FlexQoS FlexQoS 1.2.4 - Flexible QoS Enhancement Script for Adaptive QoS

[dave14305]

I also noticed that the Bandwidth values changed as well. I have a screenshot of before applying the Development branch + Hotfix, but I unfortunately changed the values back to what my screenshot was, in order for the values to be the same when they were changed, so I can't remember what they were changed to. As this is the development branch, I should have taken a screenshot to help troubleshoot if required.

I can’t remember exactly what they had all changed to, but I always have gaming as my highest priority with 35% of minimum bandwidth. Gaming had changed to 5% and file downloads had changed to 30 or 35% (previously 5%) on the download values. Pretty sure it was both download and upload values. I only checked because I had latency spikes while gaming and double checked everything was still working as it should. All ok after I edited again!

While netcontrol and web surfing were as before, min-values were mixed (max is 100% in my setup but I'd assume that these would also be involved) between the classes. As if you take the whole line (up and down limits together) and shift it to another class and so on. Sorry to say but I didn't take a screenshot so I can't tell you which line moved from class a to class b

OK, I went back to stable, and set my bandwidths in a numerically ascending order, then switched to develop. So I see there is a problem, based on the mess below.

EDIT: I found my flawed logic. I have fixed it locally, but it's partially intertwined with another change that I'm not ready to push out yet. So I need to brush up on my git stashing first.

 

[dave14305]

I pushed another fix to the develop branch. It should prevent the issue from occurring again during conversion from stable to develop. If you're already messed up on develop, you will need to manually reconfigure your rates. Sorry about that. My logic for understanding how to map the old order to new order was flawed and it made for a headache.

Lesson learned: don't try to convert settings once you've already decided on one format and sequence.

 

[raion969]

What settings would you recommend in the bandwith setting for primary gaming to dont interferee?

 

[sentinelvdx]

OK, I went back to stable, and set my bandwidths in a numerically ascending order, then switched to develop. So I see there is a problem, based on the mess below.

View attachment 30292

EDIT: I found my flawed logic. I have fixed it locally, but it's partially intertwined with another change that I'm not ready to push out yet. So I need to brush up on my git stashing first.

This screenshot comes from default/stable branch? because mine looks different on Dev Branch

Mine looks like this (Attached)

PS: If I switch back to stable rates will switch automatically to default? or better fully uninstall it and reinstall again?

 

[dave14305]

This screenshot comes from default/stable branch? because mine looks different on Dev Branch

Mine looks like this (Attached)

PS: If I switch back to stable rates will switch automatically to default? or better fully uninstall it and reinstall again?

My screenshot was from develop branch after setting the bandwidth on stable. If you switch back to stable, the old bandwidth custom setting variable doesn't exist, so it will reset to default values. Better to just fix it on develop and be done with it.

Your bandwidth settings will look different if your QoS priorities are different than mine (which they are). Is that what you're noticing about "looks different"?

 

[sentinelvdx]

My screenshot was from develop branch after setting the bandwidth on stable. If you switch back to stable, the old bandwidth custom setting variable doesn't exist, so it will reset to default values. Better to just fix it on develop and be done with it.

Your bandwidth settings will look different if your QoS priorities are different than mine (which they are). Is that what you're noticing about "looks different"?

I was referring to the min and max, in my case all looks using 100% as max rate. But was not sure how it looked like on stable.
To be honest because it was working so good before on 1.1.0 before moving to 386.1, I've never touched rates.

 

[dave14305]

I was referring to the min and max, in my case all looks using 100% as max rate. But was not sure how it looked like on stable.
To be honest because it was working so good before on 1.1.0 before moving to 386.1, I've never touched rates.

100% is the default max always. I set mine between 10-80 to be able to detect if the maximum percentages were getting mis-sorted, which they were. Now I'm back to 100% on maximums.

 

[BigTim]

Kind of on/off topic. Is there any way to change the modify the number of tracked connections to get rid of the "Disabled - too many tracked connections." on the Classification page? I'm always running well over 1000 tracked connections and like looking at both pages of graphs on the Classification page and the QoS page.

 

[dave14305]

Kind of on/off topic. Is there any way to change the modify the number of tracked connections to get rid of the "Disabled - too many tracked connections." on the Classification page? I'm always running well over 1000 tracked connections and like looking at both pages of graphs on the Classification page and the QoS page.

