Internet issues on Monterey

I experience the same issues immediately after upgrading to Monterey. The connection (wifi or Thunderbolt) dies after a few minutes of inactivity on the laptop. Restart so far is the only way to temporarily fix the connection. And then the connection dies again after a few minutes. Interestingly, on this very same laptop, when I run boocamp Windows, the connection has no issues. (Note, I have not upgraded my MacBookPro16,1 to Monterey because of the these issues)

My MacBook Pro spec:

Model Name: MacBook Pro

 Model Identifier: MacBookPro14,1

 Processor Name: Dual-Core Intel Core i5

 Processor Speed: 2.3 GHz

 Number of Processors: 1

 Total Number of Cores: 2

 L2 Cache (per Core): 256 KB

 L3 Cache: 4 MB

 Hyper-Threading Technology: Enabled

 Memory: 16 GB

 System Firmware Version: 447.40.12.0.0

 OS Loader Version: 540.40.4~45

 SMC Version (system): 2.43f11

Some info of running networking commands when the laptop gets into the bad state:

When MacBooK Pro has lost the internet connection:

$dig bbc.com

; <<>> DiG 9.10.6 <<>> bbc.com

;; global options: +cmd

;; connection timed out; no servers could be reached

$traceroute bbc.com

traceroute: unknown host bbc.com

After restarting the MacBooK Pro:

$dig bbc.com

; <<>> DiG 9.10.6 <<>> bbc.com

;; global options: +cmd

;; Got answer:

;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 17653

;; flags: qr rd ra; QUERY: 1, ANSWER: 4, AUTHORITY: 0, ADDITIONAL: 1

;; OPT PSEUDOSECTION:

; EDNS: version: 0, flags:; udp: 512

;; QUESTION SECTION:

;bbc.com. IN A

;; ANSWER SECTION:

bbc.com. 300 IN A 151.101.128.81

bbc.com. 300 IN A 151.101.192.81

bbc.com. 300 IN A 151.101.64.81

bbc.com. 300 IN A 151.101.0.81

;; Query time: 28 msec

;; SERVER: 2001:4e8:0:4009::11#53(2001:4e8:0:4009::11)

;; WHEN: Mon Dec 13 12:04:46 PST 2021

;; MSG SIZE rcvd: 100

$traceroute bbc.com

traceroute: Warning: bbc.com has multiple addresses; using 151.101.192.81

traceroute to bbc.com (151.101.192.81), 64 hops max, 52 byte packets

1 10.0.0.1 (10.0.0.1) 4.076 ms 1.646 ms 1.018 ms

2 ********* 9.823 ms 9.497 ms 9.598 ms

3 rc1st-be102-1.vc.shawcable.net (64.59.151.85) 10.825 ms 10.051 ms 14.987 ms

4 24.244.59.101 (24.244.59.101) 10.875 ms 12.245 ms 9.545 ms

5 rc1st-be214.vc.shawcable.net (24.244.58.65) 10.235 ms 9.599 ms 10.081 ms

6 rc2wt-be50-1.wa.shawcable.net (66.163.70.106) 16.275 ms 18.884 ms 14.432 ms

7 * * *

8 * * *

9 * * *

[Personal Information Edited by Moderator]

^^^^ Thanks for posting this! It appears this is "the fix" for me as well.

Once I made inactive the second (unused) instantiation of "Cisco AnyConnect Socket Filter" I have enjoyed no loss of DNS or Broadband connectivity for the last 24 hours. this is my longest unbroken connection since I installed Monterey OSX a couple of months ago.

I'll post with any further updates, but greatly appreciate all who contributed to this thread and all the diagnosis here.

My Best,

FightingIbis

by the way, most of the udp (open)sockets are created by macos mdnsresponder not AnyConnect. If you use 'netstat -anxv | grep -e udp' you can see the corresponding PID for each UDP socket, then you can have a look at your process list to check what that PID is for.

example for PID 309:

udp4 0 0 *.58521 *.* 786896 9216 309 394 0x0103 0x00010000

_mdnsresponder 309 0.0 0.0 33892620 10516 ?? Ss 8:26am 0:13.32 /usr/sbin/mDNSResponder

Wild that this is still a problem. Completely cleaned Macbook Pro 2015, and every time I upgrade to the latest Monterey update this issue persists. Maybe adjusting some other settings will fix it, but I don't have the patience to troubleshoot all that when I already know that restoring Big Sur fixes the problem.

I am using latest Cisco AnyConnect VPN and noticed that upon applying macOS 12.3 update, there have been some issues with handling VPN split tunnel and domain suffix mapping. Local DNS and access works without issue, when connecting to VPN, a local ping or Browser URL do not resolve, however, if you use DIG command or NSLOOKUP command they work.

