Historique des pannes de Okta

1. 3 septembre 2026 à 10:42 UTCresolved

   L'équipe du génie continue d'étudier la question de l'erreur 500 dans l'OP1. Pendant ce temps, les utilisateurs peuvent rencontrer des problèmes intermittents au sein de notre système. Nous fournirons une autre mise à jour dans les 30 prochaines minutes, ou plus tôt si des informations supplémentaires deviennent disponibles.

2. 3 septembre 2026 à 11:02 UTCresolved

   L'équipe du génie continue d'étudier la question de l'erreur 500 dans l'OP1. Pendant ce temps, les utilisateurs peuvent rencontrer des problèmes intermittents au sein de notre système. Nous fournirons une autre mise à jour dans les 30 prochaines minutes, ou plus tôt si des informations supplémentaires deviennent disponibles.

3. 3 septembre 2026 à 11:34 UTCresolved

   L'équipe du génie continue d'étudier la question de l'erreur 500 dans l'OP1. Pendant ce temps, les utilisateurs peuvent rencontrer des problèmes intermittents au sein de notre système. Nous fournirons une autre mise à jour dans les 30 prochaines minutes, ou plus tôt si des informations supplémentaires deviennent disponibles.

4. 3 septembre 2026 à 12:00 UTCresolved

   L'équipe du génie continue d'étudier la question de l'erreur 500 dans l'OP1. Pendant ce temps, les utilisateurs peuvent rencontrer des problèmes intermittents au sein de notre système. Nous fournirons une autre mise à jour dans les 30 prochaines minutes, ou plus tôt si des informations supplémentaires deviennent disponibles.

5. 3 septembre 2026 à 12:30 UTCresolved

   L'équipe du génie continue d'étudier la question de l'erreur 500 dans l'OP1. Pendant ce temps, les utilisateurs peuvent rencontrer des problèmes intermittents au sein de notre système. Nous fournirons une autre mise à jour dans les 30 prochaines minutes, ou plus tôt si des informations supplémentaires deviennent disponibles.

6. 3 septembre 2026 à 13:03 UTCresolved

   Okta Engineering a pu prendre les mesures nécessaires pour atténuer le problème et surveille actuellement la stabilité. La cause du problème et les détails sur les mesures prises seront affichés sur la page Trust.

7. 3 septembre 2026 à 14:09 UTCresolved

   Après un suivi prolongé, nous avons confirmé que la question n'avait pas repris. Une analyse des causes profondes sera affichée ici dans les cinq jours ouvrables. Nous nous excusons sincèrement de l'impact que cela a eu sur vous et vos utilisateurs.

Traduit automatiquement depuis la mise à jour officielle de l'incident.

1. 14 août 2026 à 15:23 UTCresolved

   Okta Engineering enquête et a déterminé que l'incident affecte actuellement la livraison de courriels à un sous-ensemble de clients. Pendant ce temps, les clients peuvent subir des reports de livraison par courriel. Nous fournirons une mise à jour dans 30 minutes, ou plus tôt si des informations supplémentaires deviennent disponibles.

2. 14 août 2026 à 15:39 UTCresolved

   Okta Engineering a déterminé que certains fournisseurs tiers n'avaient pas reçu de courriels de Okta. Notre surveillance montre que les taux de livraison se rétablissent dans des conditions normales et continuera de surveiller jusqu'à une résolution complète. Notre prochaine mise à jour sera dans 30 minutes ou plus tôt si des informations supplémentaires sont disponibles.

3. 14 août 2026 à 16:50 UTCresolved

   Our monitoring indicates that email deferrals to the impacted third-party email service have returned to the expected baseline. Additional information will be available within 5 Business days.

4. 21 août 2026 à 16:04 UTCresolved

   We sincerely apologize for any impact this incident has caused to you, your business, or your customers. At Okta, trust and transparency are our top priorities. Outlined below is the RCA summary for a recent incident where a third-party provider or downstream service experienced an issue that impacted the Okta service. We are committed to implementing improvements to the service to prevent similar occurrences. Detection and Impact: Okta detected an increase in customer support tickets regarding email delivery issues. Customers using custom email domains routed through Proofpoint experienced email delivery deferrals. During this period, affected end users were unable to receive system-generated emails, including One-Time Passwords (OTPs), account activation emails, password reset emails, factor enrollment emails, and other automated notifications. Root Cause Summary: Third-party provider Proofpoint experienced a service disruption, causing issues with email deliverability. When Okta services attempted to send emails to recipient domains whose MX records pointed to Proofpoint's infrastructure, the delivery attempts failed due to connection errors, resulting in deferred delivery. Okta's email sending infrastructure operated normally throughout the event. Remediation Steps: Okta Engineering monitored service telemetry and confirmed that email delivery returned to normal pre-incident baselines once Proofpoint restored stability to its infrastructure. Preventive Actions: Okta is defining automated alerting thresholds for email-bound and deferral rates to decrease detection time and updating runbooks to allow faster diagnosis of downstream provider disruptions. Timing: Incident start: August 14, 2026, at 5:00 AM PDT Incident resolved: August 14, 2026, at 10:04 AM PDT. Duration (# of minutes): 304 minutes Based on details provided by https://www.proofpoint.com/us/support-services

Traduit automatiquement depuis la mise à jour officielle de l'incident.

1. 24 juillet 2026 à 13:55 UTCresolved

   Après une période de stabilité et de surveillance, nous pouvons maintenant confirmer que la question des services a été entièrement résolue. Des renseignements supplémentaires sur les causes profondes seront disponibles dans les cinq jours ouvrables.

2. 1 août 2026 à 03:28 UTCresolved

   Nous nous excusons sincèrement de tout impact que cet incident a causé pour vous, votre entreprise et vos clients. À Okta, la confiance et la transparence sont nos principales priorités. Les faits concernant cet incident sont décrits ci-dessous. Nous nous engageons à apporter des améliorations à nos services afin de prévenir les événements futurs. Détection et impact Le 24 juillet 2026, vers 3h55 HAP, Okta a observé une dégradation du service pour OK7, OK12, OK14 et OP3. L'incident a eu des répercussions particulières sur les clients utilisant l'authentification AD/LDAP déléguée et les clients OAuth token-endpoint situés dans la région AWS-West-2. Tout au long de l'événement, les utilisateurs ont rencontré des erreurs d'accès aux ressources, des défaillances d'authentification, des délais de connexion des agents, des latences et des requêtes de jeton OAuth. Résumé des causes profondes La cause principale provient d'erreurs au sein d'un fournisseur de services cloud pour la région touchée. Étapes d'assainissement Le fournisseur de services cloud a récupéré la connectivité, et le service Okta s'est stabilisé par environ 4:10 PDT, et nos équipes ont confirmé un retour complet aux opérations de base à 4:19 PDT. Mesures préventives Okta examine et exécute activement plusieurs initiatives visant à prévenir les événements futurs. Ces mesures comprennent : la vérification et l'amélioration de nos processus d'intervention en cas d'incidents dans une région élargie, et l'amélioration de notre surveillance et de notre alerte en fonction de cet incident. Durée totale Durée totale (minutes) : 15 minutes Heure de début : 3 h 55 HAP Heure de fin : 4 h 10 HAP Durée (nombre de minutes) : 15

Traduit automatiquement depuis la mise à jour officielle de l'incident.

1. 22 juillet 2026 à 15:14 UTCresolved

   Veuillez noter que nos travaux de maintenance sont toujours en cours et que les équipes Okta Engineering sont activement engagées. Pendant cette fenêtre étendue, les clients peuvent éprouver des erreurs lors de la connexion à Okta Workflows. Nous apprécions votre patience car nos équipes travaillent pour terminer cette opération aussi rapidement et en toute sécurité que possible.   Pour comprendre comment cela pourrait avoir une incidence sur votre organisation et pour voir les solutions recommandées, veuillez consulter notre article de soutien détaillé: https://support.okta.com/help/s/article/okta-workflows-cell-migration-expected-comportements-and-workarounds?language=fr US . Nous fournirons d'autres mises à jour dans 30 minutes ou plus tôt, à mesure que de plus amples renseignements seront disponibles.

2. 22 juillet 2026 à 15:49 UTCresolved

   Les activités d'entretien sont toujours en cours, car nous nous assurons que la migration est terminée en toute sécurité et avec succès.   Nous continuerons à partager les mises à jour en 30 minutes ou plus tôt, à mesure qu'elles seront disponibles. Dans l'intervalle, veuillez consulter l'article de support ci-dessous pour les derniers conseils sur les solutions de rechange et les comportements attendus: https://support.okta.com/help/s/article/okta-workflows-cell-migration-expectative-comportements-and-workarounds?language=fr US

3. 22 juillet 2026 à 16:19 UTCresolved

   Nous observons maintenant des performances stables du système à la suite des activités de migration. Nos équipes d'ingénierie restent actives dans le suivi de tous les services. Les flux de travail de base se rétablissent, et les clients devraient voir des améliorations continues. Nous fournirons notre prochaine mise à jour dans 30 minutes, ou plus tôt si de nouvelles informations deviennent disponibles.

4. 22 juillet 2026 à 16:59 UTCresolved

   Après une période de stabilité et de surveillance, nous pouvons maintenant confirmer que la question des services a été entièrement résolue. Des renseignements supplémentaires sur les causes profondes seront disponibles dans les cinq jours ouvrables.

5. 29 juillet 2026 à 23:34 UTCresolved

   Nous nous excusons sincèrement de tout impact que cet incident a causé pour vous, votre entreprise et vos clients. À Okta, la confiance et la transparence sont nos principales priorités. Les faits concernant cet incident sont décrits ci-dessous. Nous nous engageons à apporter des améliorations à nos services afin de prévenir les événements futurs. Nous reconnaissons que cette maintenance a dépassé sa fenêtre initialement communiquée et que nos mises à jour d'état pendant cette extension n'ont pas tenu les clients informés en temps opportun. Nous nous excusons pour cela; l'améliorer fait partie de nos mesures préventives ci-dessous. Détection et impact Le 22 juillet 2026, à 5h15 HAP, Okta Workflows dans notre région de l'Amérique du Nord (cellule FL1) a commencé à éprouver des problèmes qui ont affecté les utilisateurs de Workflows dans les cellules 1, 2, 3, 4, 6, 7 et 11. Pendant ce temps, certaines exécutions de workflow ont échoué ou se sont terminées prématurément, et les utilisateurs ont rencontré des erreurs en chargeant la console Workflows. De plus, certaines connexions de l'application ont cessé de fonctionner, exigeant des utilisateurs d'authentifier manuellement pour restaurer la fonctionnalité. Résumé des causes profondes Le problème s'est posé lors d'une mise à niveau prévue du système. Un défaut dans le logiciel d'un fournisseur tiers a fait que la mise à niveau a fonctionné beaucoup plus longtemps que prévu. Pendant la migration vers un nouveau système de datastore, une défaillance de communication entre les composants du système a empêché le service de revenir à son état normal dans la fenêtre prévue. Pendant cette période, la console Workflows ne s'est pas dégradée gracieusement lorsque le stockage de données sous-jacent n'était pas disponible; au lieu de cela, elle a fait apparaître des erreurs directement aux utilisateurs. De plus, les connexions d'application dont les identifiants n'ont pas pu être rafraîchis pendant la perturbation ont été laissées dans un état défaillant et n'ont pas été automatiquement rafraîchies, c'est pourquoi certaines erreurs se sont poursuivies après la mise à niveau terminée, jusqu'à ce que la connexion touchée ait été réautorisée. Étapes d'assainissement Le fournisseur tiers a confirmé que le cluster était stabilisé à 8h34 HAP. Après stabilisation, Okta Engineering a récupéré les services affectés pour effacer les caches de configuration. Tous les clients qui ont subi des erreurs de rafraîchissement du jeton OAuth pendant cette période ont dû réautoriser manuellement toutes les connexions d'applications. La fonctionnalité complète a été restaurée par 9:33 PDT. Si vous êtes toujours affecté: réautorisez manuellement les identifiants de connexion affectés dans la console Workflows (voir KB: Okta Workflows: Cell Migration Attendue Behaviors and Workarounds). Les exécutions en flux de travail qui ont échoué pendant la fenêtre peuvent être ré-exécutées en utilisant votre processus normal; contactez Okta Support si vous souhaitez aider à identifier les exécutions affectées. Mesures préventives Pour empêcher que des incidents similaires se produisent, Okta a pris les mesures suivantes. Nous avons imposé une retenue à l'échelle de la flotte sur toutes les autres migrations de moteurs de datastore jusqu'à ce que notre fournisseur tiers déploie un correctif pour le défaut confirmé et confirme qu'il est sûr de procéder. Nous mettons également à jour la console Workflows pour échouer rapidement et récupérer automatiquement lorsque l'infrastructure est temporairement indisponible, assurant aux utilisateurs de voir un état de connexion plutôt que des erreurs système brutes. Parallèlement, nous ajoutons une surveillance et une alerte automatisées directes pour détecter les défaillances de la convergence des grappes afin d'accélérer la détection pendant les fenêtres de maintenance. Enfin, nous améliorons la façon dont nous communiquons pendant la maintenance qui dure longtemps, y compris une page d'état plus opportun et des mises à jour de progrès/achèvement. Pour toute question sur la fiabilité ou la feuille de route de Workflows, votre équipe de compte Okta peut vous aider. Durée totale Durée totale (minutes) : 258 minutes (4 heures et 18 minutes) Heure de début : 5 h 15 PDT Heure de fin : 9 h 33 PDT Durée (nombre de minutes) : 258 minutes

Traduit automatiquement depuis la mise à jour officielle de l'incident.

1. 13 juillet 2026 à 21:23 UTCresolved

   Notre équipe d'ingénierie a appliqué une correction potentielle à l'environnement d'aperçu impacté et surveille activement le système de récupération. Nous fournirons une autre mise à jour dans les 30 minutes, ou plus tôt si de nouvelles informations deviennent disponibles. Nous nous excusons sincèrement pour tout désagrément.

2. 13 juillet 2026 à 21:39 UTCresolved

   Notre équipe d'ingénierie a appliqué un correctif, et nous observons la restauration de service à notre environnement de prévisualisation OP2. Nous suivons de près le système et fournirons une autre mise à jour dans les 30 minutes.

3. 13 juillet 2026 à 21:53 UTCresolved

   Nous avons confirmé que l'environnement de prévisualisation OP2 s'est complètement stabilisé, et le problème antérieur est complètement résolu. Des renseignements supplémentaires sur les causes profondes seront disponibles dans les cinq jours ouvrables.

4. 21 juillet 2026 à 01:57 UTCresolved

   Nous nous excusons sincèrement de tout impact que cet incident a causé pour vous, votre entreprise et vos clients. À Okta, la confiance et la transparence sont nos principales priorités. Les faits concernant cet incident sont décrits ci-dessous. Nous nous engageons à apporter des améliorations au service afin de prévenir les incidents futurs. Détection et impact: Le 13 juillet, 2026 à 1:01 PM PT, Okta surveillance interne a alerté notre équipe à des erreurs élevées dans l'environnement EMEA Preview Cell. Au cours de cette période, tous les clients de l'environnement EMEA Preview Cell ont été touchés, subissant des défaillances d'authentification et la lenteur du service. Les administrateurs et les utilisateurs finaux peuvent avoir rencontré des messages « Erreur serveur interne » et d'autres erreurs lors de la tentative de se connecter ou d'accéder aux services Okta. Résumé des causes profondes : L'incident a été déclenché lors d'une mise à niveau du service régulier. Une erreur de configuration a provoqué la réception par inadvertance d'une version incorrecte d'un composant de routage de l'infrastructure critique par le cluster EMEA Preview Cell. Cette mise à jour a rompu le routage du réseau interne, ce qui a empêché la communication entre les composantes d'infrastructure autrement saines. Comme les demandes n'ont pas pu parvenir à l'application centrale, elles ont entraîné des erreurs de service généralisées et des défaillances d'authentification. Étapes d'assainissement: L'équipe d'ingénierie d'Okta a identifié le problème et a immédiatement commencé à déployer la version stable précédente de la configuration de routage. Après le renversement, les ingénieurs redémarrent systématiquement les services concernés (y compris les passerelles et les proxies) pour restaurer la connectivité interne. Ces mesures ont permis de régler le problème avec succès, et l'impact sur les clients a été entièrement atténué par 14 h 20 PG. Mesures préventives : Nous perfectionnons nos processus opérationnels afin d'améliorer la stabilité des services. Cela comprend des travaux continus visant à améliorer les capacités de détection et de surveillance de l'automobile, ce qui nous aide à cerner et à régler plus efficacement les problèmes potentiels. Durée totale (minutes): 79 minutes (1 heure 19 minutes) Heure de début: 1:01 PM PT Heure de fin: 2:20 PM PT

Traduit automatiquement depuis la mise à jour officielle de l'incident.

1. 7 juillet 2026 à 20:50 UTCresolved

   Okta ingénierie a identifié un problème affectant Okta Workflows et prennent actuellement des mesures pour remédier. Les clients peuvent voir 500 erreurs ou délais dans la console Okta Workflows, les paramètres d'API et les webhooks entrants. Nous fournirons une autre mise à jour dans les 30 prochaines minutes, ou plus tôt si des informations supplémentaires deviennent disponibles.

2. 7 juillet 2026 à 21:03 UTCresolved

   Le problème avec Okta Workflows a été résolu. Tous les flux de travail sont maintenant traités normalement. Une analyse des causes profondes sera affichée ici dans les cinq jours ouvrables. Nous nous excusons de tout inconvénient que cela pourrait avoir causé.

