[Release] Asuswrt-Merlin 384.15 (and 384.13_4) are now available

[dave14305]

That would indeed make the md5 check useless if the page is properly implemented (i.e. it gets the correct current_page and next_page variable set). It probably works for the devs currently using it because they don't fully allow applying settings despite not setting these two variables properly. I guess I should remove that parameter from the helper script.

I do see now that they update it via a function instead, so the page can remain static when mounting.

https://github.com/Adamm00/IPSet_ASUS/blob/master/webui/skynet.asp#L325
https://github.com/jackyaz/YazFi/blob/master/YazFi_www.asp#L340

 

[RMerlin]

I do see now that they update it via a function instead, so the page can remain static when mounting.

https://github.com/Adamm00/IPSet_ASUS/blob/master/webui/skynet.asp#L325
https://github.com/jackyaz/YazFi/blob/master/YazFi_www.asp#L340

That's a pretty good workaround then, might be worth updating the Wiki to use that method instead of editing it in the install script.

 

[Dabombber]

Would there be any issues using symlinks instead of copying the files to /www/user? That way "md5sum" could be replaced with "readlink -f --", and it'd be a bit more robust if people start using template/place-holder pages.

 

[Odkrys]

It looks httpd/web.c code need to change the path to show proper LE cert info.

%s.cer instead of cert.pem maybe?

#ifdef RTCONFIG_OPENSSL11      // Kludge as we can't link against libletsencrypt due to different OpenSSL versions
			snprintf(cert_path, sizeof(cert_path), "/jffs/.le/%s/cert.pem", nvram_safe_get("ddns_hostname_x"));
#else

 

[Val D.]

1) 48 is a DFS channel, which means at any time the clients might get bumped off it if a radar is detected

Control channel 48 should be OK actually, it's listed as DFS only for South Africa.

 

[FTC]

Hi, just a short message to post success dirty upgrading my RT-AX88U to Merlin 384.15. Everything working wonderfully, now about to cross the 4 days power up with quite a bit of traffic already processed by the router. I was coming from a record 42 days powered up with 384.14 Final, and the new (384.15 FINAL) build is proving itself even better, since I do no longer see the 'fpmx token invalid' log entries, no wan or wifi problems that I can spot.

I am actively using about 30 (mostly wifi) clients and the following functionality : IPTV (movistar triple vlan profile), samba, minidlna, diversion, openvpn, aiprotection, firewalling. Not using QoS.

In any case, a big thanks (again) and kudos to Merlin and the development team for such a good job!

Regards

 

[JGrana]

Two problems right off the bat:

1) 48 is a DFS channel, which means at any time the clients might get bumped off it if a radar is detected
2) Don't use 20/40, as most people will have too much interference for 40 MHz to work, and will end up downgrading to 20 MHz anyway

Thanks.
My WiFi Analyzer app shows the 5Ghz using channels 34-50. I’m in North America and thought 52-144 should be avoided.
Maybe change to channel 161?
I just changed 2.4 to 20Mhz.
I also will try 384.16 Alpha this weekend.
I am still not 100% convinced it is the router/firmware. My upload rate dropped to 2 Mbit this morning. A reboot brought it back to 10. This on 384.14...
I have installed iperf3 on the AX88U and will also do internal network testing the next time. This issue seems to be WAN related.

 

[MDM]

Thanks.
My WiFi Analyzer app shows the 5Ghz using channels 34-50. I’m in North America and thought 52-144 should be avoided.
Maybe change to channel 161?
I just changed 2.4 to 20Mhz.
I also will try 384.16 Alpha this weekend.
I am still not 100% convinced it is the router/firmware. My upload rate dropped to 2 Mbit this morning. A reboot brought it back to 10. This on 384.14...
I have installed iperf3 on the AX88U and will also do internal network testing the next time. This issue seems to be WAN related.

