I've been an avid lurker for quite a while here, and I need the advice of the learned and august guild of Asus owners here. I recently purchased a used (but working) RT-AC86U, and have some questions about the differences I see between this router and the reliable RT-AC1900P I have been using for the past 4 years.
My set-up: RT-AC86U running 3.0.0.4.486.41634 FW. WAN connection is symmetric 940MB/940MB via fiber to the house. Network is largely connected via CAT5e with only two LAN ports (1&2) used on the Asus, one for a VOIP adapter and the other to an 8 port gigabit switch to all the switches/computers in the house. The only WiFi connections are to the usual assortment of iPhones, iPads, and Airport Expresses for browsing and music distribution, about 12 or so total. Everything works fine, and my old RT-AC1900P was stable and worked for months between resets, which were done on principle, not due to need. I don't use any of the features, just the basics: routing and WiFi. No AiMesh, as we live in a 1000 sq. foot wooden house, so coverage (and CAT5e cable length) aren't issues.
I've only had the RT-AC86 up for about a day, but I was surprised at the system memory footprint compared to the 1900P (71-78MB used out of 256MB, ~27-30% after 3 months). The 86U starts out after reset with 284MB used (56%) out of 512MB and after I poked around looking at settings, changing icons and names of client devices 24 hours later I have 311MB used (61%). One thing that may contribute to this is a bunch of log entries related to normal WiFi activity. I regard that as a sort of memory leak, but megabytes? This seems like a lot of code/data compared to my 1900P; code bloat?
What on earth is going on, and why? If I was turning on the extra services I could understand this behavior, but all I do is log in and look around to verify that all is well, check the device lists to to look for unknown devices, look at the real-time traffic monitor, and log out. Nothing special, and and I'm not switching services on and off (loading and unloading code blocks and storage). Is there any way I can unload some of the junk in memory? Or is this just the "new normal" and I have to live with it?
Related to this, here is an example of the entries flooding the machine log; I didn't see this in my 1900P log until after a recent FW update, and fixed it as outlined below:
Feb 2 10:05:23 kernel: SHN Release Version: 2.0.1 851496c
Feb 2 10:05:23 kernel: UDB Core Version: 0.2.18
Feb 2 10:05:23 kernel: sizeof forward pkt param = 280
Feb 2 10:05:23 BWDPI: fun bitmap = 3
Feb 2 10:05:54 kernel: SHN Release Version: 2.0.1 851496c
Feb 2 10:05:54 kernel: UDB Core Version: 0.2.18
Feb 2 10:05:54 kernel: sizeof forward pkt param = 280
Feb 2 10:05:54 BWDPI: fun bitmap = 3
Feb 2 10:40:46 wlceventd: wlceventd_proc_event(507): eth6: Disassoc A8:86
D:AD:94:01, status: 0, reason: Disassociated because sending station is leaving (or has left) BSS (8), rssi:0
Feb 2 10:44:54 wlceventd: wlceventd_proc_event(507): eth6: Disassoc 20:C90:97:50:B9, status: 0, reason: Disassociated because sending station is leaving (or has left) BSS (8), rssi:0
Feb 2 10:44:54 wlceventd: wlceventd_proc_event(526): eth6: Auth 20:C90:97:50:B9, status: Successful (0), rssi:0
Feb 2 10:44:54 wlceventd: wlceventd_proc_event(555): eth6: Assoc 20:C90:97:50:B9, status: Successful (0), rssi:0
Feb 2 10:56:41 wlceventd: wlceventd_proc_event(490): eth6: Deauth_ind 7C1:C3:8A:45:32, status: 0, reason: Class 3 frame received from nonassociated station (7), rssi:0
Feb 2 10:56:41 wlceventd: wlceventd_proc_event(490): eth6: Deauth_ind 7C1:C3:8A:45:32, status: 0, reason: Class 3 frame received from nonassociated station (7), rssi:0
Feb 2 10:56:45 wlceventd: wlceventd_proc_event(526): eth6: Auth 7C1:C3:8A:45:32, status: Successful (0), rssi:0
Feb 2 10:56:45 wlceventd: wlceventd_proc_event(555): eth6: Assoc 7C1:C3:8A:45:32, status: Successful (0), rssi:0
Feb 2 11:13:24 wlceventd: wlceventd_proc_event(526): eth6: Auth A8:86D:AD:94:01, status: Successful (0), rssi:0
Feb 2 11:13:24 wlceventd: wlceventd_proc_event(555): eth6: Assoc A8:86D:AD:94:01, status: Successful (0), rssi:0
Feb 2 11:19:40 wlceventd: wlceventd_proc_event(507): eth6: Disassoc 20:C90:97:50:B9, status: 0, reason: Disassociated because sending station is leaving (or has left) BSS (8), rssi:0
Feb 2 11:19:40 wlceventd: wlceventd_proc_event(526): eth6: Auth 20:C90:97:50:B9, status: Successful (0), rssi
--------------
I included the part with "BWDPI: fun bitmap = 3" because I like the idea of a "fun bitmap", it keeps the whole thing from being didactic and boring. Anyway:
Is there a command-line entry I can make on the 86U via SSH to reduce the amount of visible (and stored) junk in the log related to normal operation? On my 1900P I set the log level to 5 to suppress the same wlceventd log entries; is the same command effective on the 86U? I'm asking as I'd rather not make a horrible mistake, and I believe the log level works inversely with level 6 including more classes of log data (current default?), but I may have this backward.
