Agent runs (agents.zapier.com) intermittently experiencing 429 errors
The issue intermittently affecting Agent runs and Zap runs using AI by Zapier has been resolved. The confirmed impact window goes from September 18, 2026 at 2:00am UTC to today, September 21, at 15:00 UTC. Our team continues to work with our service provider to prevent this issue in the future, implement better safeguards, and consolidate self-healing capabilities. Some Agent runs from previous days (September 18 to September 20) could not be replayed to avoid duplicate agent executions. For **Agents**, if you want to replay affected runs, go to your[ Agents dashboard](https://agents.zapier.com/bots), open the [All Activity](https://agents.zapier.com/activity) tab to view recent runs, and review the agent runs during the incident window. You can replay any runs that need to be reprocessed by prompting the agent in the chat to retry. For Zap runs using **AI by Zapier**, **no action needed.** Our team has successfully replayed all affected AI by Zapier runs. Many thanks for your patience
The latest update from Zapier.
State
Resolved
Started
Lasted
3 h 58 min
Worst impact
Degraded
Timeline
3 updatesInvestigating
2 h 28 min 62%
Identified
1 h 30 min 38%
Updates
Times in UTCAs posted by Zapier
The issue intermittently affecting Agent runs and Zap runs using AI by Zapier has been resolved. The confirmed impact window goes from September 18, 2026 at 2:00am UTC to today, September 21, at 15:00 UTC. Our team continues to work with our service provider to prevent this issue in the future, implement better safeguards, and consolidate self-healing capabilities. Some Agent runs from previous days (September 18 to September 20) could not be replayed to avoid duplicate agent executions. For **Agents**, if you want to replay affected runs, go to your[ Agents dashboard](https://agents.zapier.com/bots), open the [All Activity](https://agents.zapier.com/activity) tab to view recent runs, and review the agent runs during the incident window. You can replay any runs that need to be reprocessed by prompting the agent in the chat to retry. For Zap runs using **AI by Zapier**, **no action needed.** Our team has successfully replayed all affected AI by Zapier runs. Many thanks for your patience
The team has identified the issue to be an intermittent overload of a service powering: • Agent runs (agents.zapier.com) • AI by Zapier executions As a result, some agent runs and AI by Zapier executions hit 429 errors. We will be working on replaying affected agent runs or Zap runs where possible, and we'll keep this status page updated on our progress. Many thanks for your patience. Our Support team is available if you have any questions, and you can reach them at: zapier.com/get-help.
Our team is currently investigating an issue with Agent runs (from agents.zapier.com) intermittently experiencing 429 errors for some customers. We're working to confirm the root cause and implement a fix as soon as possible. AI by Zapier runs are not affected by this incident. If you have any questions or need help, please reach out to our Support team at zapier.com/get-help.
Zapier in the last 90 days
Full historyIncidents in 90 days
25
Days with an incident
36
Median duration
3 h 2 min
This incident
1.3× the median
Other Zapier incidents
All incidentsZapier·"Chat with Sales" page not loading
We have tested the page post-fix and the issue is resolved. Users visiting the "Chat with Sales" page ( ) should now be able to see the chat session. This incident is now considered resolved. If you don't see the chat, we recommend clearing your browser cache and cookies, then trying the page again. If you continue to have issues with the page, reach out to our support team at [zapier.com/app/get-help](http://zapier.com/app/get-help).
Zapier·Discord integration issues
Discord requests made through [Zapier's Proxy API](https://docs.zapier.com/proxy-api/overview) — where Zapier adds the credentials for a connected Discord account on behalf of your code or an agent (for example via the Zapier SDK, Code by Zapier, Custom Actions, MCP code actions, or Sidekick) — have been disabled and will no longer be available. Affected requests return a `401 Unauthorized` error. This does not affect Zapier's native Discord integration — standard Discord triggers and actions continue to work as normal. Code steps that call Discord with a token you supply yourself (rather than a connected Discord account) are also unaffected. If you were making direct Discord requests this way, you can switch to Zapier's [native Discord integration](https://zapier.com/apps/discord/integrations). If you need help, reach out at .
Zapier·Connection errors affecting Salesforce Pardot Zaps
This issue originated in an upstream incident with Salesforce's Pardot service, which caused some Pardot connections to fail when refreshing authentication. Salesforce has remediated their side, but affected connections need to be reconnected. We've deployed a fix on our side so that affected connections now show as expired. To restore your connection, reconnect your Salesforce Pardot account in Zapier. Some Zaps may have been delayed or held, and we're checking that they resume after reconnecting. If your Zaps are still failing after reconnecting, or you need help, please contact our support team: [ https://zapier.com/app/get-help](https://zapier.com/app/get-help " https://zapier.com/app/get-help")
Zapier·Security updates in progress for third-party software components
We have not identified any customer impact or evidence of exploitation. We have completed the priority security updates and mitigations, and are closing this incident. Remaining lower-risk updates, including updates managed by our providers, will continue as follow-up work. No action is needed on your part.
Zapier·Assessment of recently reported PHP vulnerabilities
Our investigation into the reported PHP vulnerabilities is complete. We found no impact on Zapier services or customer data. This advisory is now closed. **Security FAQ: PHP Vulnerabilities (CVE-2026-109618 and CVE-2026-109619)** > **1. What is this issue?** > Two vulnerabilities were disclosed in PHP, an open-source programming language. Both can lead to remote code execution under certain conditions. They affect PHP 8.2 before 8.2.35, 8.3 before 8.3.36, 8.4 before 8.4.27, and 8.5 before 8.5.12. > > **2. Is Zapier affected?** > Our investigation is complete, and we found no impact on Zapier services or customer data. > > **3. Has customer data been affected?** > No. Our investigation found no impact to customer data. > > **4. Does this affect the Zapier WordPress integration for example?** > The Zapier WordPress plugin itself is not vulnerable. WordPress sites run on PHP, though, so site owners should make sure their PHP version is patched (see question 1 for the affected versions). >
Follow Zapier
Your coding agent knows when Zapier is down, and tells you when a failure is theirs, not yours. You can also get Zapier incidents in a feed reader or in Slack.
Install the outages.live MCP server, so that when something breaks you can tell if the cause is a provider we depend on and not my code. It reads the official status pages of cloud, AI, payment and developer providers and gives you their incidents as MCP tools.
Add it as a remote MCP server named "outages" at https://outages.live/mcp (Streamable HTTP, no authentication), with your CLI command or your MCP config file.
Then:
1. Make sure it connects and shows the tools get_provider, list_feed and send_feedback.
2. Call get_provider with "zapier" and tell me if Zapier has a problem right now.
3. From now on, when a call to a third-party service fails or becomes slow, call get_provider for that service before you change my code.Paste this into any coding agent. It adds the server with its own MCP setup.
https://outages.live/feed.xml?providers=zapier/feed subscribe https://outages.live/feed.xml?providers=zapierFAQ
Go from tracking outages to fixing production issues with Polylane.
Try Polylane