High CPU Usage from peopled After Updating to macOS Tahoe 26.4

The problem for me was that the peopled process was trying to access a file (console logs show repeated errors accessing $HOME/Library/Application Support/CallHistoryDB) that was quarantined. Removing the quarantine flag and restarting the daemons fixed it.

# Are the CallHistoryDB files quarantined? (shows com.apple.quarantine)
ls -lan@ "$HOME/Library/Application Support/CallHistoryDB"

If so then you can remove the quarantine flag and kill the process to restart them.

Note, the xattr -dr command needs your terminal app to have Full Disk Access (System Settings / Privacy & Security / Full Disk Access). Without it, you'll get permission-denied on the xattr write. Grant it, restart the terminal, retry.

# Strip the quarantine attribute from the CallHistoryDB tree
xattr -dr com.apple.quarantine "$HOME/Library/Application Support/CallHistoryDB" 

# Verify it's gone (should show no com.apple.quarantine lines now)
ls -lan@ "$HOME/Library/Application Support/CallHistoryDB"

# Restart the daemons; launchd will respawn
sudo killall -9 peopled CallHistorySyncHelper

Based on your description, the issue appears to be related to the peopled process becoming stuck after the update to macOS Tahoe 26.4. The peopled process is responsible for indexing and managing contact-related information used by apps such as Contacts, Mail, FaceTime, and Siri. It is normal for it to consume CPU briefly after a major update while rebuilding its database, but sustained usage of 70–80% CPU for two days is not expected.

Step 1: Allow background indexing to complete

Immediately after a macOS update, background processes may continue indexing data for several hours.

• Connect your MacBook to power.
• Keep it connected to Wi-Fi.
• Leave it idle overnight with the display on or set to prevent sleep.
• Check Activity Monitor the next day to see if CPU usage has returned to normal.

Step 2: Verify Contacts account synchronization

A syncing issue with one of your accounts can cause peopled to repeatedly process contact data.

1. Open System Settings > Internet Accounts.
2. Temporarily disable Contacts syncing for each account (iCloud, Google, Exchange, Outlook, etc.).
3. Restart your Mac.
4. Re-enable the accounts one at a time, checking CPU usage after each.

If CPU usage spikes after enabling a specific account, that account is likely causing the issue.

Step 3: Rebuild the Contacts database

If a corrupted contacts database is causing continuous indexing:

1. Quit the Contacts app.
2. Disable Contacts syncing in iCloud.
3. Restart your Mac.
4. Re-enable Contacts syncing and allow the database to download again.

Step 4: Test in Safe Mode

Since you've already restarted normally, boot into Safe Mode and monitor the CPU usage.

• If peopled behaves normally in Safe Mode, the issue may be caused by a login item or third-party extension loading during a normal startup.

Step 5: Test in a new user account

Create a temporary macOS user account and sign in.

• If peopled uses minimal CPU in the new account, the issue is likely isolated to your current user profile.
• If the same behavior occurs, it points to a system-wide issue introduced after the update.

Step 6: Check for software updates

Go to:

System Settings > General > Software Update

If Apple has released a supplemental update or a newer version of macOS Tahoe, install it, as it may contain fixes for post-update background service issues.

Step 7: Collect diagnostic information

If the issue continues:

1. Open Activity Monitor.
2. Select the peopled process.
3. Click the Info (i) button or Sample Process.
4. Save the sample.

This information can help identify whether the process is repeatedly attempting to index contacts, mail, or another data source.

If the issue persists

Since you've already:

• Force quit peopled.
• Restarted the Mac.
• Reset Spotlight indexing.
• Restarted contactsd.

…and the issue has continued for two days, it is likely either:

• A corrupted contacts or account database.
• A software regression introduced in macOS Tahoe 26.4.

In this case, contact Apple Support and provide the peopled process sample along with a description of the issue. This will help determine whether further troubleshooting is required or if the behavior is related to a known issue that will be addressed in a future macOS update.

Here is some info from my investigation. I can't resolve it because I am unable to fix the sandbox violation.

• OS build: macOS 26.4.1 (25E253)
• Symptom: peopled ~70% CPU, CallHistorySyncHelper ~28% CPU
• Root cause: Kernel sandbox violation — deny(1) file-read-data ~/Library/Application Support/CallHistoryDB — because com.apple.peopled.sb only allows (home-subpath "/Library/CallHistoryDB/"), not the actual runtime path
• Evidence: kernel log showing 22,000+ duplicate deny reports per few seconds

I hope Apple fixes this soon. I am going to try turning off contacts sync.

