100% cpu usage on core 1 of AC88U

[Wagnard]

Same issue here. I'm on Official ASUS firmware: 3.0.0.4.384_20308
A complete PowerOFF must be done to fix it. Reboot is not enough. I think the reboot command is not even passing anyway.
On mine, the problem happens when I SAVE after I changed something to the 2GHZ wireless configs. (not tested to 5GHZ yet)
on TOP, user/sbin/acsd seems to be the culprit.

 

[captblaze]

Hello,

I've been using the Merlin firmware for my router for quite a while now, and just today I noticed that core 1 on my Asus AC88U is pegged at 100% all of the time, and restarting the router does not solve the issue. This must have started in the last week, because I updated to the newest firmware about a week ago, and the router was not having this issue.

Is there a way to see what is using the processor in the router (similar to task manager)? The only thing I've done recently was install Hyper-V on my home server and create a vm that runs bind9 and gnix to store copies of steam downloads so that they can be redownloaded from the server instead of the internet. However turning off the VM doesn't solve the issue, so I don't think that's the problem.

Any other info I can give to help out, please let me know. I'm not sure if it's affecting the performance of the router, but it is definitely running very warm.

 

[dsring]

Same issue here. I'm on Official ASUS firmware: 3.0.0.4.384_20308
A complete PowerOFF must be done to fix it. Reboot is not enough. I think the reboot command is not even passing anyway.
On mine, the problem happens when I SAVE after I changed something to the 2GHZ wireless configs. (not tested to 5GHZ yet)
on TOP, user/sbin/acsd seems to be the culprit.
View attachment 11821

I have tested and the only thing that triggers the problem is a change that causes the 2.4 GHz radio to restart (such as a wireless channel change, or toggling the WPS button ON or OFF). 5 GHz changes do not trigger it. And the only way to clear it is to power cycle the router. A reboot will not fix it. I wonder if the problem is unique to a certain hardware revision of the RT-AC88U / 3100. A quick test to see would be for 88U or 3100 owners to toggle WPS ON or OFF and monitor to see how CPU usage reacts.

 

[netguru]

Same to me, acsd is causing nearly 50% ..
Toggling WPS to off, acsd is gone ...

 

[Klueless]

Noticed my TV "buffering" ... something new to me since I got my Asus 1900P (Merlin 380.68_4). Logged into my Asus GUI and noticed CPU 1 flat lined at 100%. Hit reboot, never came back so I powered on/off. Router came back but still at 100%. Tried again, same thing.

Then I remembered this thread so I pulled it up. SSH? TOP? Tried telnet to my router but got an error msg. Oh well ... Klueless I am.

In any case I got lucky. Noticed something funny under Internet status on the home (network map) page, I forget exactly what now, it showed "connected" alright and I had Internet, etc. but there was "something". Anyway that led me to power off the modem (and router). Powered up modem first, waited for it to settle and then powered up the router. CPU was now normal!

CPU's been normal for a few days now and no more "buffering" ... even with three TVs steaming and grandson's Xbox going. Hope it stays that way because I really have no idea.

 

[bsdsource]

If you don't know how to telnet into your router I've provided basic instructions below. First enable telnet from the router webui. Navigate to:

Administration --> System --> select enable telnet (then save settings)

You can download putty telnet client from here:

putty

1. start putty
2. enter your router IP address into the IP address box in putty
3. select the open button at the bottom of putty (you should now be connected to your router)
4. at router command prompt enter user name (same name you use to login to your router webui)
5. at router command prompt enter password (same password you use to login to your router webui)
6. you should now be logged into your router and can use linux commands
7. use the "top" command (stats should show up showing processes, cpu, memory etc usage)
8. take a screenshot if you can or copy and paste the info. preferably screenshot.
9. return to this forum and provide the info you obtained from the top command

 

[Klueless]

If you don't know how to telnet into your router I've provided basic instructions below. First enable telnet from the router webui ...

Thank you. Will try "later" ... when I'm a little less gun shy ... after many years of "set it and forget it" I suddenly realize how little I really know!

BTW: Any idea what kind of stuff Core 1 is responsible for versus what tasks Core 2 performs?

 

[bsdsource]

TW: Any idea what kind of stuff Core 1 is responsible for versus what tasks Core 2 performs?

The Linux kernel scheduler and taskset are typically responsible for that. Also a program needs to be coded to use multi-threading which will use more than one core at a time. It appears acsd is a single threaded process.

 

[Klueless]

The Linux kernel scheduler and taskset is typically responsible for that. Also a program needs to be coded to use multi-threading which will use more than one core at a time. It appears acsd is a single threaded process.

My first thought was, "Gee, how disappointing". I woulda thunk they'd at least run esoteric tasks like file share or VPN on Core 2?

So I logged into my Asus GUI and looked at core utilizations. Ran an Internet speed test (from my 5GHz radio) and was pleased to see both cores participating. Then I copied a large file from my Asus file share; Core 2 was pegged at 100% with Core 1 jumping in and out.

 

[Mete]

Same issue on my side with a new AC88u (acsd starts hogging the CPU) when I change wifi settings. The only way to fix is power off and power on the router. There is firmware dependency, I see the problem of 100% utilization with 382 and 384 firmwares, specifically I tried these versions:
3.0.0.4.384_20379
3.0.0.4.382_18991
3.0.0.4.382_15850

I did the NVRAM erase, restart, reconfig couple times, it did not help.