3. 15 juillet 2026 à 15:22 UTCresolved

   Nous nous excusons sincèrement pour tout impact que cet incident pourrait avoir causé pour vous, votre entreprise ou vos clients. À Okta, la confiance et la transparence sont nos principales priorités. Les faits concernant cet incident sont décrits ci-dessous. Nous nous engageons à apporter des améliorations au service afin de prévenir les incidents futurs. Détection et impact: Le 7 juillet 2026, à 1:01 PDT, Okta surveillance interne et rapports clients a alerté notre équipe aux problèmes avec les utilisateurs accédant Workflows sur la cellule Okta OK14 en Amérique du Nord. Pendant cette période, les clients de FL14 ont peut-être subi des erreurs HTTP 500, 503, ou 504 Gateway Timeout lors de la navigation vers Workflows depuis le portail admin ou en essayant de se connecter à la console Workflows via SSO. Les séances existantes peuvent avoir été interrompues par erreur. Les workflows programmés et à la demande, les hooks en ligne et les webhooks peuvent avoir connu des erreurs et des délais. De plus, les exécutions en vol ont connu des retards. Résumé des causes profondes : Cet incident s'est produit au cours du déploiement prévu de la libération 2027.07.0 dans la cellule FL14. Au cours du processus de déploiement, un module de service chargé d'exécuter les mises à jour du schéma a été mis fin prématurément. Cette terminaison précoce a conduit à des connexions inattendues et de longue durée, ce qui a finalement causé l'épuisement des connexions sur la base de données primaire de la cellule. Étapes d'assainissement : Après avoir reçu des alertes et des rapports clients, l'équipe a commencé à diagnostiquer à 12h44 PDT. D'ici 13h30, ils ont isolé et retiré le module de service effectuant des mises à jour du schéma. Cela a permis aux connexions de la base de données d'effacer, de restaurer le service complet d'ici 13h45 PDT. Mesures préventives : Afin de prévenir les événements futurs, Okta s'engage à renforcer la stabilité de notre infrastructure et à renforcer nos capacités de surveillance. Nous mettons actuellement en oeuvre des améliorations stratégiques à nos processus de déploiement et systèmes d'alerte afin d'assurer une détection plus rapide, une réponse et une plus grande résilience dans l'ensemble de nos services. Durée totale : 67 minutes

Traduit automatiquement depuis la mise à jour officielle de l'incident.

1. 26 juin 2026 à 15:22 UTCresolved

   Our Engineering team has identified the issue, and we will continue monitoring the issue. We will provide another update within the next 15 minutes, or sooner if additional information becomes available.

2. 29 juin 2026 à 23:50 UTCresolved

3. 30 juin 2026 à 15:01 UTCresolved

   The Root Cause Analysis (RCA) details previously posted in this update were published in error and belonged to a separate incident that occurred on June 22nd (I-10858). We sincerely apologize for any confusion this may have caused. Our formal RCA process for this OK11 custom domains incident (I-10874) is still actively underway. We remain committed to transparency and will provide an accurate, dedicated RCA for this disruption within 5 business days (expected late July 6, 2026). Thank you for your patience.

4. 6 juillet 2026 à 22:01 UTCresolved

   We sincerely apologize for any impact this incident has caused to you, your business, and your customers. At Okta, trust and transparency are our top priorities. Outlined below are the facts regarding this incident. We are committed to implementing improvements to the service to prevent future occurrences of this incident. Detection and Impact: On June 26th at 7:43 AM (PT), Okta was alerted to customer reports of connection issues and authentication failures when accessing custom domains. The incident was isolated to OK11 and normal connectivity was restored at 7:53 AM (PT). Root Cause Summary: Okta Engineering identified that the root cause of the issue originated during a custom domain traffic migration in OK11 on June 25th at 10:47 AM (PT), the new routing infrastructure became resource-constrained under increased traffic volume the next day. This resulted in TCP connection resets and errors sent to clients attempting to access custom domains. Remediation Steps: Once the source of the issue was identified, Okta engineering immediately mitigated the issue by reverting custom domain traffic back to the previous infrastructure, restoring normal connectivity within minutes. Preventative Actions: To prevent future incidents, the routing infrastructure was scaled to prevent recurrence. Okta will work to implement enhanced monitoring and alerting, as well as updated auto-scaling policies to ensure routing infrastructure maintains adequate capacity during traffic fluctuations and future infrastructure improvements. Duration (# of minutes): Total Duration (Minutes): 53 minutes Actual Time: 7:00 AM - 7:53 AM PT

1. 22 juin 2026 à 22:02 UTCresolved

   The Okta Engineering team has taken steps to mitigate the issue. We have observed that the errors have been reduced and is continuing to subside. We are monitoring the situation closely and will provide further updates in 15-30 minutes as we move toward full mitigation.

2. 22 juin 2026 à 22:33 UTCresolved

   Internal monitoring confirmed that the EU1 environment has fully recovered and is operating normally. A full Root Cause Analysis (RCA) will be made available within 5 business days. We apologize for any disruption this may have caused.

3. 30 juin 2026 à 14:43 UTCresolved

   We sincerely apologize for any impact this incident has caused to you, your business, and your customers. At Okta, trust and transparency are our top priorities. Outlined below are the facts regarding this incident. We are committed to implementing improvements to the service to prevent future occurrences of this incident. Detection and Impact: On June 22nd, 2026, at 2:08 PM PDT, Okta internal monitoring alerted our team to errors in the EU environment. During this period, administrators and end users may have experienced intermittent 500 Internal Server Error and 429 Rate Limit responses when attempting to access Okta services. Root Cause Summary: The incident was triggered by an unexpected scaling down of the application cluster during scheduled maintenance of the environment. This change accidentally overwrote the custom capacity settings for several of our applications. This caused the system to automatically revert to its lowest default capacity, drastically reducing the amount of user traffic it could handle simultaneously. As a result, the remaining servers were immediately overwhelmed by the normal volume of user requests, leading to system slowdowns, errors, and "too many requests" blockages for our customers in the impacted cell. Remediation Steps: Okta's engineering team identified the root cause issue at 2:44 PM PDT. The team mitigated the issue by scaling up resource capacity, completely stabilizing service health back to normal levels by 2:48 PM PDT. Preventative Actions: In order to prevent similar incidents from happening again, Okta is currently reviewing the following: - Strengthen guardrails by improving our validation tools and multi-environment review processes to automatically detect and flag unexpected server capacity changes before they're applied. - Accelerate system monitoring by implementing enhanced monitoring thresholds to instantly detect and report infrastructure capacity drops before they can impact end users. - Optimize failover protocols by auditing critical alert paths for maximum reliability while establishing a clear, structured communication plan for when automated backup systems are triggered. These learnings work to prevent similar incidents from happening again. Duration (# of minutes):40 minutes

1. 2 juin 2026 à 11:21 UTCresolved

   Engineering is still investigating the issue, recent actions that were taken show errors subsiding.

