CakeQOS CakeQOS-Merlin

[Treadler]

Did that, option 1, then I go ahead and click option 3, tells me it not running, when I do a test bufferboat is a B
Was getting A before I uninstalled older version

Hmmm, one of the more learned people on here will need to address that one. Maybe try option #5? See if there’s some new goodness there for you to install/fix things?

 

[jackiechun]

Did that, option 1, then I go ahead and click option 3, tells me it not running, when I do a test bufferboat is a B
Was getting A before I uninstalled older version

EDIT: got it working , uninstalled it again, then tired installing it again(worked)

If you run into this again, could you post the output of: tc qdisc

Thanks!

EDIT:

there is a typo:
"Select Queue Prioity Type:"

Thanks, corrected

 

[New2This]

If you run into this again, could you post the output of: tc qdisc

Thanks!

Sure thing

 

[Adamm]

Any users having an issue where cake doesn't survive a reboot because the nat-start entry is missing, update the script as a hotfix was pushed out yesterday then manually start cake.

 

[ttgapers]

Did that, option 1, then I go ahead and click option 3, tells me it not running, when I do a test bufferboat is a B
Was getting A before I uninstalled older version

EDIT: got it working , uninstalled it again, then tired installing it again(worked)

Survived the reboot as well --- see above. Thanks @Adamm

 

[Treadler]

Any users having an issue where cake doesn't survive a reboot because the nat-start entry is missing, update the script as a hotfix was pushed out yesterday then manually start cake.

I did an update check, started Cake, checked status was = running, rebooted, Cake not running.

Uninstalled & reinstalled. Set up, Cake status = running.
Rebooted, Cake not running.
What am I doing wrong?

 

[kernol]

I did an update check, started Cake, checked status was = running, rebooted, Cake not running.

Uninstalled & reinstalled. Set up, Cake status = running.
Rebooted, Cake not running.
What am I doing wrong?

Not sure whether this has anything to do with the problem [I have it too] - but in my RT-AC86U I did not have any port forwards or other tweaks under the WAN Tab in settings - so interestingly in JFFS folder I did not have a nat-start script at all.

When the revised startup process was first scripted here for cake-qos [to use a nat-start script] it failed for me - but then with a further update it created the nat-start file and populated it with cake-qos start. For a while I had that cake-qos start disappearing from nat-start [after tweaking settings under menu]- but that was fixed.

I am now wondering whether the problem in not simply - nothing calls nat-start since there doesn't seem to be a reason to start it in the absence of any port forwards???

 

[Treadler]

Not sure whether this has anything to do with the problem [I have it too] - but in my RT-AC86U I did not have any port forwards or other tweaks under the WAN Tab in settings - so interestingly in JFFS folder I did not have a nat-start script at all.

When the revised startup process was first scripted here for cake-qos [to use a nat-start script] it failed for me - but then with a further update it created the nat-start file and populated it with cake-qos start. For a while I had that cake-qos start disappearing from nat-start [after tweaking settings under menu]- but that was fixed.

I am now wondering whether the problem in not simply - nothing calls nat-start since there doesn't seem to be a reason to start it in the absence of any port forwards???

No port forwards here.

 

[kernol]

No port forwards here.

Ahh .. you see in the Merlin wiki here https://github.com/RMerl/asuswrt-merlin.ng/wiki/User-scripts it suggests that nat-start only gets kicked into action after NAT rules have been applied to the NAT Table !! - so that may be it ... we don't have any Nat rules ???

Also - NB - if you stop cake-qos [command line or menu item] the startup command for cake-qos is removed from nat-start.
So unless you start cake-qos again BEFORE you reboot ... it will not automagically start after rebooting.

 

[Treadler]

Ahh .. you see in the Merlin wiki here https://github.com/RMerl/asuswrt-merlin.ng/wiki/User-scripts it suggests that nat-start only gets kicked into action after NAT rules have been applied to the NAT Table !! - so that may be it ... we don't have any Nat rules ???

Also - NB - if you stop cake-qos [command line or menu item] the startup command for cake-qos is removed from nat-start.
So unless you start cake-qos again BEFORE you reboot ... it will not automagically start after rebooting.

Makes sense, so the reboot would preserve whatever the current status quo is.
Re Nat rules, my network is very ordinary, no bells & whistles.

 

[kernol]

Makes sense, so the reboot would preserve whatever the current status quo is.
Re Nat rules, my network very ordinary, no bells & whistles.

