Your sanitized YAML should look like this:</strong></p><pre>kind: PersistentVolumeClaim<br>apiVersion: v1<br>metadata:<br>  name: image-registry-storage<br>  namespace: openshift-image-registry<br>  annotations:<br>    imageregistry.openshift.io: &#39;true&#39;<br>    volume.beta.kubernetes.io/storage-provisioner: topolvm.cybozu.com<br>    volume.kubernetes.io/selected-node: 00-50-56-9d-4e-28<br>    volume.kubernetes.io/storage-provisioner: topolvm.cybozu.com<br>  finalizers:<br>    - kubernetes.io/pvc-protection<br>spec:<br>  accessModes:<br>    - ReadWriteOnce<br>  resources:<br>    requests:<br>      storage: 100Gi<br>  storageClassName: odf-lvm-vg1<br>  volumeMode: Filesystem</pre><p>Delete the hung PVC from the OpenShift console (if it gets stuck terminating, simply remove the finalizers block from its YAML). Then, create a new PVC using your edited YAML. It should immediately enter a bound state, clearing the way for your MAS installation via the CLI.</p><h3>Part 2: Untangling JMS Message Consumption in Maximo Manage</h3><p>Once your MAS environment is running, you may encounter asynchronous integration headaches. If a JMSQSEQCONSUMER cron task reads from a continuous queue that <em>already</em> has an active Message Driven Bean (MDB), messages can be processed twice, or hours late.</p><p>To audit this and improve traceability, you can modify the Message Tracking application (INTMSGTRK) to explicitly show whether a Cron Task or an MDB processed a message.</p><p>Here is how to configure it in Application Designer (intmsgtrk.xml) based on your version.</p><h3>For Maximo Manage 9.2</h3><p>In 9.2, you simply need to expose the consumer name in the UI.</p><ul><li>Export intmsgtrk.xml from Application Designer.</li><li>Search for the ID interror_table_1_grid4_13.</li><li>Paste the following line <strong>exactly below</strong> it:</li></ul><pre>&lt;textbox id=&quot;interror_table_1_grid4_14&quot; dataattribute=&quot;CONSUMERNAME&quot; inputmode=&quot;readonly&quot;/</pre><ul><li>Save and re-import the XML.</li></ul><h3>For Maximo Manage 9.1</h3><p>Version 9.1 requires three distinct additions to clearly flag MDB processing versus Cron Task processing.</p><ul><li>Export intmsgtrk.xml.</li><li>To add the column to the list table, search for results_showlist_tablebody_8 and add:</li></ul><pre>&lt;tablecol id=&quot;results_showlist_tablebody_9&quot; dataattribute=&quot;ISJMSMDB&quot; filterable=&quot;true&quot; label=&quot;Processed by MDB&quot;/&gt;</pre><ul><li>To add the checkbox in the main detail screen, search for main_grid1_row1_col1_sec2_6 and add:</li></ul><pre>&lt;checkbox id=&quot;main_grid1_row1_col1_sec2_7&quot; dataattribute=&quot;ISJMSMDB&quot; inputmode=&quot;readonly&quot; label=&quot;Processed by MDB&quot;/&gt;</pre><ul><li>To add the specific consumer name text box, search again for main_grid1_row1_col1_sec2_6 and add:</li></ul><pre>&lt;textbox id=&quot;detail_table_tabledetails_grid1_row_col0_sec0_3&quot; dataattribute=&quot;CONSUMERNAME&quot; inputmode=&quot;readonly&quot;/&gt;</pre><p>Save and import the XML. Moving forward, your Message Tracking UI will explicitly capture and display the consumer ID for all new messages, turning a chaotic queue architecture into a fully auditable integration flow.</p><p><strong><em>If you found this helpful, clap 👏 (Medium’s version of a high-five) and follow for more on digital transformation in facility management.</em></strong></p><p>— — — — — — — — — — — — — — — — — — — — — — — — — — — — — —</p><p><strong>Contact me on LinkedIn:</strong></p><p><a href="https://www.linkedin.com/in/sayed-saeed-ab3286214/">https://www.linkedin.com/in/sayed-saeed-ab3286214/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f1e71ac27607" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Goodbye Generic, Hello Branded: Customizing the UI in Maximo Application Suite 9.0]]></title>
            <link>https://medium.com/@e.sayedsaeed5/goodbye-generic-hello-branded-customizing-the-ui-in-maximo-application-suite-9-0-01b55536dc8e?source=rss-d584a4de0b76------2</link>
            <guid isPermaLink="false">https://medium.com/p/01b55536dc8e</guid>
            <category><![CDATA[ma]]></category>
            <dc:creator><![CDATA[Sayed Saeed]]></dc:creator>
            <pubDate>Thu, 01 Oct 2026 10:45:14 GMT</pubDate>
            <atom:updated>2026-10-01T10:45:14.600Z</atom:updated>
