All grok bot routines failed to run

Where does the bug appear (feature/product)?

Grok Bot

Describe the Bug

Scheduled routines stopped firing for all my bots on Thu 8 Oct 2026. Inbox’s scans (every :29 and :59, 08:29 to 20:59 UK) ran until about 14:00, failed at 16:55, and haven’t run since, even after I paused and resumed the routine. Another 16:15 routine also failed, and a one-off routine set for 18:06 never fired. The bots work normally when I message them, so it looks like a server-side scheduler fault

Steps to Reproduce

All grok bot routines failed to fire at the scheduled times

Expected Behavior

Routines run at scheduled times

Screenshots / Screen Recordings

Operating System

MacOS
iOS

Version Information

Grok Bot 1.14.0 (17267) · runtime e9651e08edc0248a43ce2d13839fa55c7d2e1425 · embedded bundle · channel production

Additional Information

Paused and Resumed, created a copy of the routine instead both failed to solve the issue

Does this stop you from using Cursor

No - Cursor works, but with this issue

Same situation, for me and for my contacts is not working

Hey, thanks for the report and for the screenshot from Run history.

What you’re seeing isn’t related to your bots or settings. Starting around midday UK time on October 8, scheduled routines in Grok Bot were delayed and then marked as Failed during a period of very high load. Direct messages to bots were not affected, which matches what you saw.

Over the last hour, things have stabilized and routines are running again. Pausing and resuming or copying the routine won’t change anything here, so if you disabled any of them while troubleshooting, turn them back on.

Missed runs from that window won’t be replayed automatically. If any routine still shows Failed after 21:30 UK time tonight, reply here with the routine name and the time it was scheduled for, and I’ll take another look.

Same thing happening in here

I asked if the routine runs, bot send me this

Hey @Derek2, @x_panika, @Matheus_Campanha, a quick correction to my earlier update.

Routines did start running again for a while this evening, but since about 21:00 UK time scheduled routines are being held back again while we deal with very high demand on our side. Nothing is wrong with your bots or settings, and messaging your bots directly still works normally.

There’s no need to pause, resume, or recreate routines. That won’t change anything here, so please leave them switched on.

Runs missed in this window won’t be replayed automatically. We’ll post an update in this thread once routines are running normally again.

I did a summary from my desktop:

1. “Triagem Gmail a cada 20 min”, owned by the Triagem Gmail bot. This is the main routine, which labels Gmail.

  • It ran normally on its own all morning. The logged runs were at 09:00, 09:11, 09:24, 09:36, 09:43, 10:03, 10:22, 10:46 and 11:00.
  • At 11:04 the schedule changed to every 5 minutes (4,9,14,…,59 * * * *). After that, no run fired. The status kept showing the last success with no error.
  • At 14:44 it was switched to */5 * * * *. One run fired at 14:49 and succeeded, then nothing more.
  • At 15:18 it was switched to */10 * * * *, the format that worked in the morning. Nothing fired.
  • At 15:27 it was switched to @every 5m. One run fired at 16:21 and succeeded, then nothing more.
  • At about 16:50 it was switched to @every 20m. No runs had fired by 17:19.
  • Manual runs at 14:24, 15:26, 15:32 and 16:31 all worked fine, so the routine’s own logic isn’t the problem.

2. “Monitorar conserto da Triagem Gmail (hoje)”, owned by me. This is a 15-minute health check.

  • It was created at about 14:20 with 3,18,33,48 14-19 8 10 * and never fired.
  • At 14:50 it was switched to */15 * * * *. It still showed “never run” at 15:25, fired once at 16:22 and succeeded. That is the last run I’ve seen.

3. “Hello world a cada 5 min”, owned by a new bot, Teste Rotina. It’s a test that only writes one line to a file.

  • It was created at 16:33 with @every 5m. As of 17:19 it has never fired, and the file was never created.

4. “Teste hello world Triagem”, owned by the Triagem Gmail bot. It’s the same test as #3, created inside the existing bot.

  • It was created at 16:36 with @every 5m. As of 17:19 it has never fired.

Summary for the report: scheduled routines stopped firing reliably at around 11:00. It affects old and brand-new routines across 3 bots, in every schedule format. They fire sporadically (14:49, 16:21, 16:22), and there’s no error in the status. The routine panel also won’t open a routine’s details. Manual runs work, and the machine isn’t overloaded.

Still not working for me. This is an account from my client, I already spent 30% of its usage today trying to figure this out, I think the usage will hit the max soon and we won’t be able to triage emails anymore. If we can fix this and reset part of the usage it would be great :slight_smile:

Continuing to have same issue

Thanks for the update. Totally understand these things happen, but I’d appreciate it if issues like this could be communicated on the server status page. I’ve spent quite a chunk of my usage today trying to troubleshoot the problem.

Thanks for keeping us updated anyway!

Yeah, same thing here, bunch of my usage was wasted.

Let’s keep on track

Quick status update: scheduled Grok Bot routines are still affected and have not fully recovered. The team is working to restore normal operation. We don’t have a timeline to share yet, but we’ll update this thread when routines are running normally again.

Please leave routines enabled. Pausing, copying, or recreating them won’t help. For urgent tasks, messaging the bot directly may still work. Missed runs won’t be replayed automatically. I’ll post here when I have an update.

Thanks for this post - I have been going out of my mind trying to figure out why my routines are not working… the cursor website is saying resolved now is that correct? Still not seeing a positive run at my end. Thanks for your help!

Hi @Mark_S2 Sorry for the confusion. The issue resolved indicator on the https://status.cursor.com/ was referring to a different issue. The routine issue is being worked on and I’ll update you here when it is fully sorted out. Thank you for your patience.

Feel free to ask your Grok Bot to run the routines manually - that should work - I realize it’s not ideal but it’s a stopgap.

Amazing thank you - to be fair GrokBot has been doing overtime trying to troubleshoot the situation! Will keep running manually until the fix is confirmed via this thread.

Hi team, wanted to see if it’s fixed? If not, is there any ETA available?

Are you sure this issue has been resolved, because I’m still experiencing this issue? It’s been going on all day, and I know it’s not me. I have multiple routines all failing or not even starting at all.

Schedules still not working from my side in 2 accounts

Hi Kevin. I’ve got a few scheduled runs this morning so I’d assume it’s back on track now? Told the primary bot to run a health check on all the bots and they seemed like running fine now.

thanks for looking into this.

Updating this thread, I have been seeing the same behaviors, and it is still ongoing.

Both cursor automations and Grok bot schedules are still not working. They are either very late if we are lucky, or just never fire, I have been testing multiple ways to recover or work around, but it ended up using a lot of my usage allocation.

Same here, at >90% usage now still with 5 days to go after trying to troubleshoot yesterday with bots working overtime. Routines seem be running better for me at the moment (IOS log showing no failures in the history for a run that is set every 5mins). @kevinn has said he will update in this thread when its fully fixed.