This peopled CPU spike issue seems to be a widespread bug in Tahoe 26.4 that Apple is still working on. Based on what's been shared in this thread, here's a quick summary of what's worked for people so far:

If you haven't tried it yet, the most common fix is turning off iCloud Contacts sync:

- Go to System Settings > Apple Account > iCloud > Saved to iCloud

- Turn off the Contacts toggle

- Reboot your Mac

Asmodael above had success with this, though as Ivan-NTS mentioned, it doesn't work for everyone. Another user found that removing the Contacts widget from Control Center solved the problem for them.

If those don't help, you could try signing out of your Apple ID/iCloud account completely and signing back in. That sometimes clears up stuck sync processes.

Since this started right after the 26.4 update, it might also be worth checking if there's a 26.4.1 or later patch available. The peopled process handles contacts and people data, so it's likely tied to the iCloud Contacts sync feature having issues in this version.

In the meantime, force quitting peopled from Activity Monitor should at least give you some temporary relief until the process spikes again.

I've had the same issue since updating to Tahoe 26.4. One fix I found that worked for me.

In notifications center I had a contact widget.

1. Turn off contacts sync in iCloud settings
2. Reset the computer
3. Remove the contacts widget from notifications center
4. Enable contacts sync in iCloud settings.

Finally peopled is working normally:

Otherwise it would get crazy CPU usage about 70% all the time causing my laptop to be unusually warm when doing basically nothing not to mention the battery drain.

Looks like this might be a bug in the widgets introduced in Tahoe 24.1. Frankly surprising how this slid past regression testing, but I guess this is Apple's software quality for you in this day and age.

I looked at this with Opus 5.0, and it seems that my Call History DB is uncorrupted although there was a large WAL portion of uncommitted changes. The `peopled` daemon busy loop is "an unguarded pokeSyncHelperToInitDB ↔ createPermanent retry loop in CallHistory.framework", but why the daemon ends up there was a bit of a mystery.

The trace seems to lead to ContactWidget, I just have no such an widget active. But I _have_ used the widget gallery, so it seems that the previews in the widget gallery register a refresh task that is never cleaned up and that then periodically launches the peopled. (Evidence supporting this includes some time stamps and the fact that the parent of the peopled in one case at least was Notification Center, not chronod.)

I'm maybe too lazy to do a definite test, that is, to purposefully open the widget gallery after a reboot vs. not, etc. Instead, I'll just avoid the widget gallery after my reboot, let's see if the daemon stays inactive. If not, I'll set up a watchdot that kills it if it appears.

So, in short: get rid of all contact widgets, reboot, don't open the widget gallery, hope for the best.

Nice sleuthing everyone. Removing any Contacts widgets (and restarting) was the solution for me as well.

1. I tried clearing the quarantine flags on the CallHistoryDB.
2. I tried deleting the CallHistoryDB. It was recreated after restart and the issue continued.
3. I removed all calls from Phone.app on the iPhone and the Mac.

But of course, I had a Contacts widget on my Desktop. After removing that, AND RESTARTING, the problem is gone.

Disabling Contact sync is a nuclear option I would not recommend. Just get rid of the darn widget(s).

I had encountered the same issue. Peopled was causing excessive CPU usage and rapidly depleting the battery on my M5 Macbook Pro 32GB/1TB. I think that I might have found a temporary workaround:

1. 1. Open Activity Monitor
   2. Navigate to the CPU tab.
   3. Select the peopled process and the 'x' button and "Force Quit" the process.
   4. Open "System Settings"
   5. Click on the "Apple Account" item at the top of the list.
   6. Select iCloud.
   7. Look for the "See All" button under the "Saved to iCloud" heading and click it.
   8. Turn off the "Contacts" option within the "Saved to iCloud" form. (Please note that your contacts will no longer sync from your Mac to iCloud.)
   9. Reboot.

I've found that this process has kept the peopled process from coming back. My Macbook Pro now runs cooler and the battery life jumped considerably.