 

[CaptnDanLKW]

@RMerlin @Jack Yaz
With the inclusion of amtm and the new APIs for addons, more of us are going to start enabling new functionality we might see some issues crop up, for example:

I've noticed with scribe and uiScribe is that when navigating to the system log page, it is becoming unresponsive (and taking longer ) the longer I use it. There's a lot more log content to process and with the separation of log types and expand/collapse options, I suspect there's a lot of client side processing since one of the chrome.exe process pegs a single CPU. This holds up navigation to other tabs, like wireless status, etc.

Is it possible to make the "System Log" Left Side menu not start on the General Log Tab, maybe a new landing page? Not sure how much custom html your going to want to merge with each new release. Just a thought.

Thanks for all you do for the community

 

[RMerlin]

My WiFi Analyzer app shows the 5Ghz using channels 34-50. I’m in North America and thought 52-144 should be avoided.
Maybe change to channel 161?

Test with channel 36 first, the most widely supported channel (I used to have a Zenpad tablet that had major issues when connecting to a high numbered channel like 161 - Asus eventually resolved it by updating the tablet's firmware).

 

[RMerlin]

Is it possible to make the "System Log" Left Side menu not start on the General Log Tab, maybe a new landing page? Not sure how much custom html your going to want to merge with each new release. Just a thought.

Don't want to start making UI changes just for special cases. Everyone has been used to see this log there for the past 7+ years, and they still do when using stock firmware.

The firmware's logger does rotation based on size. If an addon bypasses that behaviour, then it's up to them to handle it, and not me to adapt to them.

 

[Jack Yaz]

Don't want to start making UI changes just for special cases. Everyone has been used to see this log there for the past 7+ years, and they still do when using stock firmware.

The firmware's logger does rotation based on size. If an addon bypasses that behaviour, then it's up to them to handle it, and not me to adapt to them.

Scribe has a robust global default for logrotate, to keep log file sizes down

 

[JGrana]

Test with channel 36 first, the most widely supported channel (I used to have a Zenpad tablet that had major issues when connecting to a high numbered channel like 161 - Asus eventually resolved it by updating the tablet's firmware).

Thanks, just made all the recommended changes. Time will tell ;-)

 

[CaptnDanLKW]

Scribe has a robust global default for logrotate, to keep log file sizes down

I ended up stopping syslog-ng and deleted a couple of the wlceventd.log-date and .gz files. They contained a MASSIVE number of lines... probably a result of numerous wifi changes to mesh, SC and manual tweaking. Probably something normal log rotation wasn't designed to handle.

 

[Val D.]

My WiFi Analyzer app shows the 5Ghz using channels 34-50.

Your WiFi Analyzer app shows you this:
Channel 34 = 32-36 40MHz
Channel 50 = 36-64 160MHz
Something wrong with the app, perhaps?

Most users have:
Channel 42 = 36-48 80MHz
Channel 155 = 149-161 80MHz
Those are non-DFS most common channels.

 

[bilboSNB]

my amtm seems to have stopped working, how do you go about running it now.
'amtm' does not work
'sh amtm' does not work
curl -Os https://diversion.ch/amtm/amtm && sh amtm opens it however.
I have tried the reset option after running the above command.
Any pointers please?

 

[Adamm]

my amtm seems to have stopped working, how do you go about running it now.
'amtm' does not work
'sh amtm' does not work
curl -Os https://diversion.ch/amtm/amtm && sh amtm opens it however.
I have tried the reset option after running the above command.
Any pointers please?

sh /usr/sbin/amtm

 

[fastcx]

my amtm seems to have stopped working, how do you go about running it now.
'amtm' does not work
'sh amtm' does not work
curl -Os https://diversion.ch/amtm/amtm && sh amtm opens it however.
I have tried the reset option after running the above command.
Any pointers please?

unalias amtm && amtm

 

[gmt]

Hi, after the upgrade to 384.15 on my AC-86U the memory usage seems to get pretty high over 90% https://paste.pics/41372da061f4624af89db0e6190b4c9a anyone also experiencing that? On my other router AC-68U which is still on 384.12 the memory usage with the same set of scripts is just around 50% https://paste.pics/93f66df0ce3434973a1c319680cc5050. Top command sorted by %MEM shows syslog-ng as the biggest consumer