Discovered macOS is receiving the updated DNS Server and Domain suffix and it's appearing alongside the local entires, but it's no longer mapping the domain suffix correctly and directing traffic through the tunnel properly. Had to override with entries in the local /etc/hosts file temporarily as I need to use this for connecting to work.

Discovered the problem occurs on both M1 Max and Intel based Mac systems using 12.3 update. When the system was running on older version such as 12.1, 12.2.x, was not having any issue and version of the Cisco AnyConnect VPN has remained the latest in all conditions. Starting to believe it's something with macOS Monterey update and not the Cisco AnyConnect VPN as if I run from an earlier version of 12.x - VPN works fine.

Have you checked your connection right after flushing the cache? I'm still on 12.3 due to concerns about this issue, but I have a cron job set up to flush my cache once a minute which has resolved the issue for me most of the time. However, after a brief period, the cache becomes corrupt again and DNS queries again fail, thus the cron job...

I have the same issue, problem lies on dns inability to resolve a hostname, and so far the only fix is to reboot the laptop. macos version 12.0.1, Macbook pro 13-inch 2020.

Gotcha! I listed my solution in my reply to OP. I created a new network location and it fixed the upload speed issue for me.

Same problem. Have done everything that has been suggested, and it is exactly as you have said, when I restart it works (I initially did in safe mode, then realised it didn't matter, any restart works), but only for a while, doing a Wireless Diagnostic identifies the issue as a DNS issue, have deleted the DNS cache, set up a new network, used different DNSs, removed hide IP, I can manage to fix it by restarting only, which is a major pain. I have updated on Monterey on 2 other machines and it's fine on 2 of the machines (MBP and another iMac), but this iMac simply loses its DNS. Any other answers?

Glad to see it's not only me. The fact that it's not happening to everyone on Monterey is good to know, but only part of the picture. It may still turn out to have another cause other than macOS Monterey, but I think enough evidence has been presented to not confidently dismiss it as having nothing to do with macOS.

I've been using an M1 Mac Mini running Big Sur for at least six months without any network issues. Within the last several days since upgrading to Monterey, I've suddenly had two or three incidents of inability to reach Internet resources due to DNS errors. And these are sites I've used hundreds of times.

After a few hours of trouble-free operations, my default browser (Edge) will suddenly be unable to resolve any DNS names. I then test multiple websites and browsers (Safari and Chrome) and other Internet connected apps—all with the same DNS failures. Following this, I even tried flushing DNS cache and multiple alternative DNS providers all to no avail. The only thing that has instantly resolved the issue for me (at least for quite a few hours) is restarting macOS.

Several others have already chimed in here with the same problem, so it seems worthy of further investigation.

Mine also occurred with both ethernet as well as with wifi. No, I wasn't using any ad-blocking other than the normal Safari preferences. The issue occurred regardless of browser. But in my case all resolved after restarting over several days.

Yesterday I performed a full uninstall on Cisco AnyConnect and reinstalled latest version (that my employer would provide, anyway) Version 4.10.03104

Also upgraded OSX to V12.2

I booted up my Readdle Spark application (email) and and got about 10 minutes of work until I lost Broadband connection as per usual. So 12.2 doesn't seem to solve it. I may try an uninstall of Cisco AnyConnect and see if that works...

Same old, same old... Not fun.

Best,

FightingIbis

I have the similar issue, My Mac dns resolver is broken every 3 days after upgrading to Monterey. the nslookup command shows the error:

nslookup www.google.com.tw

nslookup: isc_socket_bind: address not available

It turns out you're right. The UDP sockets aren't limited to the ephemeral ports; they can use all unrestricted ports. But port exhaustion does not seem to be the issue anyway. It's some other resource that's being exhausted, although I still haven't figured out which one. The low-level error that comes back when exhaustion has occurred is EADDRNOTAVAIL, which can be the result of several problems.

However, having a number of 93966 (and increasing) is a reliable indication that the problem is occurring on your system and that eventually, some resources will be exhausted, requiring a reboot. I have a healthy BSD server running, and that number always remains below 100.

I'm still experiencing this issue once per week on macOS 12.1 and Cisco AnyConnect 4.10.04071.

Hi, in addition to previous comments.....

I have found with my 2015 MBP if I am very close to the router wifi goes doo-lally.

After this episode either in Big Sur or Monterey, sleep or reboot required to return connection.

I have changed the hardware, no change to result.

Windows Bootcamp has NO problems with this!!! Nor any Linux I've booted from.

I've looked at the drivers in Big Sur and Monterey, they are so similar.

See pics I hope.

The only real difference I see is in the line for 'Firmware'.....???