You can re-enable the refresh rate on the FlexQoS page and it should remember it for the browser session. I don’t easily reach 750 connections here, so I haven’t tested it very often. Can’t do anything about the Classification page.

 

[drunki]

Just wanted to share my difficulties I have with the script recently. I try to keep a close eye on this threat and the one for the latest RMerlin version but I might have missed something. AC68U user, running latest 381.1 (dirty flash over the last stable version). I had been running the script without a problem for a quite some time. Recently I discovered each restart of the router wont't make possible the script to run properly and I am getting the following messages:

Feb 5 12:52:00 FlexQoS: /jffs/addons/flexqos/flexqos.sh (pid=28690) called in unattended mode with 1 args: -check
Feb 5 12:52:00 FlexQoS: iptables rules already present
Feb 5 12:52:01 FlexQoS: TC Modification Delayed Start
Feb 5 12:57:12 FlexQoS: TC Modification Delay reached maximum 300 seconds. Aborting startup!
Feb 5 12:57:12 FlexQoS: TC Modification delayed for 300 seconds
Feb 5 12:57:15 FlexQoS: Applying AppDB rules and TC rates
Feb 5 12:57:15 FlexQoS: ERROR! Check /tmp/flexqos_tcrules.log

Have I missed something or should I proceed with a factory reset and set everything from scratch ?

Thank you Dave for the great work !

 

[dave14305]

Have I missed something or should I proceed with a factory reset and set everything from scratch ?

There is a known issue in the 386.1 thread that QoS won’t initialize properly on some models during a reboot. When you see this error, run these commands to see if Adaptive QoS is setup or not:

tc qdisc ls
tc class show dev br0
nvram get bwdpi_app_rulelist

 

[BoostOver]

When link ppp go down FlexQos (Dev) won't start:
My syslog of 386.01gnuton0beta1 on DSL AC68U:

Feb 5 12:56:57 pppd[4063]: Connection terminated.
Feb 5 12:56:58 kernel: Interface ppp0 doesn't exist
Feb 5 12:57:00 pppd[4208]: pppd 2.4.7 started by Nutella05, uid 0
Feb 5 12:57:00 pppd[4208]: Connected to f0:4b:3a:43:6c:8b via interface vlan3880
Feb 5 12:57:00 pppd[4208]: Connect: ppp0 <--> vlan3880
Feb 5 12:57:03 pppd[4208]: PAP authentication succeeded
Feb 5 12:57:03 pppd[4208]: peer from calling number F0:4B:3A:43:6C:8B authorized
Feb 5 12:57:04 pppd[4208]: local IP address 79.31.209.222
Feb 5 12:57:04 pppd[4208]: remote IP address 192.168.100.1
Feb 5 12:57:05 miniupnpd[4210]: shutting down MiniUPnPd
Feb 5 12:57:05 FlexQoS: /jffs/addons/flexqos/flexqos.sh (pid=4253) called in unattended mode with 1 args: -start
Feb 5 12:57:06 miniupnpd[4330]: HTTP listening on port 42552
Feb 5 12:57:06 miniupnpd[4330]: Listening for NAT-PMP/PCP traffic on port 5351
Feb 5 12:57:06 inadyn[4333]: In-a-dyn version 2.7 -- Dynamic DNS update client.
Feb 5 12:57:06 inadyn[4333]: Update forced for alias bremebordiga73.asuscomm.com, new IP# 79.31.209.222
Feb 5 12:57:08 FlexQoS: Applying iptables static rules
Feb 5 12:57:08 inadyn[4333]: alias address=<79.31.209.222>
Feb 5 12:57:08 inadyn[4333]: Remove old cache file /var/cache/inadyn for bremebordiga73.asuscomm.com
Feb 5 12:57:08 inadyn[4333]: Updating cache for bremebordiga73.asuscomm.com
Feb 5 12:57:08 FlexQoS: Applying iptables custom rules
Feb 5 12:57:13 FlexQoS: Flushing conntrack table
Feb 5 12:57:15 FlexQoS: TC Modification Delayed Start
Feb 5 12:57:15 FlexQoS: /jffs/addons/flexqos/flexqos.sh (pid=4818) called in unattended mode with 1 args: -start
Feb 5 12:57:16 FlexQoS: [*] Killing Delayed Process (pid=4253)
Feb 5 12:57:16 FlexQoS: [*] 4253 Nutella0 1592 S sh /jffs/addons/flexqos/flexqos.sh -start
Feb 5 12:57:18 FlexQoS: Applying iptables static rules
Feb 5 12:57:19 FlexQoS: Applying iptables custom rules
Feb 5 12:57:19 FlexQoS: Flushing conntrack table
Feb 5 12:57:20 FlexQoS: TC Modification Delayed Start
Feb 5 12:57:30 kernel: SHN Release Version: 2.0.1 7812743e
Feb 5 12:57:30 kernel: UDB Core Version: 0.2.18
Feb 5 12:57:30 kernel: sizeof forward pkt param = 192
Feb 5 12:57:31 kernel: ERR[qos_start:3363] qos_ops is not registered!
Feb 5 12:57:31 kernel: ioctl_iqos_op_switch(1) fail!
Feb 5 12:57:36 FlexQoS: /jffs/addons/flexqos/flexqos.sh (pid=5434) called in unattended mode with 1 args: -start
Feb 5 12:57:36 FlexQoS: [*] Killing Delayed Process (pid=4818)
Feb 5 12:57:37 FlexQoS: [*] 4818 Nutella0 1592 S sh /jffs/addons/flexqos/flexqos.sh -start
Feb 5 12:57:40 FlexQoS: Applying iptables static rules
Feb 5 12:57:40 FlexQoS: Applying iptables custom rules
Feb 5 12:57:41 FlexQoS: Flushing conntrack table
Feb 5 12:57:42 FlexQoS: TC Modification Delayed Start
Feb 5 13:02:14 syslog: wlceventd_proc_event(526): eth1: Auth C8:28:32:2B:04:1E, status: Successful (0), rssi:0
Feb 5 13:02:14 syslog: wlceventd_proc_event(555): eth1: Assoc C8:28:32:2B:04:1E, status: Successful (0), rssi:0
Feb 5 13:02:53 FlexQoS: TC Modification Delay reached maximum 300 seconds. Aborting startup!
Feb 5 13:03:12 syslog: wlceventd_proc_event(490): eth1: Deauth_ind C8:28:32:2B:04:1E, status: 0, reason: Disassociated due to inactivity (4), rssi:0
Feb 5 13:03:13 syslog: wlceventd_proc_event(490): eth1: Deauth_ind C8:28:32:2B:04:1E, status: 0, reason: Class 3 frame received from nonassociated station (7), rssi:0
Feb 5 13:03:38 syslog: wlceventd_proc_event(490): eth1: Deauth_ind C8:28:32:2B:04:1E, status: 0, reason: Class 3 frame received from nonassociated station (7), rssi:0
Feb 5 13:07:01 FlexQoS: /jffs/addons/flexqos/flexqos.sh (pid=7352) called in unattended mode with 1 args: -check
Feb 5 13:07:01 FlexQoS: iptables rules already present
Feb 5 13:07:01 FlexQoS: TC Modification Delayed Start
Feb 5 13:12:12 FlexQoS: TC Modification Delay reached maximum 300 seconds. Aborting startup!
Feb 5 13:17:01 FlexQoS: /jffs/addons/flexqos/flexqos.sh (pid=8982) called in unattended mode with 1 args: -check
Feb 5 13:17:01 FlexQoS: iptables rules already present
Feb 5 13:17:01 FlexQoS: TC Modification Delayed Start

 

[dave14305]

When link ppp go down FlexQos (Dev) won't start:

Run the same commands in the post above yours (#352).

 

[raion969]

in FlexQoS - v1.1.1 Dev what does the Router Outbound Traffic Class mean ?
Right now it is set as file downloads.

 

[dave14305]

in FlexQoS - v1.1.1 Dev what does the Router Outbound Traffic Class mean ?
Right now it is set as file downloads.

FlexQoS - FlexQoS 1.2.4 - Flexible QoS Enhancement Script for Adaptive QoS

Also upgraded to the new 386.1 and got the following entry after upgrade and again after a reboot: Jan 31 15:42:29 FlexQoS: TC Modification Delay reached maximum 300 seconds. Aborting startup! FlexQoS isn't running afterwards, a flexqos restart command works. Tried to fix with a reinstall now...

www.snbforums.com

Before you ask, it doesn't affect gaming at all.

 

[bluzfanmr1]

The last 2 days, since switching to the develop branch, it looks like FlexQoS has stopped working at a similar time each day and I have the following messages that repeat endlessly unti I restart the script:

Feb  4 16:57:59 RT-AC66U_B1 FlexQoS: TC Modification Delay reached maximum 300 seconds. Aborting startup!
Feb  4 17:03:00 RT-AC66U_B1 FlexQoS: /jffs/addons/flexqos/flexqos.sh (pid=8326) called in unattended mode with 1 args: -check
Feb  4 17:03:01 RT-AC66U_B1 FlexQoS: iptables rules already present
Feb  4 17:03:01 RT-AC66U_B1 FlexQoS: TC Modification Delayed Start

Both times the previous log messages are identical and seem to be triggering the FlexQoS messages.

custom_script: Running /jffs/scripts/service-event (args: start samba)
Feb  4 16:52:45 RT-AC66U_B1 FlexQoS: [*] Killing Delayed Process (pid=1244)
Feb  4 16:52:45 RT-AC66U_B1 FlexQoS: [*]  1244 bluzfanm  1588 S    sh /jffs/addons/flexqos/flexqos.sh -start
Feb  4 16:52:45 RT-AC66U_B1 custom_script: Running /jffs/scripts/dnsmasq.postconf (args: /etc/dnsmasq.conf)
Feb  4 16:52:47 RT-AC66U_B1 rc_service: watchdog 621:notify_rc start_dnsmasq
Feb  4 16:52:47 RT-AC66U_B1 rc_service: waitting "start_samba" via  ...
Feb  4 16:52:47 RT-AC66U_B1 FlexQoS: Applying iptables static rules
Feb  4 16:52:47 RT-AC66U_B1 FlexQoS: Applying iptables custom rules
Feb  4 16:52:48 RT-AC66U_B1 FlexQoS: Flushing conntrack table
Feb  4 16:52:48 RT-AC66U_B1 Diversion: restarted Dnsmasq to apply settings
Feb  4 16:52:48 RT-AC66U_B1 (dnsmasq.postconf): Updating /etc/dnsmasq.conf for unbound.....
Feb  4 16:52:48 RT-AC66U_B1 FlexQoS: TC Modification Delayed Start
Feb  4 16:52:48 RT-AC66U_B1 kernel: gro enabled with interval 2
Feb  4 16:52:48 RT-AC66U_B1 Samba_Server: daemon is started
Feb  4 16:52:48 RT-AC66U_B1 custom_script: Running /jffs/scripts/service-event-end (args: start samba)
Feb  4 16:52:48 RT-AC66U_B1 wsdd2[4459]: starting.
Feb  4 16:52:49 RT-AC66U_B1 custom_script: Running /jffs/scripts/service-event (args: start dnsmasq)
Feb  4 16:52:50 RT-AC66U_B1 custom_script: Running /jffs/scripts/dnsmasq.postconf (args: /etc/dnsmasq.conf)
Feb  4 16:52:52 RT-AC66U_B1 Diversion: restarted Dnsmasq to apply settings
Feb  4 16:52:52 RT-AC66U_B1 (dnsmasq.postconf): Updating /etc/dnsmasq.conf for unbound.....
Feb  4 16:52:52 RT-AC66U_B1 custom_script: Running /jffs/scripts/service-event-end (args: start dnsmasq)

Is it something I've done or could it be something with the script?

 

[dave14305]

The last 2 days, since switching to the develop branch, it looks like FlexQoS has stopped working at a similar time each day and I have the following messages that repeat endlessly unti I restart the script:

Feb  4 16:57:59 RT-AC66U_B1 FlexQoS: TC Modification Delay reached maximum 300 seconds. Aborting startup!
Feb  4 17:03:00 RT-AC66U_B1 FlexQoS: /jffs/addons/flexqos/flexqos.sh (pid=8326) called in unattended mode with 1 args: -check
Feb  4 17:03:01 RT-AC66U_B1 FlexQoS: iptables rules already present
Feb  4 17:03:01 RT-AC66U_B1 FlexQoS: TC Modification Delayed Start

Both times the previous log messages are identical and seem to be triggering the FlexQoS messages.

custom_script: Running /jffs/scripts/service-event (args: start samba)
Feb  4 16:52:45 RT-AC66U_B1 FlexQoS: [*] Killing Delayed Process (pid=1244)
Feb  4 16:52:45 RT-AC66U_B1 FlexQoS: [*]  1244 bluzfanm  1588 S    sh /jffs/addons/flexqos/flexqos.sh -start
Feb  4 16:52:45 RT-AC66U_B1 custom_script: Running /jffs/scripts/dnsmasq.postconf (args: /etc/dnsmasq.conf)
Feb  4 16:52:47 RT-AC66U_B1 rc_service: watchdog 621:notify_rc start_dnsmasq
Feb  4 16:52:47 RT-AC66U_B1 rc_service: waitting "start_samba" via  ...
Feb  4 16:52:47 RT-AC66U_B1 FlexQoS: Applying iptables static rules
Feb  4 16:52:47 RT-AC66U_B1 FlexQoS: Applying iptables custom rules
Feb  4 16:52:48 RT-AC66U_B1 FlexQoS: Flushing conntrack table
Feb  4 16:52:48 RT-AC66U_B1 Diversion: restarted Dnsmasq to apply settings
Feb  4 16:52:48 RT-AC66U_B1 (dnsmasq.postconf): Updating /etc/dnsmasq.conf for unbound.....
Feb  4 16:52:48 RT-AC66U_B1 FlexQoS: TC Modification Delayed Start
Feb  4 16:52:48 RT-AC66U_B1 kernel: gro enabled with interval 2
Feb  4 16:52:48 RT-AC66U_B1 Samba_Server: daemon is started
Feb  4 16:52:48 RT-AC66U_B1 custom_script: Running /jffs/scripts/service-event-end (args: start samba)
Feb  4 16:52:48 RT-AC66U_B1 wsdd2[4459]: starting.
Feb  4 16:52:49 RT-AC66U_B1 custom_script: Running /jffs/scripts/service-event (args: start dnsmasq)
Feb  4 16:52:50 RT-AC66U_B1 custom_script: Running /jffs/scripts/dnsmasq.postconf (args: /etc/dnsmasq.conf)
Feb  4 16:52:52 RT-AC66U_B1 Diversion: restarted Dnsmasq to apply settings
Feb  4 16:52:52 RT-AC66U_B1 (dnsmasq.postconf): Updating /etc/dnsmasq.conf for unbound.....
Feb  4 16:52:52 RT-AC66U_B1 custom_script: Running /jffs/scripts/service-event-end (args: start dnsmasq)

Is it something I've done or could it be something with the script?

Adaptive QoS isn't initializing properly it seems. There's a known issue posted in the 386.1 thread post #2, but it isn't clear how widespread the problem is. I could add a qos restart after the 5 minute delay if it would help. Next time it gets into one of these waiting states, login via ssh and run tc qdisc ls and then service restart_qos. If FlexQoS finishes normally within a minute, that will be a positive result. I haven't seen this behavior yet myself, so I'm not sure what it looks like, so post the output of the tc command to confirm my suspicions.

 

[bluzfanmr1]

Adaptive QoS isn't initializing properly it seems. There's a known issue posted in the 386.1 thread post #2, but it isn't clear how widespread the problem is. I could add a qos restart after the 5 minute delay if it would help. Next time it gets into one of these waiting states, login via ssh and run tc qdisc ls and then service restart_qos. If FlexQoS finishes normally within a minute, that will be a positive result. I haven't seen this behavior yet myself, so I'm not sure what it looks like, so post the output of the tc command to confirm my suspicions.

Will do, appreciate it.

 

[sentinelvdx]

Adaptive QoS isn't initializing properly it seems. There's a known issue posted in the 386.1 thread post #2, but it isn't clear how widespread the problem is. I could add a qos restart after the 5 minute delay if it would help. Next time it gets into one of these waiting states, login via ssh and run tc qdisc ls and then service restart_qos. If FlexQoS finishes normally within a minute, that will be a positive result. I haven't seen this behavior yet myself, so I'm not sure what it looks like, so post the output of the tc command to confirm my suspicions.

Dave, didn't experience that on my syslog, but can confirm I'm having the same issue.