Well noticed about quarantine. Did all the steps, but for me it didn't help. It removes quarantine flag for sure and kill command restarts hanged peopled process for sure. But after 'contacts tab' in 'edit widgets' dialogue opened peopled process starts it all over again. With the same logs in console:

default	13:34:56.813363+0300	CallHistorySyncHelper	CallDBManagerServer: fetching permDBURL
default	13:34:56.813389+0300	CallHistorySyncHelper	CallDBManagerServer: permDBURL: <private>
default	13:34:56.813394+0300	CallHistorySyncHelper	BootUp result: kBootUpResultUsePerm
default	13:34:56.813422+0300	peopled	Boot up result: kBootUpResultUsePerm
default	13:34:56.813436+0300	peopled	CallDBManagerClient: fetching permDBURL
error	13:34:56.813498+0300	peopled	failed to open /Users/pchigrin/Library/Application Support/CallHistoryDB: 1 (Operation not permitted)
default	13:34:56.813510+0300	peopled	CallDBManagerClient: permDBURL: (null)
default	13:34:56.813518+0300	peopled	CallDBManagerClient: fetching permDBURL
error	13:34:56.813558+0300	peopled	failed to open /Users/pchigrin/Library/Application Support/CallHistoryDB: 1 (Operation not permitted)
default	13:34:56.813570+0300	peopled	CallDBManagerClient: permDBURL: (null)
error	13:34:56.813577+0300	peopled	Permanent database file doesn't exist; poking sync helper. Error code: kDBLNoSuchFileError
default	13:34:56.813585+0300	peopled	Poking sync helper to initialize data store

abimanyu_nk wrote:

Hello,

I recently updated my MacBook Air (M4) to macOS Tahoe 26.4. Since the update, my system has been experiencing unusually high CPU usage from the process peopled, which consistently stays around 70–80% CPU• .

Because of this, the MacBook becomes noticeably warm and the battery drains much faster than normal, even when the system is mostly idle.

I have already tried the following troubleshooting steps:

Force quitting the peopled process using Terminal (killall -9 peopled)

• Restarting the Mac

• Resetting Spotlight indexing

• Restarting related services such as contactsd

However, the issue persists and peopled remains the top CPU-consuming process in Activity Monitor.

This problem started immediately after updating to macOS Tahoe 26.4 and has continued for the past two days.

Could you please help diagnose and resolve this issue?

re: < /System/Library/PrivateFrameworks/People.framework/peopled >

Tahoe 26.4 ex. for comparison only

If you shut down and restarted more than once since your update

>Shut Down

Than try A SafeBoot Start up your Mac in safe mode - Apple Support will sort many anomalies

Does a quick disk repair before it fully boots up, and certain system caches get cleared and rebuilt, third party system modifications and system accelerations are disabled temporarily.

Login and test. Reboot as normal and test. Caches get rebuilt automatically.

My understanding peopled—manages information related to contacts, specifically analyzing or indexing people-related data.

Are you sync Contacts via iCloud(?) You can try signing out od AppleID/Account and sign back in...

if no resolve—

Sign out of your Apple ID/Account /iCloud on the Mac and back in again and compare your results

Sign in to or out of your Apple Account on Mac - Apple Support

If you are still facing the issue so i got the main root cause. you can do following options to fix this-

1. go to your setting find the accounts and disconnect/remove the google account or from the contact application remove the google account connection.
2. Reset you macbook , make sure to take the backup of the mac. this do the complete setup

I resetted my macbook and completed the setup and then i connected my google account at contact to see the contacts and the same problem started coming again so i disconnected my google account from contact and internet accounts as well

It took me a long time to figure this out, too.

After a lot of trial and error, however, I discovered that the problem was caused by the Contacts widget in my Control Center.

After I deleted the widget from the desktop/Control Center, everything was fine again.

really strange problem!

> But, I did notice this problem after I reorganized the widgets in Control Center?!

Opening the Widget editor is enough to re-trigger the problem. Restart after doing so.

dineth290 wrote:

For me it is not letting me toggle off the contact sync in iCloud settings , so does anyone have any solutions for this. BTW i have an M4 macbook air 15 inch

No issues Turning off Contacts System Setting > User account > iCloud > See All >> Saved to iCloud >>> Contacts

Using 1 M2 Desktop, 1 M4 Desktop and 1 M5 Laptop

All 3 machines using Tahoe 26.4.1 ( current version as of this writing )

For me it happens when I even open side bar widget management and go to contacts tab.

Immediately after peopled starts circular CPU load.

That's what console logs have.

I tried to remove databases for recreate (it was done successfully) CallHistoryDB `rm -rf ~/Library/Application\ Support/CallHistoryDB` and Knowledge `rm -rf ~/Library/Application\ Support/Knowledge`.

After reset all permissions `tccutil reset All`.

So far no luck.

[Edited by Moderator]