Release Asuswrt-Merlin 388.2 is now available for select models

[Deleted member 84281]

Your in good company with my GT-AX6000....
View attachment 50659

It's supposed to be fixed in the latest stock firmware. But looks like it's gonna be quite some time before Merlin gets another release out.

 

[Tech9]

If @shabbs has an issue (the "Uptime Man") - the firmware has an issue, guaranteed.

 

[Wisiwyg]

AX88U, 388.2.2
First day memory usage was high, >90%. A couple of reboots the first day and it has settled down significantly to ~ 49-50%. Only YazFi script running, USB3 swap file set to 5gb via AMTM. 8 days uptime.

 

[shabbs]

If @shabbs has an issue (the "Uptime Man") - the firmware has an issue, guaranteed.

LOL! Meh. I wouldn't call it an "issue" more of an observation. Even at the high memory usage, router is fine. Linux being Linux but it is a different behaviour than previous builds on this router.

EDIT: I was just goofing around on the router earlier today and decided to kill all the "nt_*" processes. The nt_center one was the process that had all the memory. As expected memory went back down to 48% similar to following a reboot. I've yet to see those processes come back to be honest. Which, is interesting. Nothing's blown up... yet... so we'll see. Probably not the best thing to do... was more of a curiosity move.

 

[Tech9]

Is your Traffic Analyzer enabled, TrendMicro thing? There was a bug with ever growing database in jffs, the same nt_center process. No one reported it for 388_22525 firmware though, the base for 388.2. I remember Traffic Analyzer crashing and losing data on AX86U in 388_22525 though. It works now in newer 388_23285, but Web History is still crashing. This TrendMicro bloatware never gets fixed.

 

[shabbs]

Is your Traffic Analyzer enabled, TrendMicro thing? There was a bug with ever growing database in jffs, the same nt_center process. No one reported it for 388_22525 firmware though, the base for 388.2. I remember Traffic Analyzer crashing and losing data on AX86U in 388_22525 though. It works now in newer 388_23285, but Web History is still crashing. This TrendMicro bloatware never gets fixed.

None of those are turned on, nor do I use AiProtection or App Analysis. Will see if this stays steady.

 

[AntonK]

None of those are turned on, nor do I use AiProtection or App Analysis. Will see if this stays steady.

This has to be environmental to your network, assuming nothing sinister such as broken hardware or a damaged OS/firmware install on the router. It appears that I run as little extraneous garbage as you do on your router, and yet here you are and here I am.

 

[Yiannis]

I am having some issues with latest merlin and FlexQoS. More specifically, QoS seems to be out of order , with upload speeds not reported at all. Root cause, as per @dave14305 suggestion seems to be a combination of my setup (WAN type: PPPoE ) and latest Merlin. I do hope this issue is easy to be fixed.

 

[maghuro]

Bug report?

Sometimes, the NVRAM var "custom_clientlist" gets wrong.

The correct syntax is:
custom_clientlist=<NAME>MA:CAD:RE:SS>0>0>>>>

And sometimes, the first entry (only the first one) is saved without the opening "<".

The Asus router deals correctly with it, displaying that device name correctly.
However, some third party services like NextDNS, which rely on the correct syntax, don't process correctly the custom names and doesn't show them in their page, because of that bug.

The solution is change the NAME to another one, and then change it back.
But after some time (I think after a reboot), that var trims again the first <

Thanks.

 

[shabbs]

This has to be environmental to your network, assuming nothing sinister such as broken hardware or a damaged OS/firmware install on the router. It appears that I run as little extraneous garbage as you do on your router, and yet here you are and here I am.

Yeah, probably something on my side that's either augmenting the issue or creating the scenario. I'm in no mood to do a full reset and manual config at this point...

Holding steady at 50% so far since "the killing" heh heh.

 

[John Adler]

Yeah, probably something on my side that's either augmenting the issue or creating the scenario. I'm in no mood to do a full reset and manual config at this point...

Holding steady at 50% so far since "the killing" heh heh.

View attachment 50691

what did you do exactly?
thks

 

[shabbs]

what did you do exactly?
thks

I just killed all the nt_* processes running on my router - ssh'd in, issued a "kill" on the PID associated with those processes. Issuing a "ps | grep nt_" will show you them.

The nt_center process was the big one that was growing and taking up the lion's share of the memory.

The ones I killed were: nt_center, nt_monitor, nt_actMail. So far, no detrimental impact that I can see and the memory has been stable.

Now, I don't recommend doing this. I was just experimenting knowing that if it did impact, I'd just give it a power cycle. I don't believe I'm using anything that actually relies on those processes and I know Eric has mentioned that Asus' implementation of those are "half baked". So... that's my cautionary tale.

 

[John Adler]

Thanks again, I switched AIprotection off and the ram dropped from 615 to 535...
this seems to be normal I asume...
I will follow it a couple of days to see what happens...

 

[karateca]

hi, know if update with the Firmware version 3.0.0.4.388.23285 soon?, because have a big fixed. i be waiting for this, thx for the work.

 

[L&LD]

Only Asus knows. They don't frequent these forums.

 

[karateca]

Only Asus knows. They don't frequent these forums.

asus update 1 week ago, i think must update merlin, maybe merlin wait other gpl, i dont know.

 

[shabbs]

asus update 1 week ago, i think must update merlin, maybe merlin wait other gpl, i dont know.

Will likely be a while until we see that in a Merlin build. When? When it arrives.

 

[Vax]

AIMesh broke on 388.2 couldnt find both ET12s anymore, tried the usual and also upgrading to 388.2_2, still nothing, just downgraded to 388.1, rebootet the ET12s and voila, everything fine again

 

[naznaznaz]

Your in good company with my GT-AX6000....

Remove your USB-connected disk, restart the device, ram usage drops to around 60%. If you use it without a flash disk attached to USB, you will use the device stable with around 60% ram usage. If you do not do anything on the usb with the usb disk inserted, the ram usage will appear as 60%. The slightest operation (copying, writing, moving, etc.) on the usb disk will increase your ram usage up to 97% and it will stay stuck in that state. Asus was unable to fix this situation. Tp link routers do not have such a problem. I am using Asus GT-AX6000. I also own the Tp Link Archer C4000 router.

 

[ColDen]

Remove your USB-connected disk, restart the device, ram usage drops to around 60%. If you use it without a flash disk attached to USB, you will use the device stable with around 60% ram usage. If you do not do anything on the usb with the usb disk inserted, the ram usage will appear as 60%. The slightest operation (copying, writing, moving, etc.) on the usb disk will increase your ram usage up to 97% and it will stay stuck in that state. Asus was unable to fix this situation. Tp link routers do not have such a problem. I am using Asus GT-AX6000. I also own the Tp Link Archer C4000 router.

Well in that case I must have a defective USB flash disk (an old Kingston 16 Gb) because my RAM usage is remaining very stable at 66% with several addons (Skynet, Scribe, Unbound, Connmon, scMerlin, uiScribe, Entware packages). This is not that simple (the router USB port being the issue), as other GT-AX6000 users have also reported RAM usage lower than 75%, since there can be several other modified router settings or connected (wired or wireless) devices that could also be resposible for the reported increase in RAM usage. Additionally it have been mentioned several times by other posters that today's USB flash disks are less reliable then the ones manufactures several years ago and may be one of the possible reason the the reported high RAM usage..