Thanks for your collective wisdom and help. The 86U is snappier, and it's fun to have a new toy to play with.
Cheers,
Peter
My set-up: RT-AC86U running 3.0.0.4.486.41634 FW. WAN connection is symmetric 940MB/940MB via fiber to the house. Network is largely connected via CAT5e with only two LAN ports (1&2) used on the Asus, one for a VOIP adapter and the other to an 8 port gigabit switch to all the switches/computers in the house. The only WiFi connections are to the usual assortment of iPhones, iPads, and Airport Expresses for browsing and music distribution, about 12 or so total. Everything works fine, and my old RT-AC1900P was stable and worked for months between resets, which were done on principle, not due to need. I don't use any of the features, just the basics: routing and WiFi. No AiMesh, as we live in a 1000 sq. foot wooden house, so coverage (and CAT5e cable length) aren't issues.
I've only had the RT-AC86 up for about a day, but I was surprised at the system memory footprint compared to the 1900P (71-78MB used out of 256MB, ~27-30% after 3 months). The 86U starts out after reset with 284MB used (56%) out of 512MB and after I poked around looking at settings, changing icons and names of client devices 24 hours later I have 311MB used (61%). One thing that may contribute to this is a bunch of log entries related to normal WiFi activity. I regard that as a sort of memory leak, but megabytes? This seems like a lot of code/data compared to my 1900P; code bloat?
What on earth is going on, and why? If I was turning on the extra services I could understand this behavior, but all I do is log in and look around to verify that all is well, check the device lists to to look for unknown devices, look at the real-time traffic monitor, and log out. Nothing special, and and I'm not switching services on and off (loading and unloading code blocks and storage). Is there any way I can unload some of the junk in memory? Or is this just the "new normal" and I have to live with it?
Related to this, here is an example of the entries flooding the machine log; I didn't see this in my 1900P log until after a recent FW update, and fixed it as outlined below:
Feb 2 10:05:23 kernel: SHN Release Version: 2.0.1 851496c
Feb 2 10:05:23 kernel: UDB Core Version: 0.2.18
Feb 2 10:05:23 kernel: sizeof forward pkt param = 280
Feb 2 10:05:23 BWDPI: fun bitmap = 3
Feb 2 10:05:54 kernel: SHN Release Version: 2.0.1 851496c
Feb 2 10:05:54 kernel: UDB Core Version: 0.2.18
Feb 2 10:05:54 kernel: sizeof forward pkt param = 280
Feb 2 10:05:54 BWDPI: fun bitmap = 3
Feb 2 10:40:46 wlceventd: wlceventd_proc_event(507): eth6: Disassoc A8:86
D:AD:94:01, status: 0, reason: Disassociated because sending station is leaving (or has left) BSS (8), rssi:0Feb 2 10:44:54 wlceventd: wlceventd_proc_event(507): eth6: Disassoc 20:C9
0:97:50:B9, status: 0, reason: Disassociated because sending station is leaving (or has left) BSS (8), rssi:0Feb 2 10:44:54 wlceventd: wlceventd_proc_event(526): eth6: Auth 20:C9
0:97:50:B9, status: Successful (0), rssi:0Feb 2 10:44:54 wlceventd: wlceventd_proc_event(555): eth6: Assoc 20:C9
0:97:50:B9, status: Successful (0), rssi:0Feb 2 10:56:41 wlceventd: wlceventd_proc_event(490): eth6: Deauth_ind 7C
1:C3:8A:45:32, status: 0, reason: Class 3 frame received from nonassociated station (7), rssi:0Feb 2 10:56:41 wlceventd: wlceventd_proc_event(490): eth6: Deauth_ind 7C
1:C3:8A:45:32, status: 0, reason: Class 3 frame received from nonassociated station (7), rssi:0Feb 2 10:56:45 wlceventd: wlceventd_proc_event(526): eth6: Auth 7C
1:C3:8A:45:32, status: Successful (0), rssi:0Feb 2 10:56:45 wlceventd: wlceventd_proc_event(555): eth6: Assoc 7C
1:C3:8A:45:32, status: Successful (0), rssi:0Feb 2 11:13:24 wlceventd: wlceventd_proc_event(526): eth6: Auth A8:86
D:AD:94:01, status: Successful (0), rssi:0Feb 2 11:13:24 wlceventd: wlceventd_proc_event(555): eth6: Assoc A8:86
D:AD:94:01, status: Successful (0), rssi:0Feb 2 11:19:40 wlceventd: wlceventd_proc_event(507): eth6: Disassoc 20:C9
0:97:50:B9, status: 0, reason: Disassociated because sending station is leaving (or has left) BSS (8), rssi:0Feb 2 11:19:40 wlceventd: wlceventd_proc_event(526): eth6: Auth 20:C9
0:97:50:B9, status: Successful (0), rssi--------------
I included the part with "BWDPI: fun bitmap = 3" because I like the idea of a "fun bitmap", it keeps the whole thing from being didactic and boring. Anyway:
Is there a command-line entry I can make on the 86U via SSH to reduce the amount of visible (and stored) junk in the log related to normal operation? On my 1900P I set the log level to 5 to suppress the same wlceventd log entries; is the same command effective on the 86U? I'm asking as I'd rather not make a horrible mistake, and I believe the log level works inversely with level 6 including more classes of log data (current default?), but I may have this backward.
Thanks for your collective wisdom and help. The 86U is snappier, and it's fun to have a new toy to play with.
Cheers,
Peter
Last edited:
