CPU 100% every few days

[TIm Fitz]

Have an AC68U with the latest 384.8 firmware and every few days I get CPU 1 stuck at 100%. Power reset fixes but is only temp solution. This has happened on the old version of firmware and I keep hoping a new version will correct the issue. No USB Devices, No AI Cloud or VPN's in use. We do use the Trend Micro stuff.

 

[wesbez]

Any 4K iptv. I had a issue with BT tv’s 4K channel doing this.

I found that if I didn’t change channels it would still be delivering the data to the BT box. (UK provider)

 

[TIm Fitz]

No 4K TV's in use at all.

 

[indark]

power cycle and clear nvram might help, at least that's what worked for me

power off the router for 30+ sec
hold wps button while powering on
wait till the power led start flashing/blinking
release the wps button and wait for the router to fully reboot properly
now reconfigure everything manually from scratch.

good luck

 

[Veky]

Adaptive QoS on?
HW acceleration on?

 

[Ronald Schwerer]

Just in case you are running the 68P CFE (1.0.2.5), go back to the 1.0.2.0 US AiMesh CFE.

 

[Diamond67]

Have an AC68U with the latest 384.8 firmware and every few days I get CPU 1 stuck at 100%.

When this happens, can you connect to your router IP using SSH, for example with PuTTY?

 

[ARol]

It is very similar to my problem with jffs freezes
Try to read this thread

 

[Martineau]

Have an AC68U with the latest 384.8 firmware and every few days I get CPU 1 stuck at 100%. Power reset fixes but is only temp solution. This has happened on the old version of firmware and I keep hoping a new version will correct the issue. No USB Devices, No AI Cloud or VPN's in use. We do use the Trend Micro stuff.

A lot of users appear to now suffer CPU core saturation, so I thought I would share my v384.8 99% CPU overload experience.

NOTE: Many times users have been advised to use TOP/HTOP to identify which process(es) are causing the issue.
Sadly in my experience rarely do any individual tasks show using more than >10%, and a quick visual 'sum' of the CPU total of all of the visible tasks will usually only show <20% in use yet HTOP (and NetData) report anything between 50%-99% saturation.

Fortunately in my case, the CPU saturation (when by chance I happened to notice it) appears to only last say 8-10mins and isn't 'common' (it may have happened two or three times although I reboot quite often when playing with scripts etc.) and always subsides/clears itself.

So in my environment I suspect dnmsasq and possibly cURL (and my old SanDisk Ultra USB 3.0 flash drive?) somehow together cause the perfect storm?
(dnsmasq updated in v384.8)


TL;DR



Firmware v384.8/v384.8_2 re-flashed and full Reset to Factory Default and manual re-configuration.

I recorded HTOP to be able to examine the results.
Unfortunately the full video shows cURL ('disk sleep?') persistently trying to send an email using my real email details, so I have taken three consecutive snapshots and redacted the cURL email details:

Sorry if the images are slightly too small, but my point is that video images do not lie! - well these don't..not smart enough to doctor the footage!

NetData overview..although I did decide to kill the (5?) cURL tasks but not sure if this manual intervention influenced the return to normal CPU usage.

HTOP screen refresh occurs either every second or every two seconds, so the frames (not HH:MM but represents 1 day and MM:SS since reboot)

10:34,10:35 and 10:37...show dnsmasq using 66.5%,72.1% and 72.3% respectively

and a minute later......still going strong!
11:40...................show dnsmasq using 98.7%
which may be OK since HTOP shows that CPU1 is now down to 65.4% and CPU2 is down to 47.1% rather than both showing in the mid 90%​

 

[avi68]

Hi all,

I have been having this issue for a while on my rt-ac68u. I am on 384.8, Ive noticed that the core 1 under the cpu gets stuck at 100 % after every couple of weeks. I do have a mains powered 8tb desktop hard drive connected. I created an SSH session and ran the "top" command and got this:

CPU: 0.7% usr 2.6% sys 0.0% nic 48.0% idle 44.9% io 0.0% irq 3.5% sirq
Load average: 5.13 4.99 4.63 1/124 22602
PID PPID USER STAT VSZ %VSZ CPU %CPU COMMAND
18931 1 admin S 928 0.3 0 2.4 miniupnpd -f /etc/upnp/config
230 229 admin S 5176 2.0 0 1.0 nt_monitor
265 229 admin R 5176 2.0 1 0.7 nt_monitor
9272 1 admin S 10672 4.1 0 0.3 /usr/sbin/lighttpd-monitor
301 1 admin S 9240 3.6 1 0.1 networkmap --bootwait
212 1 admin R 2060 0.8 1 0.1 protect_srv
22586 21940 admin R 1424 0.5 1 0.1 top
36 2 admin SW 0 0.0 1 0.1 [kworker/1:1]
9271 1 admin S 14892 5.8 1 0.0 /usr/sbin/lighttpd -f /tmp/lighttpd.conf -D
9324 1 admin S 11048 4.3 1 0.0 /usr/sbin/lighttpd-arpping -f br0
11202 11201 admin S N 10488 4.1 0 0.0 minidlna -f /etc/minidlna.conf -r
11199 1 admin S 10488 4.1 1 0.0 minidlna -f /etc/minidlna.conf -r
11201 11199 admin S 10488 4.1 1 0.0 minidlna -f /etc/minidlna.conf -r
276 1 admin S 6132 2.4 1 0.0 httpd -i br0
29450 1 admin S 6080 2.3 1 0.0 /usr/sbin/smbd -D -s /etc/smb.conf
911 398 admin S 5972 2.3 1 0.0 mastiff
398 322 admin S 5972 2.3 0 0.0 mastiff
322 1 admin S 5972 2.3 0 0.0 mastiff
399 398 admin S 5972 2.3 1 0.0 mastiff
29449 1 admin S 5872 2.2 0 0.0 nmbd -D -s /etc/smb.conf
1 0 admin S 5604 2.1 1 0.0 /sbin/preinit
714 1 admin D 5532 2.1 0 0.0 disk_monitor
280 1 admin S 5416 2.1 1 0.0 watchdog
206 1 admin S 5404 2.1 0 0.0 /sbin/wanduck
831 1 admin S 5400 2.1 1 0.0 bwdpi_wred_alive
351 1 admin S 5400 2.1 0 0.0 erp_monitor
514 1 admin S 5400 2.1 1 0.0 usbled
666 1 admin S 5400 2.1 0 0.0 ntp
867 1 admin S 5400 2.1 0 0.0 pc_block
324 1 admin S 5400 2.1 1 0.0 hour_monitor
323 1 admin S 5400 2.1 1 0.0 bwdpi_check
225 1 admin S 5400 2.1 1 0.0 wpsaide
91 1 admin S 5396 2.1 1 0.0 console
232 1 admin D 5328 2.0 1 0.0 nt_center
243 232 admin S 5328 2.0 1 0.0 nt_center
244 243 admin S 5328 2.0 1 0.0 nt_center
211 1 admin S 5176 2.0 1 0.0 nt_monitor
229 211 admin S 5176 2.0 1 0.0 nt_monitor
453 452 admin S 3940 1.5 0 0.0 cfg_server
452 363 admin S 3940 1.5 1 0.0 cfg_server
admin@RT-AC68U:/tmp/home/root#

Does anyone have an idea as to whats happening?

Just to let you know that I have reset to factory previously on previous firmwares but I still seem to get this issue.

Thanks all

 

[Aron Hoekstra]

I have been having this issue as well on my new RT-AC3200. For me it occurs whenever I plug in my 5TB WD USB hard drive, even if I've disabled media sharing and SMB. This happens for both stock ASUS firmware & on Merlin.

I now understand that the max partition limit is 4TB. I am going to try repartitioning the hdd and try again.

 

[Aron Hoekstra]

I now understand that the max partition limit is 4TB. I am going to try repartitioning the hdd and try again.

Repartitioning seems to have helped.

 

[consorts]

Repartitioning seems to have helped.

despite being a data hoarder, i keep all my partitions at 2tb
and try to folder nest any file collection over 250 entries.
because i know from experience even the latest soft versions
can struggle dealing with what should be legacy limitations.