The plot thickens - in my case the startup line for cake-qos within nat-start is NOT surviving a reboot. I check that it is there before rebooting - then reboot and check after everything up [firewall YasFi etc] - and ... nat-start is empty ... and cake-qos is not running for obvious reasons.

This AFTER I had placed a dummy Port Forward in to trigger Nat Table change and force nat-start to run.

Clearly still some bugs in the cake-qos script - so will chill until the coders can get to it.

 

[Stephen Harrington]

Also - NB - if you stop cake-qos [command line or menu item] the startup command for cake-qos is removed from nat-start.
So unless you start cake-qos again BEFORE you reboot ... it will not automagically start after rebooting.

Fair enough I guess - but just observing that's not consistent with how the other major AMTM scripts work is it?
Or maybe it is? Am I confused?

 

[kernol]

Fair enough I guess - but just observing that's not consistent with how the other major AMTM scripts work is it?
Or maybe it is? Am I confused?

Some do and some don't -
for e.g. Skynet has a "temporary disable" item - and in that case - a reboot will bring up Skynet again without any user intervention.

Open to correction here - but as I recall - if you stop Diversion - it stays stopped even through a reboot.

I prefer the Skynet option - so I don't have to remember on reboot to re-engage certain add-ons .

 

[kernol]

Ahh .. you see in the Merlin wiki here https://github.com/RMerl/asuswrt-merlin.ng/wiki/User-scripts it suggests that nat-start only gets kicked into action after NAT rules have been applied to the NAT Table !! - so that may be it ... we don't have any Nat rules ???

Also - NB - if you stop cake-qos [command line or menu item] the startup command for cake-qos is removed from nat-start.
So unless you start cake-qos again BEFORE you reboot ... it will not automagically start after rebooting.

Answering my own post - I know {face-plant} ... but I went back on my logs and discovered that if nat-start file exists in JFFS - it will automagically run after firewall-start. So on 24 June the cake-qos install script created my first nat-stat file in JFFS and it has been triggered to run ever since. Problem is ... apart from the bash 1st line - its been empty when it gets to reboot stage ??

 

[Treadler]

Some do and some don't -
for e.g. Skynet has a "temporary disable" item - and in that case - a reboot will bring up Skynet again without any user intervention.

Open to correction here - but as I recall - if you stop Diversion - it stays stopped even through a reboot.

I prefer the Skynet option - so I don't have to remember on reboot to re-engage certain add-ons .

It’s hardly a show stopper, just something that has to be remembered after a reboot.
(Until/unless the coders come up with an elegant solution, the way they do!)

 

[Stephen Harrington]

I prefer the Skynet option - so I don't have to remember on reboot to re-engage certain add-ons .

I think I'm with you on that one!

 

[Stephen Harrington]

It’s hardly a show stopper, just something that has to be remembered after a reboot.

You're right, ultimately not a biggie, just a potential "trap" and hopefully as this evolves there may be a more elegant way?
But it's come such a long way in such a short time, definitely not complaining!

 

[kernol]

Any users having an issue where cake doesn't survive a reboot because the nat-start entry is missing, update the script as a hotfix was pushed out yesterday then manually start cake.

I did an update check, started Cake, checked status was = running, rebooted, Cake not running.

Uninstalled & reinstalled. Set up, Cake status = running.
Rebooted, Cake not running.
What am I doing wrong?

@Adamm - followed same process as as Treadler several times this morning - no joy ... cannot get cake to survive a reboot [no matter how many uninstalls and reinstalls done].

Services-stop calls a stop to cake and that routine seems to remove cake start from nat-start [may be wrong - not a coder].

 

[Adamm]

Some do and some don't -
for e.g. Skynet has a "temporary disable" item - and in that case - a reboot will bring up Skynet again without any user intervention.

Personally that's how I believe the feature should also be, but this wasn't my decision.

Services-stop calls a stop to cake and that routine seems to remove cake start from nat-start [may be wrong - not a coder].

Good catch, that is most definitely the issue. I've pushed a hotfix which should hopefully differentiate the system and user calling the stop command and only remove system file entries for the latter. If you wouldn't mind testing it that would be appreciated, otherwise I will revert the feature to my original implementation where it is only a temporary stop.

 

[Wade Coxon]

Good catch, that is most definitely the issue. I've pushed a hotfix which should hopefully differentiate the system and user calling the stop command and only remove system file entries for the latter. If you wouldn't mind testing it that would be appreciated, otherwise I will revert the feature to my original implementation where it is only a temporary stop.

Sounds like we really need a "enable" and "disable" command to differentiate those actions from "start" and "stop"