Historial de interrupciones de Poll Everywhere

1. 2 de julio de 2026 a las 19:48 UTCidentified

   We have identified that Denmark numbers are not working for SMS. We have contacted our SMS provider and are in the process of resolving this issue. As a workaround, you can go to Profile -> International text messaging -> Coverage area to rotate to a different number.

2. 6 de julio de 2026 a las 15:38 UTCidentified

   We have disabled the Denmark SMS number and failed over to the Germany phone number while the Denmark number is not working.

3. 13 de julio de 2026 a las 22:10 UTCidentified

   The incident has been updated to include every international number that is not US / Canada. As a workaround, you can go to Profile -> International text messaging -> Coverage area to rotate to either US / Canada.

4. 13 de julio de 2026 a las 22:48 UTCresolved

   This incident has been resolved.

Traducido automáticamente desde la actualización oficial del incidente.

1. 25 de junio de 2026 a las 9:19 UTCinvestigating

   Some tasks are currently delayed (such as deleting polls) in 1.0.

2. 25 de junio de 2026 a las 9:23 UTCinvestigating

   The system has self-healed and brought the delay down to ~5 minutes and dropping. We are continuing to investigate.

3. 25 de junio de 2026 a las 9:47 UTCidentified

   We've identified the autoscaling performance issue and are working on implementing a long-term fix.

4. 25 de junio de 2026 a las 10:45 UTCresolved

   We've changed the self-healing and autoscaling system to be more tolerant of faults at the times that the system has increased load. We've confirmed that the system is functioning normally as of right now and believe the problem to be resolved in the long term.

Traducido automáticamente desde la actualización oficial del incidente.

1. 25 de junio de 2026 a las 8:53 UTCresolved

   We've discovered an automated monitoring issue that led automatic scaling to slow down certain user actions (like trashing polls) in 1.0. This issue is resolved, but we're looking into the root cause with the automatic monitoring system.

Traducido automáticamente desde la actualización oficial del incidente.

1. 24 de junio de 2026 a las 19:46 UTCresolved

   Some users when trying to login to Poll Everywhere 2.0 were redirected incorrectly resulting in a 404 page. The root cause has been resolved.

Traducido automáticamente desde la actualización oficial del incidente.

1. 15 de junio de 2026 a las 15:53 UTCresolved

   An automatic service update resulted in an outage for about 5 seconds, where approximately 80 requests failed. We have taken measures to prevent automatic service updates from creating this type of downtime in the future. Please contact support@polleverywhere.com if you have any questions about this incident.

Traducido automáticamente desde la actualización oficial del incidente.

1. 26 de mayo de 2026 a las 3:31 UTCidentified

   We've identified a disruption with our AI models. Our third-party service provider, Google, has discontinued the large language model used to generate poll content. We are working to migrate to a different model. During this time, AI features in Poll Everywhere 2.0 are unavailable. This issue does not affect Poll Everywhere 1.0.

2. 26 de mayo de 2026 a las 4:43 UTCresolved

   We have migrated to a comparable or equivalent model to the deprecated model. We will be taking further steps to understand this type of issue and prevent similar types of issues in the future. This issue started May 25, 2026 at midnight (i.e., 12:01am) PT.

Traducido automáticamente desde la actualización oficial del incidente.

1. 5 de mayo de 2026 a las 5:28 UTCmonitoring

   A security incident affecting Instructure (Canvas) has been announced. Poll Everywhere is not affected by this security incident. It is possible that later changes that affect Canvas may require re-initializing Canvas LMS connections, but currently we are not aware of customers experiencing problems with LMS integrations as a result of this security incident. Please contact Poll Everywhere if you are experiencing any issues with Canvas at support@polleverywhere.com. See: https://status.instructure.com/incidents/9wm4knj2r64z and https://status.instructure.com/incidents/ns0sjlmkvc6g. For information about the Canvas security incident, please reach out to your contact at Instructure directly.