2. 2 juin 2026 à 11:40 UTCresolved

   The Okta Engineering team has observed that the errors have been reduced. We are monitoring the situation closely and will provide further updates as we move toward full mitigation.

3. 2 juin 2026 à 12:07 UTCresolved

   Okta Engineering has confirmed that the issue has been resolved. An RCA will be provided in five business days. We apologize for any inconvenience this may have caused.

4. 10 juin 2026 à 02:19 UTCresolved

   We sincerely apologize for any impact this incident has caused to you, your business, and your customers. At Okta trust and transparency are our top priorities. Outlined below are the facts regarding this incident. We are committed to implementing improvements to the service to prevent future occurrences of this incident. Detection and Impact: On June 2nd, 2026, at 2:36 AM PT, Okta internal monitoring alerted our team to elevated errors in the OK9 environment. During this period, administrators and end users may have experienced intermittent 500 Internal Server errors when trying to access Okta services. Additionally, Admins may have encountered error messages when fetching certain records. Root Cause Summary: The incident was triggered by a misconfiguration introduced during a scheduled maintenance of the environment. This maintenance introduced an error that caused connectivity issues within specific infrastructure components. Only a small percentage of the infrastructure was affected during this period. Remediation Steps: Okta's engineering team identified and isolated impacted infrastructure at 4:01AM PT. These actions successfully addressed the issue, and normal operations were restored by 4:17 AM PT. Preventative Actions: In order to prevent similar incidents from happening again, Okta is currently reviewing the following: Incident response processes, procedures, and investigation tools to more rapidly diagnose similar issues in the future. Risk evaluation processes to include expanding error types and adding additional appropriate corresponding alerts and monitoring associated with vulnerability updates. These learnings work to prevent similar incidents from happening again and, in cases like these, allow Okta to respond faster when possible. Duration (# of minutes): 100

1. 1 mai 2026 à 00:28 UTCresolved

   The Okta Engineering team has successfully resolved the service instability that was causing system timeouts. Okta engineering has confirmed that errors have subsided, and we have transitioned into a monitoring phase to ensure continued stability. We sincerely appreciate your patience while our team worked to identify and resolve the issue.

2. 1 mai 2026 à 01:01 UTCresolved

   Internal monitoring confirmed that the OP1 environment has fully recovered and is operating normally. A full Root Cause Analysis (RCA) will be made available within 5 days. We apologize for any disruption this may have caused.

3. 8 mai 2026 à 02:51 UTCresolved

   We sincerely apologize for any impact this incident has caused to you, your business, and your customers. At Okta, trust and transparency are our top priorities. Outlined below are the facts regarding this incident. We are committed to implementing improvements to the service to prevent future occurrences of this incident. Detection and Impact: On April 30th, 2026, at 10:57 AM PT, Okta internal monitoring alerted our team of login issues for users accessing Okta Preview Cell 1. During this period, customers may have received 500 errors for query timeouts and were unable to log in. Root Cause Summary: The incident stemmed from a database query change. The updated query intermittently exceeded its timeout limit across several login endpoints. Consequently, users experienced 500 errors and were unable to log in. Remediation Steps: Upon receiving the alerts, Okta Engineering immediately launched an investigation. By 1:02 PM PT, the team took initial mitigation steps to resolve the issue, however these steps did not resolve the issue. The incident response team continued to triage and diagnose the issue and identified the root cause related to the updated database query change at 4:50 PM PT. The team reverted the new query behavior at 5:00 PM PT in Preview Cell 1, and full service functionality was restored by 5:05 PM PT. Preventative Actions: In order to prevent similar incidents from happening again, Okta is currently reviewing the following: -Reviewing incident response processes and investigation tools to more rapidly diagnose similar issues in the future -Risk evaluation processes associated with query updates -Alerts and monitors associated with the database proxy These learnings work to prevent similar incidents from happening again and, in cases like these, allow Okta to respond faster when possible. Duration (# of minutes): 375

1. 29 avril 2026 à 22:42 UTCresolved

   We sincerely apologize for any impact this incident has caused to you, your business, or your customers. At Okta, trust and transparency are our top priorities. Outlined below is the RCA summary for a recent incident where a third-party provider or downstream service experienced an issue that impacted the Okta service. We are committed to implementing improvements to the service to prevent similar occurrences. Detection and Impact: Customers using the third-party service Kolide began reporting that attempts to sign in to their orgs redirected to an unrelated tenant and received an HTTP 400 error with the message “Login Failed - 400: Bad Request Error Code: GENERAL_NONSUCCESS” Root Cause Summary: Kolide has confirmed for Okta that an incident occurred with their Device Trust SAML authentication system, resulting in some users being redirected to an incorrect service provider after completing authentication. This resulted in users receiving a 400 error. Kolide reported that users were unable to access their intended application, any unauthorized applications, or any unauthorized environments. Remediation Steps: Okta Support advised customers that they could temporarily disable their Kolide integrations to allow their users to successfully access their services. Timing: Kolide reports: Incident start: April 22, 2026, at 8:24 AM PDT Incident resolved: April 22, 2026, at 8:30 AM PDT Duration (# of minutes): 6 Minutes

1. 13 avril 2026 à 20:33 UTCresolved

   Okta has identified the issue impacting Access Request job processing and Engineering Teams are working to restore the service. We are rolling out the fix for this issue in 30 minutes and will provide another update once the fix is rolled out.

2. 13 avril 2026 à 20:59 UTCresolved

   Okta has identified the root cause of the issue affecting Access Request job processing as of 1:00 PM PT. A resolution was implemented at 1:45 PM PT, and the service is now fully operational for new submissions. The engineering team is currently working to restore and process any requests submitted during the disruption period. We will provide another update within the next 30 minutes, or sooner if additional information becomes available.

3. 13 avril 2026 à 21:29 UTCresolved

   While new Access Request job processing has been restored, Okta is also working to resolve the issue affecting in-flight Access Request job requests processed during the disruption period.  The fix for this issue is expected to be implemented around 5:00pm PT.  Okta will provide another update at 5:00pm PT or sooner if one becomes available.

4. 14 avril 2026 à 00:25 UTCresolved

   The issue with in-flight Access Request job requests processed during the disruption period has also been resolved. All Access Request jobs are now processing normally. A root cause analysis (RCA) will be posted here within five business days. We apologize for any inconvenience this may have caused.

5. 21 avril 2026 à 23:16 UTCresolved

   We sincerely apologize for any impact this incident has caused to you, your business, or your customers. At Okta, trust and transparency are our top priorities. Outlined below are the facts regarding this incident. We are committed to implementing improvements to the service to prevent future occurrences of this kind. Detection and Impact: On April 13, 2026, at 11:51 AM PST, an incident was reported where Okta Identity Governance (OIG) access requests stalled in the "In Progress" state indefinitely across all environments. The system accepted initial submissions and first-level approvals, but failed to trigger subsequent automated tasks like assigning users to apps/groups or running delegated workflows. . Root Cause Summary: During a recent update deployment, new tasks inadvertently moved to a 'LOCKED' status without implementing the necessary logic to transition them back to 'OPEN.' This caused the tasks to become permanently hidden, preventing the automated execution of the action associated with them. Remediation Steps: Okta's engineering teams promptly responded, initiating immediate detection and diagnosis. Corrective measures, including a code fix deployed to all cells, were implemented to prevent any new access requests from becoming stuck. A remediation migration successfully transitioned most affected requests to a fixed state. Normal functionality was fully restored by April 13, 2026, at 5:18 PM PST, following the completion of a final migration to unblock remaining requests.On April 14, 2026, Okta took further steps to resolve in-flight requests that were stuck, moving them to a final, terminal state. Preventative Actions: Okta is taking several steps to prevent this issue from recurring. These include enhanced testing, deployment of safeguards, and improvements to our monitoring and alerting systems. Duration (# of minutes): 434

1. 24 mars 2026 à 23:32 UTCresolved

   We sincerely apologize for any impact this incident has caused to you, your business, or your customers. At Okta, trust and transparency are our top priorities. Outlined below is the RCA summary for a recent incident where a third-party provider or downstream service experienced an issue that impacted the Okta service. We are committed to implementing improvements to the service to prevent similar occurrences. Detection and Impact: Okta internal monitoring alerted our engineering team to disruptions to the Okta status page and the commercial support portal. Okta Engineering identified a service disruption with our third-party provider and confirmed the impact. During the disruption, customer admins were unable to access the Okta Support Center (support.okta.com) to log support cases or view the primary status page (status.okta.com). Root Cause Summary: Okta's third-party provider, responsible for hosting the Okta Support Center and status page, experienced a service disruption. The third-party provider identified a version mismatch between their application and database layers as the technical trigger. This mismatch resulted in internal server errors and login failures, rendering the Okta Support Center and status page inaccessible for a portion of their disruption. Remediation Steps: Okta teams posted an admin banner within the Okta Admin Console to provide visibility into the disruption and offer a phone number for customers who needed to log support cases manually while the Okta Support Center was unavailable. Services were fully restored after the third-party provider successfully deployed a fix-forward release to the impacted infrastructure, which restored login access. Preventive Actions: As part of Okta's commitment to operational resilience, we have optimized our vendor engagement matrix to accelerate incident mitigation and downstream coordination. We are also conducting an expanded audit of external dependencies to formalize alternative response procedures within our disaster recovery plans. Simultaneously, Engineering has enhanced real-time monitoring and alerting for service endpoints, enabling faster detection and resolution. Timing: Incident start: March 16, 2026, at 08:58 PM PDT Incident resolved: March 16, 2026, at 11:34 PM PDT Duration (# of minutes): 156 minutes

1. 13 mars 2026 à 18:48 UTCresolved

   At around 11:20AM PT, Okta Engineering has confirmed that the issue has been resolved in both cells, EU1 and OK1. An RCA will be provided in five business days. We apologize for any inconvenience this may have caused.

2. 20 mars 2026 à 20:35 UTCresolved

   We sincerely apologize for any impact this incident has caused to you, your business, or your customers. At Okta, trust and transparency are our top priorities. Outlined below are the facts regarding this incident. We are committed to implementing improvements to the service to prevent future occurrences of this kind. Detection and Impact On March 13, 2026, at 10:49 AM PT, Okta detected a service degradation that prevented some customers on the OK1 and EU1 cells from performing directory operations, viewing or searching for users, groups, and devices in the Okta Admin Console. Customers may have experienced "Invalid search query" messages or 5xx errors. This issue also affected several API endpoints related to users, groups, and devices, which may have impacted customer integrations relying on this functionality. End-user authentication was not impacted. Root Cause Summary The root cause of this incident was a planned configuration change that exposed a bug in the underlying service, resulting in system errors and disrupting both directory-based APIs. Remediation Steps Upon detection, Okta's engineering teams initiated incident response procedures. Once the problematic configuration change was identified as the source of the errors, it was immediately reverted. After the reversion was applied and verified, full functionality was restored to all impacted customers. Preventative Actions Okta is taking steps to prevent future occurrences of this kind, including enhancing our testing processes, hardening and improving our configuration rollout strategy, and upgrading our monitoring and alerting systems for enhanced detection. Start Time: 03/13/2026 10:17 AM PT End Time: 03/13/2026 11:21 AM PT Duration (# of minutes): 64

1. 4 février 2026 à 12:37 UTCresolved

   Okta has implemented a fix for the LDAP issue affecting the OK9 environment. We are observing signs of service recovery and confirmed improvements from customers. We will continue to monitor the situation closely to ensure full resolution. We are monitoring the situation closely and will provide further updates within the next 30 minutes.

2. 4 février 2026 à 12:50 UTCresolved

   This incident has been resolved. Internal monitoring and customer reports confirm that the LDAPi system in OK9 has fully recovered and is operating normally. We will continue to closely monitor the service to ensure stability. A full Root Cause Analysis (RCA) will be made available within 5 days. We apologize for any disruption this may have caused.

