Elevated 504 Errors When Sending Messages via Google
Update - We have submitted a high priority bug ticket to Google for further investigation into the intermittent 502 errors (returned by Gmail, surfaced as 504s by Nylas). Along with the ticket, we included detailed logs and a sample payload to support their analysis.

We’re continuing to monitor the situation closely and will provide updates as soon as we receive a response from Google.

Thank you for your continued patience.
Jul 10, 2025 - 17:02 PDT
Update - We’ve observed a noticeable decrease in the number of 502 errors from Google when sending messages (Nylas 504s). However, some errors are still occurring.

Our team continues to monitor the situation closely, and we’ll share another update as soon as we have more information.

Thank you for your continued patience.
Jul 10, 2025 - 11:55 PDT
Investigating - We are currently investigating an increase in 504 errors when sending messages through Google accounts via the v3/message/send endpoint. This appears to be related to a temporary error on Google’s side, as we are receiving the following response from their servers:

502. That's an error.

"The server encountered a temporary error and could not complete your request.
Please try again in 30 seconds. That's all we know."

Our engineering team is monitoring the situation closely and has opened a support ticket with Google to gain further insights. We will provide updates as more information becomes available.

Thank you for your patience and understanding.
Jul 10, 2025 - 10:51 PDT

About This Site

Seeing issues that aren't reflected on this page? Contact us at support@nylas.com

Nylas Application Services Operational
API v3 ? Operational
Webhooks v3 ? Operational
Authentication v3 ? Operational
Notetaker ? Operational
Scheduler v3 ? Operational
Dashboard v3 ? Operational
IMAP v3 ? Operational
Data Connectors v3 Operational
3rd Party Providers Partial Outage
Microsoft 365 ? Operational
Outlook.com ? Operational
Google ? Partial Outage
Yahoo! ? Operational
iCloud ? Operational
Other Email Server Hosting Providers ? Operational
Mailgun ? Operational
Oxylabs ? Operational
Cloudflare ? Operational
Subprocessors Operational
* Subprocessor Change Notification ? Operational
AWS ? Operational
GCP ? Operational
AssemblyAI ? Operational
Atlassian Pty Ltd ? Operational
Apollo ? Operational
CloudFlare ? Operational
Coralogix ? Operational
DataGroomr ? Operational
Datasiv ? Operational
Discourse ? Operational
Fastly ? Operational
FiveTran ? Operational
Gong.io ? Operational
Google ? Operational
Google ? Operational
Gravitational (Teleport) ? Operational
Honeycomb.io ? Operational
Hubspot ? Operational
Inflection.io ? Operational
Ketch ? Operational
Looker ? Operational
Madkudu ? Operational
Mailgun ? Operational
Microsoft ? Operational
New Relic ? Operational
PandaDoc ? Operational
PartnerPortal ? Operational
PlayerZero ? Operational
RB2B ? Operational
Recurly Inc. ? Operational
Rootly, Inc. ? Operational
Salesforce ? Operational
Salesloft ? Operational
Slack ? Operational
Strac ? Operational
Stripe ? Operational
Tropic ? Operational
Twilio Segment ? Operational
Vercel ? Operational
Vitally ? Operational
Whistic ? Operational
Woopra ? Operational
Zendesk ? Operational
Zoom ? Operational
Operational
Degraded Performance
Partial Outage
Major Outage
Maintenance
Jul 14, 2025
Temporary workaround in place for Microsoft Graph API issue
Resolved - We’ve received an official update from Microsoft confirming that their engineering team has deployed a fix to address the Graph Calendar API issues. This fix has been fully validated for business (work/school) accounts. However, Microsoft has stated they are unable to confirm the behavior for personal Outlook.com accounts.

Based on our own monitoring, we’ve observed that while some 500 errors may still occur for personal accounts, our existing safeguards have been effective in mitigating customer impact. As a result, we’ve made these protections permanent within the Nylas platform.

We are now resolving this incident but will continue to monitor for any recurrence affecting personal accounts.

Thank you for your continued patience and understanding.
Jul 14, 11:25 PDT
Update - We’ve received confirmation from Microsoft that their backend team has made partial progress toward resolving the upstream Graph Calendar API issue. Specifically, they’ve reported that the API fix is approximately halfway saturated through their infrastructure.

Microsoft has extended the tentative ETA for full production rollout to June 30, 2025.

In the meantime, our existing workaround remains in place, and core functionality continues to operate with minimal disruption. We will continue to monitor for any changes and provide updates as we receive them from Microsoft.

Thanks for your continued patience
Jun 17, 18:29 PDT
Update - We’re still monitoring the issue impacting Microsoft’s Graph Calendar API, which affects access to some calendar data.

