ASUS RT-AC68U Firmware version 3.0.0.4.384_81039

[Smokindog]

Mine looks pretty similar on both APs. I even changed the scan to 1 second and nothing popped out at me.

It is odd that if I switch to a different browser tab for a period of time and then come back the spikes are much lower for the time the tab did not have "focus". I'm stumped but that doesn't surprise me

Here you go:
View attachment 19106
81039: clearly httpd, the Apache HTTP server takes most CPU time, also explaining the slugish GUI response.


View attachment 19107
45717: nt_center is the top, but others are close.

 

[RMerlin]

Here you go:
View attachment 19106
81039: clearly httpd, the Apache HTTP server takes most CPU time, also explaining the slugish GUI response.


View attachment 19107
45717: nt_center is the top, but others are close.

According to top, nothing uses more than 4% of CPU, and your load average is only 0.2 - nothing abnormal there. And if you switch to a page that does not redraw the CPU/RAM info every few secs, that httpd CPU load will drop to below 1%.

Your nt_center uses 0.7% CPU - that's not even noticeable.

 

[OzarkEdge]

According to top, nothing uses more than 4% of CPU, and your load average is only 0.2 - nothing abnormal there. And if you switch to a page that does not redraw the CPU/RAM info every few secs, that httpd CPU load will drop to below 1%.

Your nt_center uses 0.7% CPU - that's not even noticeable.

81039 on my 86U pegged CPU2 at 100% on the status page, multiple times in succession... this after a clean install and before much configuration or usage. The graphs were crazy busy with linework.

OE

 

[wouterv]

According to top, nothing uses more than 4% of CPU, and your load average is only 0.2 - nothing abnormal there. And if you switch to a page that does not redraw the CPU/RAM info every few secs, that httpd CPU load will drop to below 1%.

Your nt_center uses 0.7% CPU - that's not even noticeable.

I disagree, the shown numbers are taken with certain intervals, also "top" showed now and than much higher numbers than I could capture, the real peak values are not shown.
I cannot believe you are trying to say that the spikes different people see in the CPU graph and she sluggish GUI are absolutely normal while older firmware versions do not show that behavior.

 

[wouterv]

Mine looks pretty similar on both APs. I even changed the scan to 1 second and nothing popped out at me.

It is odd that if I switch to a different browser tab for a period of time and then come back the spikes are much lower for the time the tab did not have "focus". I'm stumped but that doesn't surprise me

View attachment 19108

What you see can be explained by the fact it is the HTTP server of the router that seems to have an issue by presenting the GUI.

 

[RMerlin]

I disagree, the shown numbers are taken with certain intervals, also "top" showed now and than much higher numbers than I could capture, the real peak values are not shown.
I cannot believe you are trying to say that the spikes different people see in the CPU graph and she sluggish GUI are absolutely normal while older firmware versions do not show that behavior.

I based what I wrote on what people provide me. The top output posted showed nothing using a high CPU usage, and this is what I referred to as being normal.. If top regularly shows otherwise, then people need to post that information, and not the top output when everything is quiet.

On my own RT-AC68U, the graphic never showed any spike above 10-12% for me.

 

[VANT]

I disagree, the shown numbers are taken with certain intervals, also "top" showed now and than much higher numbers than I could capture, the real peak values are not shown.
I cannot believe you are trying to say that the spikes different people see in the CPU graph and she sluggish GUI are absolutely normal while older firmware versions do not show that behavior.

86u have same problem
https://www.snbforums.com/threads/a...on-3-0-0-4-384-81039.58178/page-3#post-511802
this fw is horrible slow

 

[wouterv]

I based what I wrote on what people provide me. The top output posted showed nothing using a high CPU usage, and this is what I referred to as being normal.. If top regularly shows otherwise, then people need to post that information, and not the top output when everything is quiet.

On my own RT-AC68U, the graphic never showed any spike above 10-12% for me.