2. 11 de mayo de 2026 a las 4:52 UTCmonitoring

   We'd like to express our deepest sympathy for our partners, customers, teachers, and students affected by the Canvas security incident, particularly following the events of late last week. We understand that this is a particularly challenging time for everyone impacted, and we are sorry for the profound impact this has had on everyone during final exams and end-of-semester work. Poll Everywhere remains unaffected by the security incident with Canvas. Poll Everywhere has been conducting internal investigations and reviews, and confirmed that there is no impact to the Poll Everywhere platform at this time. Additionally, Instructure has not notified Poll Everywhere that any data related to its services has been impacted. Poll Everywhere's connection to Canvas uses the LTI Advantage standard, an industry-common integration pathway with a variety of LMS platforms, including (but not limited to) Canvas. We rely on a trusted set of keys provided by Instructure that validates that incoming data from Canvas is authentic. This key material is only changed by Instructure, and they have not changed it. However, if Instructure rotates this key material, Poll Everywhere will seamlessly update its data to trust the newly issued key material. The LTI standard also defines a set of keys that are maintained by Poll Everywhere. Poll Everywhere's key material (private keys) remains safely stored with Poll Everywhere, and has not been compromised. As a result, no API rotation, credential rotation, or other security-adjacent credential changes are necessary for any customer or end-user of Poll Everywhere to undertake, with respect to the Canvas integration, at this time. Please reach out to support@polleverywhere.com if you have any questions related to the Canvas security incident and how it affects Poll Everywhere. Please refer to https://www.instructure.com/incident_update and https://status.instructure.com/ for the status of the Canvas platform.

3. 14 de mayo de 2026 a las 5:37 UTCresolved

   Thank you for your continued use and support of Poll Everywhere. We're sorry for the impact that the incident at Instructure has had on everyone impacted. With respect to the security incident at Instructure regarding Canvas, we believe that Instructure has directly communicated to all affected customers with resolution information about this incident. We believe this situation to be closed, and we will continue to monitor the situation in the background. For more information, on the Canvas side, please see: https://www.instructure.com/incident_update. Once again, Poll Everywhere was not impacted during the Canvas security incident. In terms of future impact to Poll Everywhere, we expect no action to be required in terms of rotation of LMS setup or configuration. We alter and adjust our security program, including organizational and technical controls, in response to security incidents in the industry, including this one. We will incorporate this incident into our future planning. We remain in contact with Instructure and will monitor their response to this situation going forward. Again, to be extremely clear, Poll Everywhere has had no impact from this incident at Instructure. However, we believe it is our role as members of the security community to learn from all incidents, including this one. If you have any questions, please feel free to reach out to support@polleverywhere.com.

Traducido automáticamente desde la actualización oficial del incidente.

1. 4 de mayo de 2026 a las 8:28 UTCresolved

   A small number of customers may have encountered an error upgrading their plans as a result of a tax calculation issue. This issue has been resolved, but the upgrade process will not automatically retry to prevent erroneous charges. Please retry those upgrades. If you encounter any issues upgrading, please contact support@polleverywhere.com.

Traducido automáticamente desde la actualización oficial del incidente.

1. 22 de abril de 2026 a las 5:06 UTCresolved

   Live updates were delayed for ~14 minutes due to a deployment configuration error

Traducido automáticamente desde la actualización oficial del incidente.

1. 14 de abril de 2026 a las 20:08 UTCinvestigating

   We've observed a elevated number of error across the site. We are currently investigating a root cause.

2. 14 de abril de 2026 a las 20:23 UTCresolved

   A fix has been implemented. This incident was caused by a framework upgrade causing database timeouts in requests. During this outage, activation, poll creation / editing, as well as several other core features experienced disruption due to database requests timing out.

Traducido automáticamente desde la actualización oficial del incidente.

1. 9 de febrero de 2026 a las 16:42 UTCidentified

   There is an incompatibility with Keynote 15 and the Poll Everywhere app. Currently, the Mac Presenter version of Poll Everywhere for Mac does not support Keynote 15, which was released with Apple Creator Studio. If you use Poll Everywhere, please note that if you install Keynote 15, you will not be able to use Poll Everywhere until we publish a software update. As a workaround, please postpone updating to Keynote 15 or using the Apple Creator Studio version of Keynote.