Microsoft has completed internal testing and is now working on finalizing a fix. Although this issue lies outside of our product, we remain in close communication with their support team and are pushing for timely updates.

We’ll continue to share more information as it becomes available. Thanks for your patience.
Jun 6, 10:07 PDT
Update - We’ve been in contact with Microsoft support regarding the ongoing issue with their Graph Calendar API. Microsoft has acknowledged the problem and confirmed that their engineering team is actively working on a resolution.

We’ll continue to monitor the situation and provide updates as new information becomes available. Thank you for your patience.
Jun 4, 10:21 PDT
Monitoring - We’ve implemented a workaround to mitigate the impact caused by an issue with Microsoft Graph API. As a result, the affected functionality has been restored, and users should no longer experience errors when accessing related features.

Please note that this is a temporary mitigation, and the root cause remains unresolved on Microsoft’s side. We are actively monitoring for updates and will revert our changes once Microsoft provides a permanent fix.

We appreciate your patience and will continue to keep you informed.
Jun 3, 10:04 PDT
Update - We are still observing the ongoing issue with the Microsoft Graph API, which continues to affect some Microsoft grants. Impacted users may experience 500 and 504 errors on calendar-related endpoints, including delays or failures in webhook delivery.

As of now, there has been no resolution or update from Microsoft. Our team will continue monitoring and will provide further updates as soon as we detect meaningful changes.

This will be our final status page update for today unless there is a significant development. Thank you for your patience.
Jun 2, 17:30 PDT
Update - We are continuing to monitor the ongoing issue with the Microsoft Graph API, which is impacting some Microsoft grants. The issue is still causing 500 and 504 errors on calendar-related endpoints, and delays or failures in webhook delivery for affected users.

There has been no meaningful change since our last update. Our team remains in contact with Microsoft and is actively monitoring for improvements. We will provide further updates as soon as we see progress or receive new information.
Jun 2, 16:13 PDT
Identified - We are currently experiencing service degradation caused by an issue with the Microsoft Graph API. As a result, some Microsoft graph grants are experiencing 500 and 504 errors. For affected users, this may impact webhooks and the following API calls:

calendars
event
scheduler

Our team is actively monitoring the situation and has reported the issue to Microsoft. We will continue to provide updates as more information becomes available
Jun 2, 12:00 PDT
US & EU: Intermittent Microsoft Graph Errors Creating Calendars & Events
Resolved - Microsoft has confirmed that the Graph API errors affecting calendar and event creation were caused by a backend issue related to the deprecation of Skype. This introduced a conflict in the code used to retrieve calendar information, particularly for shared calendars.

A fix has now been fully deployed across their infrastructure, and internal testing by Microsoft has validated that the issue was resolved as of July 8, 2025.

We have observed no further errors on our side since the fix was applied. Microsoft has also indicated that they are reviewing the impacted code paths to improve testing and prevent similar issues in the future.

This incident is now considered resolved. Thank you for your patience throughout this issue.
Jul 14, 07:54 PDT
Update - The issue remains ongoing, with no change in behavior observed. We have submitted a Severity A support ticket with Microsoft and are currently awaiting their response regarding the Graph API errors impacting a subset of calendar and event creation requests.

Our engineering team continues to monitor closely and will provide updates as soon as new information becomes available.
Jun 23, 14:03 PDT
Identified - We are seeing `ErrorInternalServerError` API errors from a small number of Microsoft grants when attempting to create new events or new calendars in the Nylas API.

We have confirmed that these errors are coming from Microsoft Graph API email servers and have reached out to Microsoft to address this issue.

Additionally, if you retry a failed API call to create a new calendar, a different error 409 `ErrorFolderExists` may be returned. This error is also from Microsoft Graph API.

This does not appear to be affecting any on-premise Exchange (EWS) grants or other email providers.
Jun 23, 10:55 PDT
Jul 13, 2025

No incidents reported.

Jul 12, 2025

No incidents reported.

Jul 11, 2025

No incidents reported.

Jul 10, 2025

Unresolved incident: Elevated 504 Errors When Sending Messages via Google.

Jul 9, 2025

No incidents reported.

Jul 8, 2025

No incidents reported.

Jul 7, 2025

No incidents reported.

Jul 6, 2025

No incidents reported.

Jul 5, 2025

No incidents reported.

Jul 4, 2025

No incidents reported.

Jul 3, 2025

No incidents reported.

Jul 2, 2025

No incidents reported.

Jul 1, 2025

No incidents reported.

Jun 30, 2025

No incidents reported.

← Incident History Powered by Atlassian Statuspage