3. 11 février 2026 à 18:45 UTCresolved

   On February 4th, 2026, at 3:33 AM PT, Okta received customer reports of authentication failures within the OK9 (EMEA) cell. Customers attempting to access LDAP interface services encountered certificate validation errors related to hostname mismatches, preventing successful authentication. Root Cause Summary: The incident was triggered by a misconfiguration introduced during a scheduled update of our LDAP interface to a new configuration source on February 4th 2:24 AM PT. This update inadvertently applied an incorrect SSL certificate profile to the LDAP interface within the OK9 environment. This misconfiguration caused certificate validity errors, resulting in connection issues when attempting to use LDAP services. Remediation Steps: Once the source of the issue was identified, Okta engineering team rolled back changes at 4:07 AM PT. These actions successfully addressed the issue and restored normal operations by 4:26 AM PT. Preventative Actions: To prevent future incidents, Okta has updated synthetic tests and enhanced monitoring. Teams also plan to add new tooling to improve changes to the LDAP service certificates. In addition, we are reviewing our internal processes and procedures, refining our automated rollback processes to further reduce the customer impact time for similar issues in the future. Duration (# of minutes): Total Duration (Minutes): 122 minutes Actual Time: 02:24 AM PT - 04:26 AM PT

1. 28 janvier 2026 à 19:24 UTCresolved

   We are seeing significant and steady system stability as our third-party provider recovery continues in the EU-West-1 region. We will continue monitoring the service to ensure a full recovery. We will provide another update in 60 minutes or as soon as more information becomes available.

2. 28 janvier 2026 à 19:36 UTCresolved

   The third-party provider has confirmed that the issue has been resolved. A root cause analysis (RCA) will be posted here within five business days.

3. 5 février 2026 à 01:26 UTCresolved

   We sincerely apologize for any impact this incident has caused to you, your business, and your customers. At Okta, trust and transparency are our top priorities. Outlined below are the facts regarding this incident. We are committed to implementing improvements to the service to prevent future occurrences of this incident. Detection and Impact: On January 29th, 2026, at 8:28 AM PT, Okta internal monitoring alerted our team of intermittent connectivity issues accessing Okta Cell 9 (EMEA). During this period, customers may have experienced intermittent errors establishing network connections and latency when trying to access Okta services. Clients configured to retry failed connections likely saw successful connectivity on secondary attempts. Root Cause Summary: The incident stemmed from a failure of a component that manages configuration changes to the networking infrastructure with our Cloud Service Provider (CSP) in the region hosting Okta Cell 9. Remediation Steps: Immediately upon receiving the alerts, Okta Engineering began investigating several isolated spikes but determined that cell health was stable. At 9:35 AM PT, Okta Engineering received more alerts and immediately initiated incident response procedures. At 9:51 AM PT, Okta Engineering began working with the CSP to diagnose the issue. At 10:13 AM PT, the latency began to stabilize. At 11:22 AM PT, the issue was resolved. Preventative Actions: To prevent similar incidents from recurring, Okta Engineering is improving our monitoring, working with our CSP to enhance their monitoring, and reviewing incident response processes associated with this event. Duration (# of minutes): 186

4. 11 février 2026 à 18:40 UTCresolved

1. 22 janvier 2026 à 21:46 UTCresolved

   We are continuing to monitor the third-party email service provider's progress toward a resolution. At this time, we have not received an estimated time for recovery (ETR). If you are attempting to sign in and have configured alternative MFA factors (such as Okta Verify, SMS, or Security Keys/Biometrics), we recommend using those options to avoid delays associated with email delivery. Our engineering teams remain on standby to verify service restoration the moment it begins. We will continue to provide updates as soon as the impacted provider shares new information.

2. 22 janvier 2026 à 23:24 UTCresolved

   We're continuing to track the ongoing service disruption with the third-party email provider. We appreciate your patience while we await the provider's resolution of the root cause. We'll post our next update in 90 minutes or as soon as we have a significant change in status.

3. 23 janvier 2026 à 00:46 UTCresolved

   The third-party email provider has identified the root cause as a service infrastructure issue within the North America region. They are currently implementing load-balancing configurations to stabilize the environment. While recovery is underway, users may continue to experience intermittent delays or failures in receiving emails. We are monitoring the provider's progress and will provide another update as they move toward full resolution.

4. 23 janvier 2026 à 03:55 UTCresolved

   The third-party provider continues to see improvements across their impacted North American infrastructure. While our internal systems remain healthy, our team is on standby to validate the restoration of all external dependencies. We will provide a final update once the provider confirms that their services have fully returned to normal operations.

5. 23 janvier 2026 à 06:46 UTCresolved

   The third-party provider has confirmed that a fix has been successfully deployed. A root cause analysis (RCA) will be posted here within ten business days. We sincerely apologize for the impact this had on you and your users.

6. 30 janvier 2026 à 22:34 UTCresolved

   Following is the RCA summary for a recent incident where a third-party provider or downstream service experienced an issue that impacted the Okta service Timing: Incident start: January 22, 2026, at 10:55 AM PST Incident resolved: January 22, 202,6 at 7:50 PM PST Detection: The issue was initially identified through customer reports that they were not receiving expected email notifications. Okta Engineering confirmed the service disruption after internal monitoring indicated a high volume of deferred email delivery attempts to specific third-party domains. Impact: Customers across all Okta cells with email addresses hosted by Microsoft Exchange Online may have experienced intermittent delays or failures in receiving automated email notifications, such as Multi-Factor Authentication (MFA) codes and enrollment links. Okta services remained healthy and attempted to resend deferred messages. Root Cause: Microsoft reports that it experienced a service issue within its North American data centers during a planned maintenance activity. As a result, emails sent from Okta to the affected provider resulted in “soft bounce” errors, indicating that the provider's servers were temporarily unable to accept the messages. Remediation Steps: To mitigate the impact on users attempting to sign in, Okta recommended using alternative MFA factors, such as Okta Verify, security keys, or SMS, to bypass delays associated with email delivery. Okta Engineering continuously monitored the third-party provider's recovery efforts and validated that mail delivery returned to baseline levels once the provider stabilized their environment.

