{"id":6334,"date":"2026-11-16T07:45:00","date_gmt":"2026-11-16T07:45:00","guid":{"rendered":"https:\/\/www.wakilsahab.in\/lawyers\/?p=6334"},"modified":"2026-11-18T16:45:00","modified_gmt":"2026-11-18T16:45:00","slug":"directory-recovery-glossary","status":"publish","type":"post","link":"https:\/\/www.wakilsahab.in\/lawyers\/directory-recovery-glossary\/","title":{"rendered":"DisasterRecoveryOps for Online Directories: A Primer and Glossary"},"content":{"rendered":"<p>People who use an online directory to find an advocate expect it to be there when they need it. A user searching for a specialist in a particular city is often in a hurry, and a blank page or a missing profile loses their trust quickly. For the teams who run such directories, and for readers curious about how serious outages are handled behind the scenes, this primer introduces DisasterRecoveryOps in measured, jargon-light terms. It is a technology explainer, not legal commentary, and it closes with a glossary.<\/p>\n<h2>Why a directory is a good test case<\/h2>\n<p>A legal directory holds several kinds of data with different needs. Profiles and listings change occasionally. Enquiry forms capture new messages all day. Search indexes are rebuilt regularly. Because each part has a different tolerance for loss and delay, a directory shows clearly why recovery cannot be one-size-fits-all.<\/p>\n<p>Pages such as <a href=\"https:\/\/www.wakilsahab.in\/lawyers\/top-10-best-intellectual-property-lawyers-advocate-in-guwahati\/\">the Guwahati intellectual property advocates page<\/a> and <a href=\"https:\/\/www.wakilsahab.in\/lawyers\/top-10-best-intellectual-property-lawyers-advocate-in-madurai\/\">its Madurai counterpart<\/a> are the kind of content visitors expect to load at any hour. Behind pages like these sits a database, a web server and a set of scheduled jobs, each of which can fail.<\/p>\n<h2>The core idea in three questions<\/h2>\n<ol>\n<li><strong>How long can we be down?<\/strong> The answer is a time limit, measured from the moment service stops.<\/li>\n<li><strong>How much recent change can we afford to lose?<\/strong> The answer is a point in time to which the data can be restored.<\/li>\n<li><strong>Who decides to switch, and by when?<\/strong> The answer is a named person and a deadline.<\/li>\n<\/ol>\n<p>The Ops Atlas entry on this discipline organises its guidance around these questions and adds the practice of rehearsing the answers with a stopwatch. The reference page is <a href=\"https:\/\/www.xopsschool.com\/disasterrecoveryops\/\">the Atlas explanation of DisasterRecoveryOps<\/a>, and the rest of this piece summarises its main points in beginner-friendly language.<\/p>\n<h2>Why waiting counts<\/h2>\n<p>The most counterintuitive lesson concerns time. Beginners imagine that recovery time is how long computers take to start on another site. In the Atlas it covers the whole span from the start of the failure, which includes the minutes people need to notice the problem and to decide. Its worked drill had six minutes of detection, twenty-five of deliberation and thirty-three of technical work, giving sixty-four against a sixty-minute goal.<\/p>\n<h2>Levels of protection<\/h2>\n<p>Four levels of readiness are usually described. The simplest keeps copies and rebuilds afterwards. The next keeps a minimal live core ready to enlarge. Then there is a smaller but working replica, and finally two sites sharing live traffic. Each level shortens the outage and raises the cost. For a directory, the enquiry form might deserve a higher level than the archive of old listings, since lost enquiries are lost opportunities for the people who sent them.<\/p>\n<h2>A quiet example of tiering<\/h2>\n<p>Consider the platform&#8217;s parts one at a time. Public listing pages are read constantly but change rarely, so they can be rebuilt from a recent copy and served from a standby with little drama. The enquiry inbox changes every few minutes and deserves a much tighter data-freshness limit, because every missing message is a person who waited for a reply that never came. Internal reporting tools can wait a day. Putting those three in different tiers, with different standby levels and different drill frequencies, is the sort of proportionate thinking the Atlas sample matrix encourages, with its tiers ranging from one hour down to seventy-two.<\/p>\n<h2>What can go wrong that people forget<\/h2>\n<ul>\n<li><strong>Copies stored next to the originals.<\/strong> A building problem takes both.<\/li>\n<li><strong>Mirrors that copy mistakes.<\/strong> If a bad update corrupts the listings, the mirror copies the damage. Keep older versions as well.<\/li>\n<li><strong>Logins that depend on the failed system.<\/strong> Emergency access belongs somewhere independent.<\/li>\n<li><strong>Untested return trips.<\/strong> Going back to the original site needs its own plan.<\/li>\n<\/ul>\n<h2>A note on professional services and trust<\/h2>\n<p>Directories that list professionals carry a duty of care towards the public. Listing pages should not claim more than the operator can support, and operational plans should be written in the same spirit: modest, specific and tested. The Atlas itself is careful in this way, labelling a term as rare when it is rare and marking vendor terminology as vendor framing.<\/p>\n<h2>Mini-glossary<\/h2>\n<ul>\n<li><strong>Recovery time objective:<\/strong> the longest acceptable outage.<\/li>\n<li><strong>Recovery point objective:<\/strong> the point in time to which data can be restored.<\/li>\n<li><strong>Failover:<\/strong> an urgent move to the standby that may lose very recent changes.<\/li>\n<li><strong>Switchover:<\/strong> a planned move with writes paused first.<\/li>\n<li><strong>Failback:<\/strong> the return to the original site.<\/li>\n<li><strong>Fencing:<\/strong> stopping the old side from accepting new work.<\/li>\n<li><strong>Tier:<\/strong> a rank deciding how much protection a service receives.<\/li>\n<li><strong>Decision deadline:<\/strong> the latest time at which the team must commit to the move.<\/li>\n<\/ul>\n<h2>How to start without a big budget<\/h2>\n<p>Begin by listing the parts of the platform and writing a tolerable outage next to each. Choose the first service to rehearse, usually the one that receives enquiries. Compute a decision deadline from the objective, the slowest recent restore and a modest margin. The Atlas example subtracts thirty-three and five from sixty to reach twenty-two minutes; yours will differ.<\/p>\n<p>Then time a drill, write down what happened and fix the biggest delay. The Atlas ladder starts with objectives that have been signed off and moves through checked plans and exercised plans towards switchovers carried out routinely in production. Most small operators will be content with the third step.<\/p>\n<h2>A word on who owns what<\/h2>\n<p>Recovery planning also clarifies responsibilities. A hosting company runs the machines; the operator runs the service. One cloud vendor&#8217;s own documentation, as quoted in the Atlas, makes clear that customers must arrange DR themselves rather than expecting it to arrive with the account. If the operator has not written down who decides, who communicates with users and who restores what, nobody else will do it. A short responsibility table, reviewed twice a year, is enough for most small platforms.<\/p>\n<h2>What to remember<\/h2>\n<p>Before the summary, one practical reminder: write the plan so that a stressed stranger could follow it. Short sentences, numbered steps and named contacts beat a long narrative every time. The takeaways are modest. Know your limits in minutes. Include the humans in the clock. Match protection to importance. Practise before you need it. If a directory with thousands of profiles can follow those four rules, so can almost any other service.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>How a professional listings directory can think about outages, recovery limits and decision deadlines, explained in plain terms with a short glossary.<\/p>\n","protected":false},"author":2,"featured_media":6335,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[515],"tags":[],"class_list":["post-6334","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-legal-tech"],"_links":{"self":[{"href":"https:\/\/www.wakilsahab.in\/lawyers\/wp-json\/wp\/v2\/posts\/6334","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.wakilsahab.in\/lawyers\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.wakilsahab.in\/lawyers\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.wakilsahab.in\/lawyers\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.wakilsahab.in\/lawyers\/wp-json\/wp\/v2\/comments?post=6334"}],"version-history":[{"count":1,"href":"https:\/\/www.wakilsahab.in\/lawyers\/wp-json\/wp\/v2\/posts\/6334\/revisions"}],"predecessor-version":[{"id":6452,"href":"https:\/\/www.wakilsahab.in\/lawyers\/wp-json\/wp\/v2\/posts\/6334\/revisions\/6452"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.wakilsahab.in\/lawyers\/wp-json\/wp\/v2\/media\/6335"}],"wp:attachment":[{"href":"https:\/\/www.wakilsahab.in\/lawyers\/wp-json\/wp\/v2\/media?parent=6334"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.wakilsahab.in\/lawyers\/wp-json\/wp\/v2\/categories?post=6334"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.wakilsahab.in\/lawyers\/wp-json\/wp\/v2\/tags?post=6334"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}