            <content:encoded><![CDATA[<p>Maximo Application Suite (MAS) 9.0 delivers a sleek, unified experience backed by <a href="https://www.carbondesignsystem.com/"><strong>IBM’s Carbon Design System</strong></a>. However, when you are designing a Computerized Maintenance Management System for high-profile enterprise facilities — whether that involves the vast infrastructure of an international airport or the prestigious galleries of a major museum — delivering a generic interface often doesn’t cut it. Your users want an environment that feels like their own.</p><p>Fortunately, MAS 9.0 introduces native User Interface Customization directly within the Suite administration configurations, making it easier than ever to inject your own branding, adjust visual styles, and visually distinguish your environments.</p><p>Here is a breakdown of how you can easily transform the MAS UI without needing complex backend deployments.</p><h3>The Quick Win: Header Configuration</h3><p>Swapping out the default IBM branding for your own company logo is the easiest way to make MAS feel like an in-house application.</p><ol><li>Navigate to your <strong>Suite administration page</strong>.</li><li>Go to <strong>Configurations</strong> &gt; <strong>User interface customization</strong>.</li><li>On the <strong>Header configuration</strong> tab, enter your company and product name.</li><li>Upload your logo (ensure it is a .svg or .png file under 1 MB, ideally 40 pixels in height or less).</li></ol><p>Instantly, the common header reflects your organizational identity across the core asset management applications.</p><h3>Deep Customization: CSS Overrides</h3><p>If you want to go beyond a logo swap, MAS 9.0 allows you to inject custom Cascading Style Sheets (CSS). Because MAS applications share a single Carbon-based skin, injecting a single block of CSS can fundamentally alter colors, typography, and spacing across the suite.</p><p>On the <strong>CSS customization</strong> tab, simply toggle <strong>Enable CSS customization</strong> to <em>on</em>. You can write CSS directly in the browser or upload a file.</p><h3>Example 1: Differentiating Environments</h3><p>A common best practice in enterprise asset management is ensuring users know exactly which environment they are in to prevent accidental data entry in production. You can easily turn the header color of your development or test environment to a distinct color, like dark cyan:</p><pre>.bx--header {<br>  background-color: darkcyan;<br>}</pre><h3>Example 2: Rebranding the Login Page</h3><p>You can fully customize the entry point for your users by adding a custom background image and modifying the welcome text on the login page:</p><pre>/* Add a custom background image */<br>.login-background-image {<br>  background-image: url(https://medium.com/&quot;your_image_url_here&quot;);<br>}</pre><pre>/* Override the default login title text */<br>p#x7ynq::after { <br>  visibility: visible; <br>  content: &quot;Welcome to the Enterprise CMMS&quot;; <br>  position: absolute; <br>  left: 0;<br>  top: 0; <br>} <br>p#x7ynq { <br>  visibility: hidden; <br>  position: relative; <br>}</pre><h3>Example 3: Adjusting the Carbon Skin for Density</h3><p>Some users migrating to MAS 9.0 prefer the tighter spacing and smaller fonts of older Maximo skins to reduce scrolling. IBM provides a template file named Manage_Overrides_tempate.css which includes key CSS classes you can upload to condense the Carbon skin specifically within Maximo Manage, without impacting other apps like Maximo Health.</p><h3>Important Caveats for Architects</h3><p>Before you start overhauling the UI, keep a few technical constraints in mind:</p><ul><li><strong>SaaS Limitations:</strong> If you are running MAS as a Service based on the 9.0 release, this UI customization feature is not currently available.</li><li><strong>Support and Upgrades:</strong> Custom CSS is not officially supported by IBM Support. Because CSS classes can change between releases (especially in newer, role-based apps), you will need to re-test your CSS overrides after any suite upgrade.</li><li><strong>Scope of Impact:</strong> While CSS is shared across most suite applications, your custom changes will <em>not</em> impact Maximo Monitor or the Maximo Mobile apps downloaded from app stores.</li><li><strong>Content Security Policy (CSP):</strong> If you apply CSS changes and they do not render on the frontend, you will likely need to configure your CSP to allow the inline styles or external image URLs.</li></ul><p>Customizing the MAS 9.0 UI bridges the gap between powerful core asset management capabilities and a tailored, user-friendly experience. A few simple CSS rules and a logo upload are all it takes to make the suite feel like a proprietary, purpose-built tool.</p><p><strong><em>If you found this helpful, clap 👏 (Medium’s version of a high-five) and follow for more on digital transformation in facility management.</em></strong></p><p>— — — — — — — — — — — — — — — — — — — — — — — — — — — — — —</p><p><strong>Contact me on LinkedIn:</strong></p><p><a href="https://www.linkedin.com/in/sayed-saeed-ab3286214/">https://www.linkedin.com/in/sayed-saeed-ab3286214/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=01b55536dc8e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Catch Database Drift Before It Kills Your Maximo Upgrade: A Deep Dive into Integrity Checker]]></title>
