Joined: 09 Nov 2014 Posts: 312 Location: Bakersfield, CA
Posted: Sat Oct 29, 2016 18:03 Post subject:
My two R7000s are still up. No issues so far, gateway has QoS setup too.
Gateway:
up 21:56, load average: 0.00, 0.00, 0.00
Server AP:
up 21:50, load average: 0.00, 0.01, 0.00
I've not noticed any odd things, and it seems pages load faster with this build (that just might be me in my head. Lol) _________________ [size=9]
Deployed Routers:
Netgear R7800 - 1x build 45493
- Gateway (USB /w Entware, CAKE QoS)
Netgear R7000 - 5x build 45493
Router: R8000
Firmware: DD-WRT v3.0-r30800M kongac (10/27/16)
Kernel: Linux 4.4.25 #134 SMP Thu Oct 27 16:01:52 CEST 2016 armv7l
Previous: R30700 kong
Status: ok
Reset: No
Errors: No
Couple of R7000 here, updated with 30800 (ddup --flash-latest), I did erase the nvram via command line afterwards.
Both units unstable before the nvram erase, both units completely stable afterwards (8h).
Only problem encountered, VAPs don't work - at all - when set via GUI (not tried to set them via SSH). I opened a separate thread about this.
--
Update:
Something seems to trigger frequent reboots (the affected unit stops responding correctly afterwards and I've got to reboot it again manually with the power button). I did not isolate a particular device that's causing it, because there were ~20 different devices connected between 2.4 and 5ghz.
Settings other than default at reboot time:
reg. domain Europe, same SSID for 2.4/5, multicast, turboquam, i&e beamforming, a.t. fairness, all those settings enabled on both radios, WPA2PSKY AES, 2.4 beacon interval 200 dtim 4, 5 beacon 100 dtim 5, auto preamble for both, bt preemption on 2.4 - ht40 ch1 for 2.4, vht80 ch116 lower lower for 5 - 2 bridged AP isolated 2.4/5 VAPs with the same SSID - htb+fq_codel active - LLTD enabled - PPPoE on WAN.
CPU load was unusually very high with just a few devices connected (~18% vs 2-3% with 30700).
Flashing 30700m on both units (GUI set to default, no command line nvram erase this time), apparently solved the problem.
I loaded 30800 (then erase nvram) on one R7000 and during the manual configuration process the R7000 would reboot when accessing the GUI and I was unable to complete the configuration and gave up.
I'll stick with 30700 for now... _________________ Home Network on Telus PureFibre - 10GbE Copper Backbone
2x R7800 - Gateway & WiFi & 3xWireGuard - DDWRT r45928 Std
R7000 - Wired ISP 4K IPTV Gateway - DDWRT r44483 Std
Off Site 1
R7000 - Gateway & WiFi & WireGuard - DDWRT r44483 Std
WRT610Nv1 - Client Bridge - DDWRT r33679 Mega K2.4
Off Site 2
R7000 - Gateway & WiFi - DDWRT r44483 Std
E2000 - Wired ISP IPTV PVR Blocker - DDWRT r35531
Joined: 21 Jul 2006 Posts: 1867 Location: Fortaleza Ce Brazil
Posted: Sun Oct 30, 2016 10:26 Post subject:
After 2 and 1/2 days working ... r8500 crashes wifi 2.4
Problems still there
Router still working but some wifi 2.4 connections drops and cant stablish connect again ... This happens with just wifi 2.4 .. others clients ( 5 gz ) still connected and working . _________________ DDwrt ...it rocks ....
1 R7800 45229 AP Wireguard webserver JFFS SAMBA FTP usb HD Mesh
1 R7800 45229 Cli Mesh
1 WZR1750 45229 AP Webserver OpenVPN Wireguard
1 DDx86 Vmware 40260
1 E3000 41027 K3 @500 WAP wifi client
Joined: 08 Oct 2007 Posts: 47 Location: Winnipeg, MB
Posted: Sun Oct 30, 2016 18:19 Post subject: Re: Not Good
<Kong> wrote:
cc999 wrote:
R8500
Updated thru telnet
Lasted 2 hours, streaming to my laptop (5 ghz) caused a complete router crash.This is after about 30 minutes into the stream.
ver 30700 ran flawlessly. I went back to 30700.