2. 9 de febrero de 2026 a las 23:30 UTCresolved

   We have updated the guidance and error messaging for using the Poll Everywhere Mac Presenter App with Keynote. You can find more information here: https://support.polleverywhere.com/hc/en-us/articles/46464006721307-Keynote-system-requirements-for-Poll-Everywhere

3. 23 de febrero de 2026 a las 18:36 UTCpostmortem

   Keynote Creator Studio changed the bundle ID for Keynote, removing a reference to the legacy iWork suite. Mac Presenter version 4.0.1 has been released and is available via the in-app “Check for Updates” menu and via our website, which will support Keynote 14 and Keynote 15.1 \(in Keynote Creator Studio\). See: [https://www.polleverywhere.com/app/powerpoint/mac](https://www.polleverywhere.com/app/powerpoint/mac)

Traducido automáticamente desde la actualización oficial del incidente.

1. 5 de febrero de 2026 a las 2:51 UTCidentified

   We have identified the issue and are deploying a fix.

2. 5 de febrero de 2026 a las 3:00 UTCresolved

   This incident has been resolved.

Traducido automáticamente desde la actualización oficial del incidente.

1. 30 de enero de 2026 a las 7:01 UTCinvestigating

   We're investigating a situation where pe.app is serving pages that indicate a 500 internal server error has occurred.

2. 30 de enero de 2026 a las 7:05 UTCresolved

   We've resolved the issue, and the outage has been resolved. Please contact support@polleverywhere.com if you continue to encounter any errors or defects.

3. 30 de enero de 2026 a las 7:29 UTCpostmortem

   An underlying defect in the latest version of the framework we use – Ruby on Rails – led to a temporary failure during a routine update. We are working to get the fix merged in Ruby on Rails to resolve the defect in the long term.

Traducido automáticamente desde la actualización oficial del incidente.

1. 26 de enero de 2026 a las 18:09 UTCidentified

   We've identified an issue where some billing emails and links on Poll Everywhere 2.0 go to "checkout.stripe.com", "buy.stripe.com", or "billing.stripe.com", or Stripe, our payment processor, instead of billing.pe.app. We're working on resolving the issue. In the interim, these links are safe. If you are curious about the authenticity of any email or link, please contact support@polleverywhere.com.

2. 26 de enero de 2026 a las 21:05 UTCresolved

   This has been resolved

Traducido automáticamente desde la actualización oficial del incidente.

1. 9 de enero de 2026 a las 15:45 UTCidentified

   We have identified the problem that is preventing Google Slides insertion. We are working on a new release that will fix the problem.

2. 9 de enero de 2026 a las 16:43 UTCmonitoring

   We have submitted a new version of the Poll Everywhere for Google Slides app to the Google extension store. We will update this status to Resolved once v1.20.4 has been approved and is available.

3. 12 de enero de 2026 a las 1:40 UTCresolved

   Version 1.20.4 of our Chrome extension has been approved in the Chrome Web Store, which resolves the problem.

Traducido automáticamente desde la actualización oficial del incidente.

1. 8 de enero de 2026 a las 19:19 UTCresolved

   This issue is now resolved.

Traducido automáticamente desde la actualización oficial del incidente.

1. 18 de diciembre de 2025 a las 0:26 UTCresolved

   This incident has been resolved

Traducido automáticamente desde la actualización oficial del incidente.

1. 5 de diciembre de 2025 a las 9:38 UTCresolved

   An incident with our third-party service provider, WorkOS, resulted in a small number of errors today, around 12:50AM, primarily experienced when logging into or using pe.app. We have contacted our third-party service provider for details about this outage. As this issue has been resolved on their side, we do not expect to see further impacts at this time. Please refer to this incident on their status page: https://status.workos.com/incidents/llh8hvkdm9mb Additionally, though our status page is integrated with their system status, it did not appear as an outage on our page. We have also contacted them about this incident. If you continue to experience any errors, please contact support@polleverywhere.com. Edit (1:44 PT): This was a Cloudflare global outage: https://www.cloudflarestatus.com/incidents/lfrm31y6sw9q

Traducido automáticamente desde la actualización oficial del incidente.

