What's new

[Beta 384/NG] Asuswrt-merlin 384.4 Beta is now available

  • SNBForums Code of Conduct

    SNBForums is a community for everyone, no matter what their level of experience.

    Please be tolerant and patient of others, especially newcomers. We are all here to share and learn!

    The rules are simple: Be patient, be nice, be helpful or be gone!

Status
Not open for further replies.
I've tried reboot, didn't help. I've deleted everything under /etc/cfg_mnt. Router created new ones. I've also deleted everything under /jffs/.le but none of them helped. Just before this "watchdog" uses high CPU and this comes with using more CPU.

If you don't have any idea it's OK I can live with it :)

I see I still use a mixture of Asus's new cfgsync code and my previous workaround, which probably causes the cert to be re-generated everytime a service tries to restart cfgsync (since the location is different). There's a good chance that's what is causing your issue.
 
in my AC86U I have Runner : Disabled!!!How I can re-enable runner???thanks....I tried reboot without success
 
in my AC86U I have Runner : Disabled!!!How I can re-enable runner???thanks....I tried reboot without success

It gets automatically enabled or disabled based on what features the router is running, just like Flow Accelerator in previous generations. You can't control it.
 
The problem was AI Protection.....if I enable it then runner goes to disabled.....if I disable AI Protection and reboot router then runner goes to enable.If I use AIProtection NAT Harware acceleration goses disabled?What are problems with runner disabled?reduced performance?Merlin can you fix the problem in the next releases?
 
The problem was AI Protection.....if I enable it then runner goes to disabled.....if I disable AI Protection and reboot router then runner goes to enable.If I use AIProtection NAT Harware acceleration goses disabled?What are problems with runner disabled?reduced performance?Merlin can you fix the problem in the next releases?

There is nothing to fix there, the two are simply incompatible, which is why runner gets disabled automatically.
 
The size varies for each individual nvram setting, they are documented in shared/defaults.c.

On the RT-AC86U, there's an additional limit of 1000 chars enforced by Broadcom. Asus bypasses that limit for some settings such as dhcp_staticlist by storing them in nvram.

I don't have a list of which, they are handled in the closed source libnvram.

Length is indeed in bytes/char.

Looking at release/src/router/shared/defaults.c from the NG release vs. legacy I see that nvram variable length limitations are defined by the new "CKN_STR" params (which aren't present in the legacy file). So CKN_STR2048 means a 2048 char. limit, etc.

What does CKN_STR_DEFAULT mean here? How about CKN_STR_MAX?

These two entries seem a bit contradictory:
Code:
#   { "wl_maclist", "", CKN_STR_DEFAULT, CKN_TYPE_DEFAULT, CKN_ACC_LEVEL_DEFAULT, CKN_ENC_DEFAULT, 0 },   /* xx:xx:xx:xx:xx:xx ... */
#   { "wl_maclist_x", "", CKN_STR2048, CKN_TYPE_DEFAULT, CKN_ACC_LEVEL_DEFAULT, CKN_ENC_DEFAULT, 0 },   /* xxxxxxxxxxxx...xxxxxxxxxxx */

My notes indicate that these two lists really only differ in the delimiter between MAC addresses. wl_maclist uses a space while wl_maclist_x uses a "<" character. So the lists are essentially the same length, give or take a char. *

Why is one list limited to 2048 chars but not the other?

*(Even if we take the comments on these lines verbatim, then the supposed no-colon format of the MAC addresses in wl_maclist_x would result in a shorter string than wl_maclist, for the same addresses. But those comments are also present in the legacy code, and I think they are out-of-date. In any case, wl_maclist_x seems to accept standard colon-delimted addresses with no complaints...)
 
My notes indicate that these two lists really only differ in the delimiter between MAC addresses. wl_maclist uses a space while wl_maclist_x uses a "<" character. So the lists are essentially the same length, give or take a char. *

Why is one list limited to 2048 chars but not the other?
No...
wl_maclist is a space delimited list of MACs only
wl_maclist_x is a char delimited list including both MACs and hostnames
 
Is the trend micro signature supposed to update automatically?
It seems as of late I have to do a manual update check.
 
What does CKN_STR_DEFAULT mean here? How about CKN_STR_MAX?

Check the other header files in that folder, all constants are defined there.

However, I don't see such a setting on my AC86U. Is this model specific?

Yes. HND models don't need it.

it's work googed in AC87U, but 384.4 beta without telnet, should add telnet option in new version?

Why do you need telnetd?
 
I've tried reboot, didn't help. I've deleted everything under /etc/cfg_mnt. Router created new ones. I've also deleted everything under /jffs/.le but none of them helped. Just before this "watchdog" uses high CPU and this comes with using more CPU.

If you don't have any idea it's OK I can live with it :)

Try this build, I fully reverted that temporary workaround that was no longer necessary:

https://1drv.ms/u/s!AuCcWdNeYuXMjwBXc3pAcPpLtf1r
 
Slapped around the new IPSec VPN off and on most of the day from work to home. Things seem good so far.
 
No...
wl_maclist is a space delimited list of MACs only
wl_maclist_x is a char delimited list including both MACs and hostnames
On my RT-AC68P, wl_maclist_x is (currently) a list of MAC addresses with "<" character prepended to each address (no spaces). But these were previously assigned via script, and I could have made a mistake -- though it all seems to be working correctly at the moment.

Maybe you mean custom_clientlist ?
 
I see it on my AC-68U under tools->Other Settings->Advanced tweaks and Hacks... but NOT on my AC86U. Both are running 384.3
I used an old script from my 87 on my 86 that turns off the memory management (RAM usage is around 75%) compared to when its on (around 50% or less) by changing the NVRAM setting.

My guess is the setting is still there on the 86 in NVRAM and it works. There is just no UI method to toggle it.

nvram set drop_caches=0

That shuts it off IIRC.
 
On my RT-AC68P, wl_maclist_x is (currently) a list of MAC addresses with "<" character prepended to each address (no spaces). But these were previously assigned via script, and I could have made a mistake -- though it all seems to be working correctly at the moment.

Maybe you mean custom_clientlist ?
Ahhh....now it's coming back. It's as I described on my fork (374 base). Around 378 they removed the hostname from maclist_x and did a lookup into clientlist (that's why a lot of people broke during that upgrade)
 
  • Like
Reactions: SNR
Status
Not open for further replies.

Sign Up For SNBForums Daily Digest

Get an update of what's new every day delivered to your mailbox. Sign up here!
Top