            <link>https://medium.com/@e.sayedsaeed5/catch-database-drift-before-it-kills-your-maximo-upgrade-a-deep-dive-into-integrity-checker-040a89c43977?source=rss-d584a4de0b76------2</link>
            <guid isPermaLink="false">https://medium.com/p/040a89c43977</guid>
            <category><![CDATA[ma]]></category>
            <category><![CDATA[maximo-db]]></category>
            <category><![CDATA[db-integrity]]></category>
            <dc:creator><![CDATA[Sayed Saeed]]></dc:creator>
            <pubDate>Tue, 29 Sep 2026 11:18:27 GMT</pubDate>
            <atom:updated>2026-09-29T11:18:27.390Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Why </strong><strong>updatedb fails, how Report vs. Repair mode works under the hood, and the real-world errors you have to fix manually.</strong></p><p>Every Maximo administrator has experienced that distinct sinking feeling during an upgrade window: it is 2:00 AM, updatedb is halfway through executing schema migrations, and suddenly a red stack trace halts the script. The culprit is rarely the new release files—it is almost always <strong>database drift</strong>.</p><p>Maximo Asset Management is an abstraction layer built on top of an RDBMS (Oracle, IBM Db2, or Microsoft SQL Server). It tracks its own universe in metadata tables like MAXOBJECT, MAXATTRIBUTE, MAXTABLE, and MAXRELATIONSHIP. When that internal dictionary disagrees with the native database catalog, upgrades do not just fail; they leave environments in an inconsistent, half-migrated state.</p><p>This is why running the <strong>Integrity Checker</strong> before touching an upgrade script is mandatory. Here is a breakdown of how it works, what it can and cannot fix, and the errors you will inevitably encounter in production.</p><h3>The Core Problem: The Split-Brain Data Dictionary</h3><p>Maximo does not query native SQL catalogs on the fly to understand table structures. Instead, it consults its own metadata:</p><ol><li><strong>Maximo Data Dictionary:</strong> System tables (MAXATTRIBUTE, MAXOBJECT, etc.) that define lengths, data types, required flags, and relationships.</li><li><strong>Native Database Catalog:</strong> The physical engine structures (USER_TAB_COLS, sys.columns, or SYSCAT.COLUMNS).</li></ol><p>Over years of system lifecycle changes, drift creeps in. A DBA runs an emergency ALTER TABLE via SQL Developer without touching Database Configuration. A hotfix script terminates halfway through. A custom table is deleted, but its attributes remain logged.</p><p>When an upgrade script attempts to add constraints, reshape indexes, or re-type columns during an upgrade, it assumes the starting point matches the base Maximo spec. If there is a mismatch, the upgrade crashes.</p><h3>Report Mode vs. Repair Mode</h3><p>The utility can be run via the GUI (install_home\maximo\tools\maximo\integrityui.bat) or headless CLI (integritychecker.bat). It operates in two phases:</p><pre>[ Run Report Mode (-r) ] ──&gt; Generates log: integrity&lt;timestamp&gt;.log<br>                                       │<br>                ┌──────────────────────┴──────────────────────┐<br>                ▼                                             ▼<br>        Errors Detected?                              Zero Errors<br>                │                                             │<br>      ┌─────────┴─────────┐                           Ready for Upgrade<br>      ▼                   ▼<br> Auto-Repairable       Manual SQL Required<br>      │                   │<br>[ Repair Mode (-f) ]   [ Resolve data/schema manually ]<br>      │                   │<br>[ Run configdb ]       [ Re-run Report Mode ]</pre><h3>1. Report Mode (-r)</h3><p>Report Mode is strictly read-only. It scans dictionary tables, compares them against the database engine catalog, and categorizes findings into <strong>Info</strong>, <strong>Warning</strong>, and <strong>Error</strong>.</p><blockquote><strong><em>Rule of thumb:</em></strong><em> Warnings (like custom performance indexes) can often remain. Upgrade scripts will only abort on </em><strong><em>Errors</em></strong><em>.</em></blockquote><h3>2. Repair Mode (-f)</h3><p>Repair mode alters metadata and database schemas to bring them back into alignment. It handles mechanical fixes: restoring dropped primary key constraints, aligning sequence numbers, or deleting orphaned attribute definitions.</p><p>Running Repair mode is only half the operation. <strong>You must run </strong><strong>configdb immediately afterward</strong> to apply structural modifications and sync views across the application.</p><h3>Three Real Scenarios: What Gets Fixed and What Blows Up</h3><p>Understanding how Integrity Checker treats discrepancies determines whether an upgrade window stays on schedule.</p><h3>Scenario 1: The Phantom Column (Auto-Repairable)</h3><ul><li><strong>The cause:</strong> Months ago, a consultant created a custom attribute PR_CUSTOM_REF in Database Configuration, but later dropped it directly from Oracle with ALTER TABLE PR DROP COLUMN PR_CUSTOM_REF.</li></ul><p><strong>The log entry:</strong></p><pre>ERROR: MAXATTRIBUTE column PR_CUSTOM_REF exists for table PR, but column does not exist in database catalog.</pre><p><strong>How Repair mode handles it:</strong> Repair mode identifies that the physical column is gone, cleans out the dangling rows from MAXATTRIBUTE and MAXATTRIBUTECFG, and updates the dictionary safely.</p><h3>Scenario 2: The Expanded Column (Auto-Repairable with configdb)</h3><ul><li><strong>The cause:</strong> An integration payload exceeded character limits, so an engineer directly updated SQL Server to VARCHAR(50), while Maximo still believed the field was ALN with length 25.</li><li><strong>The log entry:</strong></li></ul><pre>ERROR: Column length mismatch for ASSET.SERIALNUM.Database catalog length: 50, Maximo dictionary length: 25.</pre><ul><li><strong>How Repair mode handles it:</strong> Repair mode aligns the dictionary length up to 50. Running configdb validates the structural change and updates dependent views without data loss.</li></ul><h3>Scenario 3: The Broken Foreign Key (Manual Resolution Required)</h3><ul><li><strong>The cause:</strong> A poorly constructed data import inserted 300 rows into WORKORDER referencing a LOCATION value that does not exist in the parent table.</li></ul><p><strong>The log entry:</strong></p><pre>ERROR: Foreign key constraint validation failed between WORKORDER and LOCATIONS.   300 orphaned records found.</pre><p><strong>Why Repair mode cannot fix this:</strong> The utility will never guess which location should replace the bad data, nor will it drop 300 business work orders.</p><p><strong>The manual fix:</strong> You must write custom SQL to inspect the affected IDs, re-assign them to a fallback location (or purge test rows), and re-run Integrity Checker in Report mode until the validation passes cleanly.</p><h3>The Pre-Upgrade Playbook</h3><p>Before scheduling downtime or running updatedb:</p><ul><li><strong>Take a Cold Snapshot:</strong> Always back up the database before running Repair mode. Repair mode issues direct DDL and DML statements.</li><li><strong>Never Skip </strong><strong>configdb:</strong> If you run Repair mode, configdb is not optional—skipping it leaves structural DDL half-applied.</li><li><strong>Loop Until Zero Errors:</strong> Re-run Report mode after every batch of fixes. Do not start the upgrade until the log reports 0 Errors.</li></ul><blockquote><strong><em>Treating the Integrity Checker as a routine gate rather than an emergency measure is the difference between a routine morning deployment and a 4:00 AM rollback.</em></strong></blockquote><p><strong><em>If you found this helpful, clap 👏 (Medium’s version of a high-five) and follow for more on digital transformation in facility management.</em></strong></p><p>— — — — — — — — — — — — — — — — — — — — — — — — — — — — — —</p><p><strong>Contact me on LinkedIn:</strong></p><p><a href="https://www.linkedin.com/in/sayed-saeed-ab3286214/">https://www.linkedin.com/in/sayed-saeed-ab3286214/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=040a89c43977" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Breaking the Monolith: The Architectural Evolution from Maximo 7.6 to Maximo Manage]]></title>
            <link>https://medium.com/@e.sayedsaeed5/breaking-the-monolith-the-architectural-evolution-from-maximo-7-6-to-maximo-manage-022303ab3e78?source=rss-d584a4de0b76------2</link>
            <guid isPermaLink="false">https://medium.com/p/022303ab3e78</guid>
            <category><![CDATA[ma]]></category>
            <category><![CDATA[openshift]]></category>
            <dc:creator><![CDATA[Sayed Saeed]]></dc:creator>
            <pubDate>Sun, 27 Sep 2026 14:16:59 GMT</pubDate>
            <atom:updated>2026-09-27T14:16:59.349Z</atom:updated>
            <content:encoded><![CDATA[<p>Enterprise Asset Management (EAM) is undergoing a massive transformation, and nowhere is this more evident than in the evolution of IBM Maximo. If you’ve been working with Maximo over the years, you are likely intimately familiar with the deployment mechanics of Maximo Asset Management 7.6.</p><p>But with the introduction of Maximo Manage within the Maximo Application Suite (MAS), the underlying architecture has fundamentally shifted. We are moving away from legacy application servers and manual builds, fully embracing a cloud-native, containerized ecosystem built on Red Hat OpenShift.</p><p>Let’s dive into what this architectural shift actually looks like and why it matters.