When running:
sentinelvdx@RT-AC88U:/tmp/home/root# tc qdisc
qdisc pfifo_fast 0: dev fwd1 root refcnt 2 bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1
qdisc pfifo_fast 0: dev eth0 root refcnt 2 bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1
qdisc pfifo_fast 0: dev eth1 root refcnt 2 bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1
qdisc pfifo_fast 0: dev eth2 root refcnt 2 bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1
qdisc pfifo_fast 0: dev tun21 root refcnt 2 bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1

After restarting QoS through AMTM:
sentinelvdx@RT-AC88U:/tmp/home/root# tc qdisc
qdisc pfifo_fast 0: dev fwd1 root refcnt 2 bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1
qdisc htb 1: dev eth0 root refcnt 2 r2q 10 default 0 direct_packets_stat 160
qdisc fq_codel 800a: dev eth0 parent 1:2 limit 1024p flows 1024 quantum 1514 target 5.0ms interval 100.0ms ecn
qdisc fq_codel 800c: dev eth0 parent 1:10 limit 1024p flows 1024 quantum 1514 target 5.0ms interval 100.0ms ecn
qdisc fq_codel 800e: dev eth0 parent 1:11 limit 1024p flows 1024 quantum 1514 target 5.0ms interval 100.0ms ecn
qdisc fq_codel 8011: dev eth0 parent 1:12 limit 1024p flows 1024 quantum 1514 target 5.0ms interval 100.0ms ecn
qdisc fq_codel 8013: dev eth0 parent 1:13 limit 1024p flows 1024 quantum 1514 target 5.0ms interval 100.0ms ecn
qdisc fq_codel 8015: dev eth0 parent 1:14 limit 1024p flows 1024 quantum 1514 target 5.0ms interval 100.0ms ecn
qdisc fq_codel 8017: dev eth0 parent 1:15 limit 1024p flows 1024 quantum 1514 target 5.0ms interval 100.0ms ecn
qdisc fq_codel 8019: dev eth0 parent 1:16 limit 1024p flows 1024 quantum 1514 target 5.0ms interval 100.0ms ecn
qdisc fq_codel 801b: dev eth0 parent 1:17 limit 1024p flows 1024 quantum 1514 target 5.0ms interval 100.0ms ecn
qdisc pfifo_fast 0: dev eth1 root refcnt 2 bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1
qdisc pfifo_fast 0: dev eth2 root refcnt 2 bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1
qdisc htb 1: dev br0 root refcnt 2 r2q 10 default 0 direct_packets_stat 0
qdisc fq_codel 8001: dev br0 parent 1:2 limit 1024p flows 1024 quantum 1514 target 5.0ms interval 100.0ms ecn
qdisc fq_codel 800b: dev br0 parent 1:10 limit 1024p flows 1024 quantum 1514 target 5.0ms interval 100.0ms ecn
qdisc fq_codel 800d: dev br0 parent 1:11 limit 1024p flows 1024 quantum 1514 target 5.0ms interval 100.0ms ecn
qdisc fq_codel 800f: dev br0 parent 1:12 limit 1024p flows 1024 quantum 1514 target 5.0ms interval 100.0ms ecn
qdisc fq_codel 8012: dev br0 parent 1:13 limit 1024p flows 1024 quantum 1514 target 5.0ms interval 100.0ms ecn
qdisc fq_codel 8014: dev br0 parent 1:14 limit 1024p flows 1024 quantum 1514 target 5.0ms interval 100.0ms ecn
qdisc fq_codel 8016: dev br0 parent 1:15 limit 1024p flows 1024 quantum 1514 target 5.0ms interval 100.0ms ecn
qdisc fq_codel 8018: dev br0 parent 1:16 limit 1024p flows 1024 quantum 1514 target 5.0ms interval 100.0ms ecn
qdisc fq_codel 801a: dev br0 parent 1:17 limit 1024p flows 1024 quantum 1514 target 5.0ms interval 100.0ms ecn
qdisc pfifo_fast 0: dev tun21 root refcnt 2 bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1

 

[dave14305]

I'm pushing an update to develop branch to try to figure out what's wrong in this scenario. It reduces the wait from 5 to 3 minutes and will force a qos restart if 3 minutes are reached. It also logs extra info to the syslog while it's waiting. Would be interested in what appears in the log when this timeout is reached.

I just rebooted by AC86U and cannot reproduce this problem.