1. 20 janvier 2026 à 16:04 UTCresolved

   We've identified that the service interruption is due to an issue with a third-party provider. We are actively working with the provider to ensure a swift resolution and are closely monitoring progress. We will provide further updates as more information becomes available.

2. 20 janvier 2026 à 16:35 UTCresolved

   We are continuing to monitor progress with our third-party provider. The provider is currently implementing mitigation steps, and we are seeing signs of recovery. Several customers have already confirmed that service has been restored on their end. We are working to ensure this resolution is consistent across all accounts. We will provide the next updates within 30 minutes, or sooner if additional information becomes available.

3. 20 janvier 2026 à 17:10 UTCresolved

   Our third-party provider is continuing their recovery efforts. While we are no longer seeing active errors, we remain in a period of close monitoring as they work toward a final resolution. We recognize that some users may still experience residual effects; our Support teams are working with them to resolve these errors. Our next update will be in 60 minutes or sooner if additional information becomes available.

4. 20 janvier 2026 à 17:21 UTCresolved

   Our third-party provider has confirmed that a fix has been successfully deployed. A root cause analysis (RCA) will be posted here within ten business days. We sincerely apologize for the impact this had on you and your users.

5. 26 janvier 2026 à 18:31 UTCresolved

   Following is the RCA summary for a recent incident where a third-party provider or downstream service experienced an issue that impacted the Okta service. Timing: Incident start: January 19, 2026, at 10:10 AM PST Incident resolved: January 20, 2026, at 11:49 AM PST Detection: Okta teams were notified via customer cases and an increase in Google Workspace import errors. Impact: Any customers performing Google Workspace imports during the incident window may have encountered an error stating “UserImportJob: Failed to download users. Json field=primaryEmail does not exist for object=[...]” Root Cause: Google Workspaces, a third-party provider, confirmed that a faulty code change introduced caused certain fields, such as primary email addresses, to be missing in API call results. As a result, Okta imports from Google Workspace could not complete, leading to import failures. Google provides additional details for this disruption: https://www.google.com/appsstatus/dashboard/incidents/9q5n4sfDnxYzHj8FxEua Remediation Steps: Okta teams soon identified that a provider-side disruption was the cause of the incident and opened a formal support case to track the resolution. The third-party provider resolved the incident by rolling back a faulty code change and restoring the missing metadata. Recovery occurred in stages, with some customers successfully remediated before others. Okta teams monitored the recovery in real-time and confirmed successful import signals across multiple impacted customers before standing down.

1. 3 janvier 2026 à 01:52 UTCresolved

   We sincerely apologize for any impact this incident has caused to you, your business, or your customers. At Okta, trust and transparency are our top priorities. Outlined below are the facts regarding this incident. We are committed to implementing improvements to the service to prevent future occurrences of this kind. Detection and Impact: On December 26, 2025, at 3:27 AM PST, Okta engineering was alerted to increased errors in US Cell 6. During this time, US Cell 6 experienced errors, connection timeouts, and slow responses while attempting to access resources. Admins would have seen a banner on the admin portal stating that the cell was in read-only mode. Root Cause Summary: Okta has determined that the issue was caused by an outage with our cloud service provider. All instances fully recovered by 5:43 AM PT. Remediation Steps: Upon detection, Okta's engineering teams immediately began diagnosing the issue and worked with our cloud service provider to resolve the issue. In addition, the Incident response teams continued to review and analyze the incident, applying additional measures to improve service availability for end-user authentications. Preventative Actions: Okta is taking steps to prevent a recurrence of this issue. We are making improvements to our monitoring and response procedures to more quickly diagnose and recover from this kind of failure in the future. Total Duration: 2 hours and 16 minutes Start Time: December 26, 2025, 03:27 AM PT End Time: December 26, 2025, 05:43 AM PT

1. 14 décembre 2025 à 16:40 UTCresolved

   The Okta Engineering team is continuing to work with our third-party vendor and addressing this issue with urgency related to the custom domain email delivery issues.

2. 14 décembre 2025 à 16:42 UTCresolved

   The next update will be posted in the next 60 minutes.

3. 14 décembre 2025 à 17:50 UTCresolved

   Our Engineering team is continuing to work with the third-party provider on mitigation actions to help resolve this outstanding issue. We are treating the issue with the utmost urgency to restore our services as soon as possible. The next update will be provided within the next 60 minutes or sooner, if additional information becomes available.

4. 14 décembre 2025 à 18:48 UTCresolved

   We have identified a potential root cause of this issue, and are currently investigating next steps towards resolution. We will provide the next update in 60 minutes, unless new information becomes available.

5. 14 décembre 2025 à 19:39 UTCresolved

   This issue has been successfully resolved as of 11:07AM PDT. All services have been restored. If you continue to see issues, please update your support case for assistance. Additional root cause information will be available within 5 Business days.

6. 23 décembre 2025 à 14:49 UTCresolved

   We sincerely apologize for any impact this incident has caused to you, your business, or your customers. At Okta, trust and transparency are our top priorities. Outlined below are the facts regarding this incident. We are committed to implementing improvements to the service to prevent similar future occurrences. Detection and Impact: On December 14, 2025, at 3:01 a.m. PST, Okta detected an issue that prevented email delivery for customers using custom email domains. This affected end-users' ability to receive emails for Multi-Factor Authentication (MFA), password resets, registration activation, account lockout notifications, and other notification emails, such as rate-limit notifications for administrators. Root Cause Summary: Okta has determined that the issue was caused by an unintended configuration change on our Mail Transfer Agents (MTAs). This change incorrectly routed emails from custom domains without proper DMARC sender authentication. As a result, email service providers rejected the emails, causing the delivery failures. Remediation Steps: Upon detection, Okta's engineering teams immediately began diagnosing the delivery failures. After identifying the incorrect routing, we updated the configuration to ensure emails were sent through the appropriate authenticated channels. Normal email functionality was fully restored after the updated configuration was applied. Preventative Actions: Okta is taking several steps to prevent a recurrence of this issue. We are enhancing our monitoring and alerting for email services and improving our configuration management tools to add further safeguards. Duration (# of minutes):1140