As I wrote: the top output does show "flashes" of high CPU load, I have seen numbers 20 to 30 but too short visable to capture in the screen dump.
You clearly see that top takes snapshots with certain intervals, with some luck you may see a high value.
I do not believe it is capable to really capture the Max value of CPU load over a certain time period.
At least you see that httpd takes much more CPU time than it did in previous firmware releases.

 

[heysoundude]

As I wrote: the top output does show "flashes" of high CPU load, I have seen numbers 20 to 30 but too short visable to capture in the screen dump.
You clearly see that top takes snapshots with certain intervals, with some luck you may see a high value.
I do not believe it is capable to really capture the Max value of CPU load over a certain time period.
At least you see that httpd takes much more CPU time than it did in previous firmware releases.

Surely there is a way to enable a log in top. anyone?

 

[Jonathan Dunlap]

Looks like 86U is having the same issue with the same version firmware: https://www.snbforums.com/threads/asus-rt-ac86u-firmware-version-3-0-0-4-384-81039.58178/

 

[wshuman3]

Surely there is a way to enable a log in top. anyone?

top -b > output

 

[MicroNate]

This issue is with the web gui. If you close out the web browser while looking at TOP data, you'll notice the CPU usage goes to nearly zero. I wouldn't be worried about it.

 

[wouterv]

This issue is with the web gui. If you close out the web browser while looking at TOP data, you'll notice the CPU usage goes to nearly zero. I wouldn't be worried about it.

Duh, httpd is the Apache web server, that was already mentioned.
I am not worried, just indicating it is an anomaly in this firmware version.

 

[Starbuck99]

Security fix

- Fixed a DDoS vulnerability. Thanks for Altin Thartori's contribution.

Bug fix

- Fixed Samba server compatibility issue.
- Fixed OpenVPN related bugs.
- Fixed schedule reboot bugs.
- Improved AiMesh compatibility.
- Improved system stability.

Link:

https://dlcdnets.asus.com/pub/ASUS/wireless/RT-AC68U/FW_RT_AC68U_300438481039.zip

Update broke Samba Share on Windows 10.
I have been using the samba server with guest login enabled.
Windows 10 cannot connect to the samba server after this update. Even though Windows 10 has been set to Samba V1 - or am I the only one experiencing this?

 

[AllanO]

Completely lost my 2.4ghz band after installing on my aimesh cluster of two ac-rt86u’s and one ac-rt68u. Need to fall back to previous version or stop using aimesh and go to extender configuration. Is the previous version still available to download.

 

[Andy1932]

Completely lost my 2.4ghz band after installing on my aimesh cluster of two ac-rt86u’s and one ac-rt68u. Need to fall back to previous version or stop using aimesh and go to extender configuration. Is the previous version still available to download.

https://www.asus.com/us/Networking/RTAC68U/HelpDesk_BIOS/

 

[LimJK]

Hmm,

I tried doing update from GUI today for RT-AC68U at my daughter's place, looks like it has been withdrawn (?) it recognise 45717 as the latest firmware.

 

[Stephen Harrington]

I tried doing update from GUI today for RT-AC68U at my daughter's place, looks like it has been withdrawn (?) it recognise 45717 as the latest firmware.

Yep, looks like they might be in the middle of pulling it?
I checked on my RT-AC68U running 45717 a couple of days ago and 81039 was showing as an available update, but I didn't do it after reading the complaints here. Just checked again and got:-

"The router cannot connect to ASUS server to check for the firmware update".

81039 is still available on the Australian server as a manual download though ...

https://dlcdnets.asus.com/pub/ASUS/wireless/RT-AC68U/FW_RT_AC68U_300438481039.zip

 

[LimJK]

81039 is still available on the Australian server as a manual download though ...

Stephen,
Thanks ... I actually have updated FW 81039 for AiMesh "nodes" last week both RT-AC68U and RT-AC86U and it appear to be running without issue so far. However, like you, after reading the complaints for new Firmware for RT-AC68U as AiMesh Router, I decided not to update the firmware for now.

 

[bicubic]

After installing 81039 I lost 5Ghz for first 15 min. Go back to 45717 and 5Ghz show immediately