1. 24 de noviembre de 2025 a las 18:43 UTCresolved

   From 4:57 PM Pacific through 5:45 PM Pacific, there were intermittent issues with live updates on Poll Everywhere 2.0. The root cause of this issue has been identified and fixed.

Traducido automáticamente desde la actualización oficial del incidente.

1. 20 de octubre de 2025 a las 7:28 UTCidentified

   Ongoing incident with various providers. Significantly affecting authentication in 2.0 (pe.app). Degraded performance in 1.0 (polleverywhere.com) affecting directory sync and SMS. See: - https://status.workos.com/incidents/b178c1kxt0g3 - https://health.aws.amazon.com/health/status

2. 20 de octubre de 2025 a las 8:52 UTCidentified

   We are continuing to monitor this issue

3. 20 de octubre de 2025 a las 9:16 UTCidentified

   Ongoing incident with various providers. We've identified further systems affected. Significantly affecting authentication in 2.0 (pe.app). Degraded performance in 1.0 (polleverywhere.com) affecting directory sync and SMS. Additionally our support systems for email and phone support are experiencing outages. See: - https://status.workos.com/incidents/b178c1kxt0g3 - https://health.aws.amazon.com/health/status (various services) - https://status.zendesk.com/ (email support) - https://status.talkdesk.com/ (phone support)

4. 20 de octubre de 2025 a las 9:51 UTCidentified

   Various components of our service are recovering as a result of the Amazon Web Services US-West-1 outage that occurred earlier today. Some services are partially recovered or are performing as expected on an intermittent basis right now. In-particular, sign-in to pe.app via WorkOS and directory sync services have been partially restored. We continue to be unable to accept inbound support phone calls, and some SMS numbers provided by Twilio remain inoperable. Additionally, as services are restored in phases, we expect there to be small amounts of errors in the platforms as the cascade failure is recovered from. Currently, users with existing sessions on pe.app and polleverywhere.com, as well as responses on pollev.com remain fully functional.

5. 20 de octubre de 2025 a las 9:58 UTCmonitoring

   Services look to be restored by our providers, we are continuing to monitor

6. 20 de octubre de 2025 a las 11:45 UTCresolved

   This incident has been resolved. This incident was caused by the failure of an upstream provider, Amazon Web Services (us-east-1). When AWS us-east-1 failed, a cascade failure occurred with several of our service providers. Poll Everywhere is hosted in Amazon Web Services (us-east-2) and although some of our third-party service providers, including Canvas, Talkdesk, Slack, Zendesk, Twilio, and WorkOS failed, our core platform on PollEverywhere.com remained fully functional. With respect to Poll Everywhere 2.0 at pe.app, we are working with our upstream provider, WorkOS, to determine solutions to future AWS outages. Please reach out to us at support@polleverywhere.com if you are continuing to experience issues with any of our services.

7. 27 de octubre de 2025 a las 6:26 UTCpostmortem

   During the AWS us-east-1 outage that occurred on October 19 and 20th of 2025, Poll Everywhere experienced degraded services, particularly with our 2.0 product at pe.app, and some internal systems. Our third-party service providers, Amazon Web Services and WorkOS, have published technical write ups related to the incident. We have read these reports and understand the impact that the AWS us-east-1 outage had on our services and how we can improve our response to these incidents in the future. For more technical details, please see: * [Update - AWS services operating normally](https://www.aboutamazon.com/news/aws/aws-service-disruptions-outage-update) * [WorkOS - Service disruption on October 20, 2025](https://workos.com/blog/service-disruption-on-october-20-2025) Immediately following the outages that impacted Poll Everywhere, we contacted our partner vendors, including WorkOS, to understand their response to our inquiry and the greater outage. In-particular, we understand that WorkOS will be conducting multi-region deployments of their critical services in a hot standby configuration by the end of Q1 2026, which will eliminate this type of failure as a possibility in the future. With the exception of WorkOS in 2.0, our core product and services remained online with only minor disruption to some SMS providers. As a result, we do not have other remediation work planned as a result of this outage. We will continue to monitor WorkOS and work with the WorkOS team to provide feedback and dialog as they work to implement hot standby support and multi-region deployments over the coming months.

Traducido automáticamente desde la actualización oficial del incidente.