</p><h3><strong>The Legacy Approach: Maximo Asset Management 7.6</strong></h3><p>To understand where we are going, we first have to look at where we’ve been. The architecture of Maximo 7.6 is highly centralized around a concept that veteran admins know all too well: <strong>the administrative workstation</strong>.</p><p>In the 7.6 world, this workstation is the hub of all deployment activities. The workflow goes something like this:</p><ol><li><strong>Download:</strong> Software packages, iFixes, and add-ons are manually downloaded from IBM Passport Advantage or Fix Central to the local admin workstation.</li><li><strong>Generate:</strong> The administrator uses command-line tools on this node to compile the massive EAR (Enterprise Archive) and WAR (Web Application Archive) files.</li><li><strong>Deploy:</strong> These files are then deployed to an application server — typically distributed across a WebSphere Application Server Network Deployment to balance the load for the Classic UI, Work Centers, Cron tasks, and BIRT reporting.</li></ol><p>Meanwhile, the Maximo database and external integration systems operate in strict isolation on their own servers. It is a monolithic, manual, and often time-consuming process.</p><h3><strong>The Modern Paradigm: Maximo Manage on Red Hat OpenShift</strong></h3><p>Enter Maximo Manage. The most striking difference in the new architecture is the complete removal of the traditional WebSphere application servers and the legacy admin workstation.</p><p>Instead, the infrastructure layer is entirely powered by the <strong>Red Hat OpenShift Container Platform</strong>.</p><h4>The Rise of the Manage Operator</h4><p>The manual administrative workstation has been replaced by the <strong>Maximo Manage Operator</strong>. Think of the Operator as your automated, always-on deployment manager that natively lives inside OpenShift.</p><p>Here is how deployment flows in the modern era:</p><ul><li><strong>The Entitled Registry:</strong> Instead of manually downloading ZIP files, the Manage Operator pulls container images directly from the IBM Entitled Registry (including base images and industry solutions).</li><li><strong>Customization Archives:</strong> If you have custom Java code, scripts, or configurations, they are stored in a customer repository. The Operator pulls this customization archive securely via HTTPS.</li><li><strong>Automated Builds:</strong> The Operator takes these pieces, builds a configured administrative image, and spins up the necessary workloads.</li></ul><h4>Workload Clusters</h4><p>Instead of relying on WebSphere JVMs, Maximo Manage deploys discrete, lightweight container pods for different workloads:</p><ul><li><strong>UI Deployment</strong></li><li><strong>Cron Deployment</strong></li><li><strong>Report Deployment</strong></li><li><strong>Maximo Enterprise Adapter (MEA) Deployment</strong></li></ul><p>Because these are containerized workloads, they can be scaled up or down elastically based on demand. There is also a dedicated <strong>Admin Command Processor</strong> pod that handles backend tools, like the integrity checker, without ever disrupting the user-facing UI pods.</p><h3>A Unified Ecosystem</h3><p>This architectural shift isn’t just about deploying the core EAM application differently; it’s about integration. Maximo Manage is co-deployed within the broader <strong>Maximo Application Suite</strong>. This means your asset management system shares the exact same OpenShift infrastructure as other powerful tools like Maximo Health, Maximo Predict, and Maximo Monitor.</p><p>Furthermore, this modernization extends to data storage and messaging. Maximo Manage allows you to easily configure IBM Cloud Object Storage for storing document links, and utilize scalable messaging providers like Kafka or IBM MQ clusters.</p><h3>Final Thoughts</h3><p>The transition from Maximo 7.6 to Maximo Manage is a true paradigm shift. By trading monolithic application servers for Kubernetes-based container orchestration, organizations gain unprecedented scalability, automated deployments, and a robust foundation for integrating AI and IoT directly into their asset management workflows.</p><p>The days of manually building EAR files are behind us. The cloud-native future of EAM is here.</p><p><em>Have you started your migration journey to the Maximo Application Suite? What architectural changes are you most excited (or nervous) about? Let me know in the comments below!</em></p><p><strong><em>If you found this helpful, clap 👏 (Medium’s version of a high-five) and follow for more on digital transformation in facility management.</em></strong></p><p>— — — — — — — — — — — — — — — — — — — — — — — — — — — — — —</p><p><strong>Contact me on LinkedIn:</strong></p><p><a href="https://www.linkedin.com/in/sayed-saeed-ab3286214/">https://www.linkedin.com/in/sayed-saeed-ab3286214/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=022303ab3e78" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Bypassing BMXAA4042E: How to Bulk Cancel Workflow-Locked Work Orders in IBM Maximo]]></title>
            <link>https://medium.com/@e.sayedsaeed5/bypassing-bmxaa4042e-how-to-bulk-cancel-workflow-locked-work-orders-in-ibm-maximo-9b3b288073a2?source=rss-d584a4de0b76------2</link>
            <guid isPermaLink="false">https://medium.com/p/9b3b288073a2</guid>
            <category><![CDATA[escalation]]></category>
            <category><![CDATA[maximo]]></category>
            <dc:creator><![CDATA[Sayed Saeed]]></dc:creator>
            <pubDate>Sun, 20 Sep 2026 13:45:48 GMT</pubDate>
            <atom:updated>2026-09-20T13:45:48.537Z</atom:updated>
            <content:encoded><![CDATA[<blockquote>Every Maximo administrator or consultant eventually encounters a bulk cleanup request: <em>“Cancel several hundred legacy work orders sitting in backlog across our sites.”</em></blockquote><p>It sounds trivial. You build a query, navigate to Work Order Tracking, select all, and trigger <strong>Change Status</strong> to CAN.</p><p>Instead of success, Maximo throws a roadblock:</p><blockquote><strong><em>BMXAA4042E</em></strong><em> — </em>The status cannot be changed for the selected records.</blockquote><p>Direct database deletions (DELETE FROM WORKORDER) violate referential integrity and leave orphaned records across child tables ( WOACTIVITY, WPMATERIAL, DOCLINKS). The application layer refuses to cooperate because business validation halts the batch if even a single work order is locked in an active workflow.</p><p>Here is an architectural breakdown of why this happens and how to orchestrate a two-stage escalation pattern to safely purge the backlog without writing complex code.</p><h3>Why Bulk Status Changes Fail on Backlog Work Orders</h3><p>Work orders stuck in states like WAPPR (Waiting on Approval), WMATL (Waiting on Material), or WRESCH (Waiting on Reschedule) frequently carry active workflow assignments.</p><p>When a record is routed into workflow:</p><ol><li>Maximo provisions an active record inside the workflow engine table (WFINSTANCE.ACTIVE = 1).</li><li>An interactive lock governs state transitions. The MBO validation framework enforces that status progression occurs through workflow assignment completions (actions, task nodes, manual routing) — not through standard UI status updates.</li><li>If an escalation or batch update attempts to force STATUS = &#39;CAN&#39;, the validation framework rejects the record to protect workflow integrity.</li></ol><p>Attempting to change status and stop workflows in a single standard action fails because <strong>Change Status</strong> actions lack built-in hooks to gracefully kill running workflow processes first.</p><h3>The Solution: A Decoupled, Two-Stage Escalation Pattern</h3><p>To cancel these records cleanly through the supported business framework, the process is decoupled into two automated, sequential escalations:</p><pre><strong>[Escalation 1: Target WFINSTANCE] ──► Force-stop active workflows cleanly<br></strong>                                               │<br>                                               ▼<br><strong>[Escalation 2: Target WORKORDER]  ──► Transition unlocked records to CAN</strong></pre><h3>Stage 1: Stop Active Workflows (WFINSTANCE)</h3><p>Rather than inspecting every work order from the parent table, target the active workflow instances directly.</p><h3>1. The Scoping Query</h3><p>Because WFINSTANCE tracks processes across the entire enterprise (PRs, POs, Service Requests, Contracts), scoping requires two mandatory constraints:</p><ul><li>OWNERTABLE = &#39;WORKORDER&#39; (prevents impacting other modules sharing the same numerical IDs).</li><li>An optimized subquery restricting the target site, statuses, and scheduling threshold.</li></ul><pre>ACTIVE = 1 <br>  AND OWNERTABLE = &#39;WORKORDER&#39; <br>  AND EXISTS (<br>    SELECT 1 <br>    FROM WORKORDER W <br>    WHERE W.WORKORDERID = WFINSTANCE.OWNERID <br>      AND W.SITEID = &#39;Alex&#39; <br>      AND W.STATUS IN (&#39;WAPPR&#39;, &#39;WRESCH&#39;, &#39;WMATL&#39;) <br>      AND W.TARGSTARTDATE &lt; &#39;2026-07-30&#39;<br>  )</pre><blockquote><strong><em>Performance Tip:</em></strong><em> An </em><em>EXISTS semi-join allows Db2 to leverage the primary key index on </em><em>WORKORDERID rather than scanning large subsets using </em><em>IN (...).</em></blockquote><h3>2. The Termination Action</h3><ul><li>Navigate to <strong>Platform Configuration &gt; Automation Scripts</strong>.</li><li>Create a script with an Action launch point targeting the WFINSTANCE object:</li></ul><pre># Launch Point on WFINSTANCE<br>if mbo.getBoolean(&quot;ACTIVE&quot;):<br>    mbo.stopWorkflow(&quot;Automated bulk cancellation preparation via Escalation&quot;)</pre><ul><li>Navigate to <strong>Platform Configuration &gt; Escalations</strong>.</li><li>Create an escalation against WFINSTANCE using the condition above.</li><li>Link the point to your action script and activate the escalation.</li></ul><p>Once triggered, this clears the workflow runtime records and releases the edit locks on all target work orders.</p><h3>Stage 2: Mass Status Transition to CAN (WORKORDER)</h3><p>With active workflows cleared, the work orders behave as standard editable records and can safely transition through the business layer.</p><h3>1. The Action Definition</h3><ul><li>Open the <strong>Actions</strong> application.</li><li>Create action WOCANCEL:</li><li><strong>Object:</strong> WORKORDER</li><li><strong>Type:</strong> Change Status</li><li><strong>Value:</strong> CAN</li></ul><h3>2. The Work Order Escalation</h3><ul><li>Open <strong>Escalations</strong> and create an escalation targeting WORKORDER.</li><li><strong>Condition:</strong></li></ul><pre>SITEID = &#39;Alex&#39;    AND STATUS IN (&#39;WAPPR&#39;,&#39;WMATL&#39;)    AND TARGSTARTDATE &lt; &#39;2026-07-30&#39;</pre><ul><li>Under <strong>Escalation Points</strong>, create a record, check <strong>Repeat</strong>, and bind the WOCANCEL action.</li><li>Validate and activate.</li></ul><p>When the crontask executes, it transitions each record to CAN, updates status audit logs in WOSTATUS, cascades down to tasks, and bypasses the BMXAA4042E validation error.</p><h3>Key Takeaways for Maximo Architects</h3><ul><li><strong>Avoid Direct SQL Deletes:</strong> Bypassing the MBO layer corrupts system relationships and prevents necessary transaction triggers from executing.</li><li><strong>Scope </strong><strong>WFINSTANCE Carefully:</strong> Forgetting OWNERTABLE = &#39;WORKORDER&#39; risks aborting workflows on purchase orders or requisitions that happen to share internal integer keys.</li><li><strong>Decouple Complex Batch Jobs:</strong> Attempting to handle workflow shutdowns, status assignments, and child task closures in a single monolithic rule increases failure surface area. Splitting cleanup into discrete, repeatable escalations yields observable, auditable batch processing.</li></ul><p><strong><em>If you found this helpful, clap 👏 (Medium’s version of a high-five) and follow for more on digital transformation in facility management.</em></strong></p><p>— — — — — — — — — — — — — — — — — — — — — — — — — — — — — —</p><p><strong>Contact me on LinkedIn:</strong></p><p><a href="https://www.linkedin.com/in/sayed-saeed-ab3286214/">https://www.linkedin.com/in/sayed-saeed-ab3286214/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9b3b288073a2" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Architecting for Reality, Part 3: Elasticity, Metrics, and the Math of Autoscaling]]></title>
            <link>https://medium.com/@e.sayedsaeed5/architecting-for-reality-part-3-elasticity-metrics-and-the-math-of-autoscaling-691ec6fb5473?source=rss-d584a4de0b76------2</link>
            <guid isPermaLink="false">https://medium.com/p/691ec6fb5473</guid>
            <category><![CDATA[openshift]]></category>
            <dc:creator><![CDATA[Sayed Saeed]]></dc:creator>
            <pubDate>Mon, 07 Sep 2026 21:01:01 GMT</pubDate>
            <atom:updated>2026-09-07T21:01:01.112Z</atom:updated>
            <content:encoded><![CDATA[<p><em>In Part 2 of this series, we explored how OpenShift handles chaos. We built a self-healing architecture using loose coupling, ReplicaSets, and aggressive Liveness and Readiness probes.</em></p><p>But what happens when your system is perfectly healthy, but completely overwhelmed?</p><p>When you are architecting an enterprise asset management (EAM) system or scaling out an IBM Maximo deployment for a mega-project, predicting exact resource utilization is nearly impossible. You might have 50 maintenance technicians logging work orders today, and 5,000 sensors triggering automated alerts tomorrow.</p><p>If you statically provision virtual machines for peak load, you waste thousands of dollars in idle compute. If you provision for average load, your system crashes on opening day.</p><p><strong>In OpenShift, the answer is elasticity. We don’t guess the required capacity; we let the cluster calculate it in real-time.</strong> Here is how modern OpenShift utilizes metrics, enforces resource limits, and orchestrates the Horizontal Pod Autoscaler (HPA) to survive massive traffic spikes.</p><h4>The Foundation: Observability and the Metrics Stack</h4><p>Before OpenShift can dynamically scale an application, it has to know exactly what that application is doing.</p><p>In older versions of OpenShift (3.x), this was handled by a clunky stack of Hawkular, Heapster, and Cassandra. In modern OpenShift 4, observability is built around <strong>Prometheus</strong>.