I do not see any issue with 380 firmware (3.0.0.4.380.7743).

I noticed that if I disable either the 2.4G or the 5G radio, the router starts working fine again with 382 and 384 firmwares...

Alternatively if I disconnect all the GE ports except the WAN interface, then I could change the Wifi parameters (e.g. WPS enable/disable) without hitting high utilization or lost Wifi while both bands are up with 384 firmware (I have not tried 382)...

My HW rev is A6. It might be wild guess right now but maybe there is some power issue in the circuit...

 

[rodney reed]

Same issue here. 3.0.0.4.384_20379. /usr/sbin/acsd

 

[Mete]

Same issue on my side with a new AC88u (acsd starts hogging the CPU) when I change wifi settings. The only way to fix is power off and power on the router. There is firmware dependency, I see the problem of 100% utilization with 382 and 384 firmwares, specifically I tried these versions:
3.0.0.4.384_20379
3.0.0.4.382_18991
3.0.0.4.382_15850

I did the NVRAM erase, restart, reconfig couple times, it did not help.

I do not see any issue with 380 firmware (3.0.0.4.380.7743).

I noticed that if I disable either the 2.4G or the 5G radio, the router starts working fine again with 382 and 384 firmwares...

Alternatively if I disconnect all the GE ports except the WAN interface, then I could change the Wifi parameters (e.g. WPS enable/disable) without hitting high utilization or lost Wifi while both bands are up with 384 firmware (I have not tried 382)...

My HW rev is A6. It might be wild guess right now but maybe there is some power issue in the circuit...

I figured out that my issue was related to a Netgear 3700 device that was being used as an access point at another place in the house. This AP was connected via Ethernet cable to my router. For some reason AC88U and Netgear 3700 do not seem to get along well (at least in my configuration). Once I removed the access point from my network AC88U started working fine... No more 100% utilization after wifi parameter change...

 

[Mete]

I figured out that my issue was related to a Netgear 3700 device that was being used as an access point at another place in the house. This AP was connected via Ethernet cable to my router. For some reason AC88U and Netgear 3700 do not seem to get along well (at least in my configuration). Once I removed the access point from my network AC88U started working fine... No more 100% utilization after wifi parameter change...

I contacted to ASUS and below is the reply I got after explaining the high CPU utilization:
....

I appreciate your patience, I sincerely apologize for the inconvenience. I have just receive an update from our higher level of support we have confirmed it is a chipset defect device. We are recommending you to send in the product for repair.

Kindly provide the serial number of your RT-AC88U for me to verify its warranty status
....​

Since I removed the Netgear 3700 from my network, I do not see the issue any longer. At this point I am having hard time to believe that this is HW issue but, I could still return the router back to the place where I bought... Does anybody have any more insight into this high CPU utilization?

 

[curlysir]

I have not had the problem since I installed 384.3. Did not do anything else.

 

[David Potter]

Hello,

I've been using the Merlin firmware for my router for quite a while now, and just today I noticed that core 1 on my Asus AC88U is pegged at 100% all of the time, and restarting the router does not solve the issue. This must have started in the last week, because I updated to the newest firmware about a week ago, and the router was not having this issue.

Is there a way to see what is using the processor in the router (similar to task manager)? The only thing I've done recently was install Hyper-V on my home server and create a vm that runs bind9 and gnix to store copies of steam downloads so that they can be redownloaded from the server instead of the internet. However turning off the VM doesn't solve the issue, so I don't think that's the problem.

Any other info I can give to help out, please let me know. I'm not sure if it's affecting the performance of the router, but it is definitely running very warm.

 

[David Potter]

I have had this same problem several times now with RT-AC5300, Reboot does not solve the problem but power cycle does resolve. In my case it was after making AI change to allow WiFi access for 1 client (parental controls). From the posts this seems to be a problem that has been around for awhile. Hope someone can fix ......

 

[Aaron Freedman]

Similar issue with a brand new ac88u on the latest firmware. All of a sudden internet slowed to a halt, 2.4ghz band disappeared. I did a manual power down and back up except my 5 and 2.4 bands turned on and off a few times then eventually disappeared. No wifi at all now..

Just out of curiosity - do any of you use Google Homes or Home Minis?

 

[firebladeRR1k]

Contact Asus support via the feedback link at the bottom. This may be a known issue. I was experiencing exactly this same issue and they were able to help.

 

[Haraldinho]

Contact Asus support via the feedback link at the bottom. This may be a known issue. I was experiencing exactly this same issue and they were able to help.

Hey firebladRR1k, can you share how Asus support was able to help? Did they suggest some parameter change or did you have you your unit replaced?

@RMerlin, were you able to distill a root cause out the supplied top overviews? As many over here, I am experiencing the same. I saw the acsd process is scanning for low-interference wifi channels. Can it have anything to do having the control channel setting being on "Auto"?

Curious issue.

 

[RMerlin]

Hey firebladRR1k, can you share how Asus support was able to help? Did they suggest some parameter change or did you have you your unit replaced?

@RMerlin, were you able to distill a root cause out the supplied top overviews? As many over here, I am experiencing the same. I saw the acsd process is scanning for low-interference wifi channels. Can it have anything to do having the control channel setting being on "Auto"?

Curious issue.

No, because the vast majority of users never reported back with their ntop output. A few did, quite often it showed IO activity, which meant most likely network or USB traffic loading their CPU.

But in general, no, there's no common cause, just different cases.