I did NOT clear nvram just updated from the telnet prompt as stated before. Whether or not that is a problem when moving up to this version I guess only KONG can answer.
CC
I do not recall a lot of changes that have any potential to cause this. Mostly little fixes that cannot cause crashes.
Ndpi was the biggest change which is a kernel module and therefore would be my first candidate for inspection. Are you doing any protocol prioritization/blocking?
Similar issue here. Went from 30700 to 30800 using ddup. Immediately noticed router was crashing/restarting every few minutes. I had already read this thread, so I went and disabled the QoS service.
Router seemed stable when QoS disabled. Tried re-enabling QoS and rebooting. Router didn't come back online. Power cycled, router came back up. Router started crashing/restarting every 2-3 minutes again.
Again, disabled QoS... Stable for 15 mins now.
I'm using TCP packet, services, netmask and mac priorities on the QoS page.
Posted: Sun Oct 30, 2016 18:24 Post subject: Re: Not Good
Tarv wrote:
<Kong> wrote:
cc999 wrote:
R8500
Updated thru telnet
Lasted 2 hours, streaming to my laptop (5 ghz) caused a complete router crash.This is after about 30 minutes into the stream.
ver 30700 ran flawlessly. I went back to 30700.
I did NOT clear nvram just updated from the telnet prompt as stated before. Whether or not that is a problem when moving up to this version I guess only KONG can answer.
CC
I do not recall a lot of changes that have any potential to cause this. Mostly little fixes that cannot cause crashes.
Ndpi was the biggest change which is a kernel module and therefore would be my first candidate for inspection. Are you doing any protocol prioritization/blocking?
Similar issue here. Went from 30700 to 30800 using ddup. Immediately noticed router was crashing/restarting every few minutes. I had already read this thread, so I went and disabled the QoS service.
Router seemed stable when QoS disabled. Tried re-enabling QoS and rebooting. Router didn't come back online. Power cycled, router came back up. Router started crashing/restarting every 2-3 minutes again.
Again, disabled QoS... Stable for 15 mins now.
I'm using TCP packet, services, netmask and mac priorities on the QoS page.
After 2 and 1/2 days working ... r8500 crashes wifi 2.4
Problems still there
Router still working but some wifi 2.4 connections drops and cant stablish connect again ... This happens with just wifi 2.4 .. others clients ( 5 gz ) still connected and working .
What type of client device has problems? I noticed one connection problem in the last 30 days with a ralink device, had some trouble with packet aggregation. _________________ KONG PB's: http://www.desipro.de/ddwrt/
KONG Info: http://tips.desipro.de/
Joined: 21 Jul 2006 Posts: 1867 Location: Fortaleza Ce Brazil
Posted: Mon Oct 31, 2016 8:51 Post subject:
<Kong> wrote:
stalonge wrote:
After 2 and 1/2 days working ... r8500 crashes wifi 2.4
Problems still there
Router still working but some wifi 2.4 connections drops and cant stablish connect again ... This happens with just wifi 2.4 .. others clients ( 5 gz ) still connected and working .
What type of client device has problems? I noticed one connection problem in the last 30 days with a ralink device, had some trouble with packet aggregation.
1 noticed problems with at least 3 devices , 1 samsung TV .. ,1 playstaion 4 , 1 notebook ( Atheros Communications Inc. AR9565 Wireless Network Adapter ) and 1 Samsung cell phone ...
What is " trouble with packet aggregation " . How can i notice this ?
_________________ DDwrt ...it rocks ....
1 R7800 45229 AP Wireguard webserver JFFS SAMBA FTP usb HD Mesh
1 R7800 45229 Cli Mesh
1 WZR1750 45229 AP Webserver OpenVPN Wireguard
1 DDx86 Vmware 40260
1 E3000 41027 K3 @500 WAP wifi client
repeatedly unstable on two EA6400 (with modded CFE), even with reset.
i use these devices as AP's with no special services, but with 2 virtual SSID's on each radio.
the devices would work for a few mintues, but become progressively bad: unable to access gui, and unable to accept new radio clients (wpa2 personal encryption). finally existing clients would lock up.
the predictably-unpredictable time-dependent failure suggests a memory leak to me.
i see Kong has a newer build - so he's probably on top of this.
i suggest staying away from 30800-----
thanks!