</p><p>When you deploy a cluster, the Prometheus Operator is automatically provisioned. It continuously scrapes metrics directly from the kubelet process running on every single worker node. These metrics—specifically CPU and memory consumption—are fed into the Kubernetes Metrics API. This API is the central nervous system that OpenShift uses to make its scaling decisions.</p><h4>The Guardrails: Requests, Limits, and QoS</h4><p>Before you can tell OpenShift to scale your application when it reaches “75% capacity”, you have to define what 100% actually means for that specific pod.</p><p>We do this by establishing Resource Requests and Limits in our Deployment configuration. In the Kubernetes world, CPU is measured in “millicores” (m), where 1000m equals one full CPU core.</p><ul><li><strong>The Request (The Guarantee):</strong> This is the minimum amount of resources OpenShift promises to reserve. If your Node.js pod requests 400m of CPU, the scheduler will strictly place it on a worker node that has at least 40% of a core sitting idle.</li><li><strong>The Limit (The Ceiling):</strong> This is the absolute maximum the pod can consume. If a memory leak causes the pod to exceed its memory limit, the Linux kernel ruthlessly terminates the container (OOMKilled). If it hits its CPU limit, the process is aggressively throttled.</li></ul><p>By configuring these two values, you inadvertently assign a <strong>Quality of Service (QoS)</strong> tier to your pod:</p><ol><li><strong>Guaranteed:</strong> Request and Limit are exactly the same. OpenShift treats this as a VIP workload. It will rarely, if ever, be evicted.</li><li><strong>Burstable:</strong> The Limit is higher than the Request. The pod is guaranteed its baseline, but can “burst” into unallocated node resources during a traffic spike.</li><li><strong>Best Effort:</strong> No Requests or Limits are defined. These pods are the first to be killed if a worker node starts running out of memory.</li></ol><p>Here is what a Burstable configuration looks like in your Deployment YAML:</p><pre>resources:<br>  requests:<br>    cpu: 400m<br>    memory: 256Mi<br>  limits:<br>    cpu: 800m<br>    memory: 512Mi</pre><h4>The Engine: The Horizontal Pod Autoscaler (HPA)</h4><p>With our resource baseline established, we can automate the scaling process using the <strong>Horizontal Pod Autoscaler (HPA)</strong>.</p><p>The HPA continuously queries the Metrics API. It compares the <em>actual</em> CPU or memory usage of your pods against a target threshold.</p><p>Setting this up via the CLI requires a single command:</p><pre><strong>oc autoscale deploy/app-cli --min=2 --max=10 --cpu-percent=75</strong></pre><p>Let’s break down the math. We set our CPU Request to 400m. We told the HPA to target 75%. This means the target CPU usage per pod is 300m. If a massive wave of API traffic hits the cluster and pushes the average CPU consumption across your pods to 500m, the HPA calculates that it needs more replicas to bring the average back down to 300m. It immediately instructs the ReplicaSet to spin up more pods (up to our maximum of 10).</p><h4>The Reality Check: Preventing “Thrashing”</h4><p><em>Scaling up is easy. Scaling down is where architectures fail.</em></p><p>When your traffic spikes, OpenShift reacts within seconds. But when the traffic suddenly drops, OpenShift does <em>not</em> immediately kill off the extra pods.</p><p>This delay is a deliberate architectural defense mechanism against <strong>thrashing</strong>. Thrashing is the chaotic, inefficient cycle of continuously starting and stopping pods due to minor metric fluctuations. Deploying a new pod is expensive — it requires scheduling overhead, image pulls, and application startup sequences.</p><p>To stabilize the cluster, the HPA controller enforces a “cooldown” window. By default, after a traffic spike subsides, OpenShift waits <strong>5 minutes</strong> before terminating any pods. This ensures the drop in traffic was an actual end to the load, rather than a brief dip before a second massive spike.</p><h4>The Bottom Line</h4><p>Virtual machines force you to architect for the “worst-case scenario.” Container orchestration allows you to architect for the “current scenario.”</p><p>By establishing strict resource boundaries and trusting the HPA to read Prometheus metrics, your applications become elastic. They breathe in when the users arrive, and they breathe out when the users leave — saving compute, saving money, and ensuring your mega-projects stay online.</p><p><strong><em>If you found this helpful, clap 👏 (Medium’s version of a high-five) and follow for more on digital transformation in facility management.</em></strong></p><p>— — — — — — — — — — — — — — — — — — — — — — — — — — — — — —</p><p><strong>Contact me on LinkedIn:</strong></p><p><a href="https://www.linkedin.com/in/sayed-saeed-ab3286214/">https://www.linkedin.com/in/sayed-saeed-ab3286214/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=691ec6fb5473" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>