{"id":1154,"date":"2026-10-31T01:25:00","date_gmt":"2026-10-31T01:25:00","guid":{"rendered":"https:\/\/www.jetexe.com\/blog\/?p=1154"},"modified":"2026-10-31T01:25:00","modified_gmt":"2026-10-31T01:25:00","slug":"secops-slipping-fixes-diagnosis","status":"publish","type":"post","link":"https:\/\/www.jetexe.com\/blog\/secops-slipping-fixes-diagnosis\/","title":{"rendered":"Security Fixes Keep Slipping? The SecOps Diagnosis"},"content":{"rendered":"<p>Every ops conversation we have with charter and jet operators eventually hits the same complaint: the security findings list never shrinks. Items roll from this quarter to next, the same three servers appear on every report, and nobody can quite say whose desk the fix sits on. Here is the uncomfortable part \u2014 that is not a tooling problem, a budget problem, or a laziness problem. It is a contract problem, and it has a name. What the industry keeps calling <strong>SecOps<\/strong> is, at the Atlas&#8217;s definition, a working deal between the security function and the operations function: baselines both sides accept, one remediation queue where every row carries an owner and a date, exceptions with shelf lives, and numbers both teams own. Diagnose your slipping fixes against that deal and the fault usually surfaces in one of four places. We will walk through each, then the repair.<\/p>\n<h2>The symptom, described honestly<\/h2>\n<p>Run this quick check at your next ops meeting. Ask for three numbers: how many open findings have no named owner; how many approved exceptions are past their review date; and what percentage of fixes closed on time in your most exposed cell \u2014 internet-facing systems at high urgency. In the Atlas&#8217;s September worked example those numbers were 18 of 600 unowned (three percent), 6 of 35 exceptions stale, and a headline &#8220;82.5% on time&#8221; that concealed a 75% cell because its maintenance window kept moving. If your versions of those numbers are unknown, or worse than that, you do not have a queue; you have a rumour of a queue.<\/p>\n<p>Charter and jet operators feel this sharper than most industries, and not because their technology is exotic. An aircraft on the ground is revenue gone, so maintenance items get owners and clocks almost by reflex. A security finding is the same arithmetic wearing a suit: exposure accrues every day it sits unowned, the cost is simply invisible until the day it is not. The habit is already in the building. It just needs pointing at a different kind of risk.<\/p>\n<h2>Fault one: the SecOps report goes over the wall<\/h2>\n<p>The classic failure. Security runs a scan, exports four thousand rows to a spreadsheet, emails it to &#8220;ops&#8221;, and considers the job done. Nobody accepts the report because nobody was asked to. The repair is routing, not volume: each finding lands on one person&#8217;s queue with a date, ideally automatically \u2014 platforms such as ServiceNow&#8217;s Yokohama release drop findings straight onto the desk of whichever support group owns the configuration item concerned, reading the CMDB so the assignment survives staff changes. That removes the &#8220;I didn&#8217;t see it&#8221; defence for everyone. Broadcast reports feel like communication; they are actually delegation refused.<\/p>\n<h2>Fault two: SecOps deadlines exist on paper but not in the rota<\/h2>\n<p>A deadline nobody can staff is a wish. The contract insists timeframes are resourced per owner group, which in practice means the change calendar has real slots for security work. This is where the per-cell view earns its keep: the September example&#8217;s lagging cell was lagging for exactly one reason \u2014 its change window kept slipping. Internet-facing high-urgency work hit 95% on time in the same estate, so the capability existed; the calendar just never protected the datacentre&#8217;s turn. Solution: book security fixes like any other maintenance slot, protected, with the same no-bump courtesy you give an inspection.<\/p>\n<h2>Fault three: hardening changes get switched on cold<\/h2>\n<p>Somewhere along the way, someone decides Friday afternoon is a great time to enforce a new baseline across everything. The predictable result is an outage, a rollback, and a standing excuse to skip the next hardening round. Configuration guidance has preached the staged path for years \u2014 SP 800-128 describes select, tailor, test, staged enforcement \u2014 and the Atlas counts &#8220;big-bang hardening without a pilot&#8221; among the canonical anti-patterns. Its disruption evidence: 3 incidents out of 120 security changes in one month, every single one caused by an estate-wide baseline shoved out with zero pilot work behind it. Staging is not caution; it is how the second half of the programme keeps its permission to operate.<\/p>\n<h2>Fault four: &#8220;done&#8221; gets declared at installation, not at clearance<\/h2>\n<p>The subtlest fault, and the one that bites hardest under vendor pressure. Installing an update is not the finish line; clearing the exceptions is. Microsoft&#8217;s Netlogon work on CVE-2020-1472 made that distinction famous: enforcement ran in phases from 2020-08-11, event IDs 5827 and 5828 recorded denied connections while 5830 and 5831 recorded the permitted ones, and the whole effort counted as a success only because the exception list had emptied out ahead of the hard cutoff on 2021-02-09. Had any legacy machine stayed on that list, the estate would have been &#8220;updated&#8221; and still exposed. Define done as cleared, and your queue&#8217;s dates start meaning something.<\/p>\n<h2>The repair, in one page<\/h2>\n<p>Good news: the fix for all four faults is the same document, and it fits on one page:<\/p>\n<ol>\n<li>Name the single queue and the person who chairs it.<\/li>\n<li>Attach one owner and one date to every finding \u2014 no nameless rows, ever.<\/li>\n<li>Reserve protected change slots per owner group, and let the per-cell numbers judge whether they hold.<\/li>\n<li>Pilot every hardening change, log-only first, then ring by ring.<\/li>\n<li>Give every exception an expiry, and count the stale ones weekly.<\/li>\n<\/ol>\n<p>Notice what the list does not contain: a purchase order. Tools can accelerate every line above, and some are worth their price, but none of them can substitute for a named human and a defended calendar slot. When the numbers improve \u2014 and within a quarter they will \u2014 resist the urge to celebrate the headline percentage. Publish the per-cell table instead, with the unowned count and the stale-exception count alongside it, because those three figures together are what stop the slips from creeping back.<\/p>\n<p>That list is the working contract in miniature; the full version, with its lifecycle and metrics, is set out in <a href=\"https:\/\/www.xopsschool.com\/secops\/\">the Atlas&#8217;s SecOps reference page<\/a>. Operators who want the underlying habits in their own domain will recognise them everywhere \u2014 the yield-management thinking behind <a href=\"https:\/\/www.jetexe.com\/blog\/how-empty-leg-jet-flights-work\/\">how empty leg jet flights work<\/a>, the planning discipline in this <a href=\"https:\/\/www.jetexe.com\/blog\/empty-leg-flight-guide-for-private-jet-travelers\/\">empty leg guide for private jet travelers<\/a>, and the preparation mindset of <a href=\"https:\/\/www.jetexe.com\/blog\/motorcycle-rental-in-canada-choose-your-ride-and-make-every-mile-count\/\">choosing your ride for a long Canadian run<\/a>. Aircraft teaches the pattern: nothing flies on vibes, and neither should your fix queue. Diagnose the four faults, sign the one-pager, and watch next quarter&#8217;s list finally shrink.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Four faults explain every slipping security fix: reports over the wall, unresourced deadlines, cold hardening, and &#8220;done&#8221; declared too early.<\/p>\n","protected":false},"author":7,"featured_media":1155,"comment_status":"","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[913],"tags":[],"class_list":["post-1154","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-aviation-operations"],"_links":{"self":[{"href":"https:\/\/www.jetexe.com\/blog\/wp-json\/wp\/v2\/posts\/1154","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.jetexe.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.jetexe.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.jetexe.com\/blog\/wp-json\/wp\/v2\/users\/7"}],"replies":[{"embeddable":true,"href":"https:\/\/www.jetexe.com\/blog\/wp-json\/wp\/v2\/comments?post=1154"}],"version-history":[{"count":1,"href":"https:\/\/www.jetexe.com\/blog\/wp-json\/wp\/v2\/posts\/1154\/revisions"}],"predecessor-version":[{"id":1159,"href":"https:\/\/www.jetexe.com\/blog\/wp-json\/wp\/v2\/posts\/1154\/revisions\/1159"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.jetexe.com\/blog\/wp-json\/wp\/v2\/media\/1155"}],"wp:attachment":[{"href":"https:\/\/www.jetexe.com\/blog\/wp-json\/wp\/v2\/media?parent=1154"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.jetexe.com\/blog\/wp-json\/wp\/v2\/categories?post=1154"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.jetexe.com\/blog\/wp-json\/wp\/v2\/tags?post=1154"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}