Mem: 407228K used, 33192K free, 2632K shrd, 11152K buff, 70652K cached
CPU:  0.1% usr  0.2% sys  0.0% nic 99.6% idle  0.0% io  0.0% irq  0.0% sirq
Load average: 2.32 2.24 2.20 2/167 9063
  PID  PPID USER     STAT   VSZ %VSZ CPU %CPU COMMAND
 2104  2099 user      S     155m 36.0   0  0.0 syslog-ng
 2212     1 nobody   S    47676 10.8   0  0.0 pixelserv-tls 192.168.1.2
25196     1 nobody   S    47144 10.6   1  0.0 dnsmasq --log-async
 1135     1 user      S    19980  4.5   1  0.0 amas_lib
  285     1 user      S    18504  4.1   1  0.0 /bin/swmdk
 1089     1 user      S    13832  3.1   1  0.0 roamast
 3297     1 user      S    12368  2.8   0  0.0 minidlna -f /etc/minidlna.conf -r
 3278     1 user      S    11608  2.6   1  0.0 /usr/sbin/smbd -D -s /etc/smb.conf
 3263     1 user      S    11472  2.6   0  0.0 /usr/sbin/nmbd -D -s /etc/smb.conf
 3264  3263 user      S    11172  2.5   0  0.0 /usr/sbin/nmbd -D -s /etc/smb.conf
  917     1 user      S    11036  2.5   1  0.0 nt_monitor
 1119     1 user      S    10820  2.4   0  0.0 cfg_server
  919     1 user      S    10760  2.4   1  0.0 /sbin/netool
  979     1 user      S    10688  2.4   1  0.0 httpd -i br0
 1008   983 user      S    10436  2.3   0  0.0 vis-dcon

 

[dave14305]

Hi, after the upgrade to 384.15 on my AC-86U the memory usage seems to get pretty high over 90% https://paste.pics/41372da061f4624af89db0e6190b4c9a anyone also experiencing that? On my other router AC-68U which is still on 384.12 the memory usage with the same set of scripts is just around 50% https://paste.pics/93f66df0ce3434973a1c319680cc5050. Top command sorted by %MEM shows syslog-ng as the biggest consumer

Mem: 407228K used, 33192K free, 2632K shrd, 11152K buff, 70652K cached
CPU:  0.1% usr  0.2% sys  0.0% nic 99.6% idle  0.0% io  0.0% irq  0.0% sirq
Load average: 2.32 2.24 2.20 2/167 9063
  PID  PPID USER     STAT   VSZ %VSZ CPU %CPU COMMAND
 2104  2099 user      S     155m 36.0   0  0.0 syslog-ng
 2212     1 nobody   S    47676 10.8   0  0.0 pixelserv-tls 192.168.1.2
25196     1 nobody   S    47144 10.6   1  0.0 dnsmasq --log-async
 1135     1 user      S    19980  4.5   1  0.0 amas_lib
  285     1 user      S    18504  4.1   1  0.0 /bin/swmdk
 1089     1 user      S    13832  3.1   1  0.0 roamast
 3297     1 user      S    12368  2.8   0  0.0 minidlna -f /etc/minidlna.conf -r
 3278     1 user      S    11608  2.6   1  0.0 /usr/sbin/smbd -D -s /etc/smb.conf
 3263     1 user      S    11472  2.6   0  0.0 /usr/sbin/nmbd -D -s /etc/smb.conf
 3264  3263 user      S    11172  2.5   0  0.0 /usr/sbin/nmbd -D -s /etc/smb.conf
  917     1 user      S    11036  2.5   1  0.0 nt_monitor
 1119     1 user      S    10820  2.4   0  0.0 cfg_server
  919     1 user      S    10760  2.4   1  0.0 /sbin/netool
  979     1 user      S    10688  2.4   1  0.0 httpd -i br0
 1008   983 user      S    10436  2.3   0  0.0 vis-dcon

If syslog-ng is installed by Scribe, you may be better off asking in that thread. It wouldn’t be related directly to the 384.15 firmware, unless it’s a symptom of rampant syslog messages from the new version.