Asuswrt-Merlin 378.53 is now available

[kvic]

Thanks! I experienced the same Apple TV issue. This resolves my problem.

Since Apple introduced many network problems in Yosemite/iOS 8 release, I didn't bother to track down the issue. Resorted to a remote for waking up Apple TV.

So great that you figured out the cause.

On an anecdotal note, this seems to also resolve another wake up from sleep issue. I have mounted a Mac folder to RT-AC56U using NFS. The router runs a script periodically to write data to this folder. Once the Mac enters sleep, data write to the folder won't wake it up. After change to Asus mode, the Mac wakes up and data write succeeds. Still watching. Not 100% sure since the script only runs a few times after change.

Just wanted to jump in and share that I have finally resolved my AirPlay not working/Apple TV not waking from sleep with iOS remote issues that have been bothering me on and off for the last month or two (always running latest Merlin), I was convinced my Apple TV was the problem but after setting NAT loopback to Asus the device works perfectly for the last few days (tried disabling first with no joy).

I went through a lot of trial an error with router settings and saw the issue regardless if the device was connected via cable or wifi.

I would expect this to be affecting quite a few others, but if you don't have an Apple TV then sorry for the spam!

 

[ChikkSpot]

NAT loopback was not working for me in this release.... It was set to Merlin, I had to change it to ASUS and it started working again.

Asus's seem to be working again, although Merlin's works for me too. (AC68R)

 

[northumberland]

Having some issues with wireless suddenly dropping out. Had this issue some months back too. Wireless shows to clients but will never connect. Reboot or change wifi settings/apply fixes it temp.

I'm running 378.53 on an RT-N66U. 5G Seems to work okay, but 2.4 frequently drops (~3 in oast 7 days). Making a change to wireless settings restarts the interface. I do have a lot of clients connected, but most of them are wired. CPU and RAM have plenty resources available.

This morning when I checked, the 'NETWORK MAP' tab it didn't even show wireless at all (because it was down?). The upper icon showing the wifi status (2.4/5) was lit as if interfaces were available.

best way from command line to updown the interface as a workaround in the meantime? also anywhere I shoud be checking logs so I can better report on this?

Thanks
Steve

 

[kvic]

I think Merlin still stands well on his NAT loopback. It's on the same league as "Asus" and "Disable" with regard to wake up Apple TV. From my further check, this setting doesn't create nor resolve the issue for me.

What I found is that when Apple TV fails to wake up it doesn't have an IP address in Bonjour service discovery. Though the device is still associated with the wlan and is assigned an IP address. Also router's arp cache has a relevant entry. I don't know what's the root cause but seems Bonjour SD cache is corrupted shortly after Apple TV goes into sleep mode.

Above is the good old way Airplay works over wlan/lan. In Yosemite/iOS 8, Apple introduced peer-to-peer airplay. It uses bluetooth for service discovery and move trunk data over wifi/lan. I suspect people not seeing the issue may have their bluetooth on and thus uses the peer-to-peer mode. I'm consistently able to wake up Apple TV when play a video this way.

Yosemite/iOS8 create many network/wifi issues from migrating mDNS to discoveryd. Keep whatever tricks work for you and let Apple sort out the mess...

Thanks! I experienced the same Apple TV issue. This resolves my problem.

Since Apple introduced many network problems in Yosemite/iOS 8 release, I didn't bother to track down the issue. Resorted to a remote for waking up Apple TV.

So great that you figured out the cause.

On an anecdotal note, this seems to also resolve another wake up from sleep issue. I have mounted a Mac folder to RT-AC56U using NFS. The router runs a script periodically to write data to this folder. Once the Mac enters sleep, data write to the folder won't wake it up. After change to Asus mode, the Mac wakes up and data write succeeds. Still watching. Not 100% sure since the script only runs a few times after change.

 

[Sanchu]

QOS functionality is not working correctly on 87u and i am using the latest merlin version (378.53). If i enable smart QOS download speeds drops to 1Mbps and if i disable qos i get 13 Mbps. Does any body have experienced similar problem. Is this a hardware fault or the software problem ..

 

[triglia118]

@RMerlin : dear Merlin, thank you for your hard work, I'm running 378.53 on an RT-N66U everything is ok, except for this strange issue: http://www.snbforums.com/threads/usb-disk-configuration-folders-tree-missing.24274/ can you have a look? thank you for your time

 

[chudak]

QOS functionality is not working correctly on 87u and i am using the latest merlin version (378.53). If i enable smart QOS download speeds drops to 1Mbps and if i disable qos i get 13 Mbps. Does any body have experienced similar problem. Is this a hardware fault or the software problem ..

I used the same router and did not see QOS relations, but WiFi performance is definitely degraded in 378.53, 24. has shorter range and strenth

 

[Sanchu]

I used the same router and did not see QOS relations, but WiFi performance is definitely degraded in 378.53, 24. has shorter range and strenth

I had the same problem with my first ac87u then i got a replacement ac87u router. Both routers behavior is same . I am not sure where the problem is. This behavior is exactly same. @RMerlin Did you see this behavior before ?

 

[chudak]

I had the same problem with my first ac87u then i got a replacement ac87u router. Both routers behavior is same . I am not sure where the problem is. This behavior is exactly same. @RMerlin Did you see this behavior before ?

I am considering downgrading to 378.50, my hope was that newer firmware is better, but alas.

@RMerlin is WiFi issues in 378.53 confirmed problem?

 

[Sanchu]

I am considering downgrading to 378.50, my hope was that newer firmware is better, but alas.

@RMerlin is WiFi issues in 378.53 confirmed problem?

I agree with the above. WIFI performance is degraded for sure...

 

[RMerlin]

I am considering downgrading to 378.50, my hope was that newer firmware is better, but alas.

@RMerlin is WiFi issues in 378.53 confirmed problem?

You never mentioned which router model you are referring to, so I can't answer that question...

 

[chudak]

You never mentioned which router model you are referring to, so I can't answer that question...

@RMerlin ^

It was somewhere above ☺
It's ac87u

Do you plan to release a new update soon or maybe an experiment build to try out?

 

[Sanchu]

@RMerlin Did you come across the mentioned QOS problem (POST 226 and 229 ) of this thread for my rt 87u . if you need i can provide the system logs. Please help me

 

[Sanchu]

I am seeing another strange behavior on my rt 87u on this merlin firmware . If the enable vpn and then disable vpn couple of times then the router cant resolve the domain names. i have set the wan dns as google dns. I can ping ip addresses and cannot ping domain names Any body come across this

May 10 12:37:22 openvpn[1706]: RESOLVE: Cannot resolve host address: del-c02.ipvanish.com: Name or service not known
May 10 12:37:27 rc_service: httpd 1767:notify_rc stop_vpnclient1
May 10 12:37:36 rc_service: httpd 1767:notify_rc start_vpnclient1
May 10 12:37:37 kernel: tun: Universal TUN/TAP device driver, 1.6
May 10 12:37:37 kernel: tun: (C) 1999-2004 Max Krasnyansky <maxk@qualcomm.com>
May 10 12:37:38 openvpn[1873]: DEPRECATED OPTION: --tls-remote, please update your configuration
May 10 12:37:38 openvpn[1873]: OpenVPN 2.3.6 arm-unknown-linux-gnu [SSL (OpenSSL)] [LZO] [EPOLL] [MH] [IPv6] built on Apr 26 2015
May 10 12:37:38 openvpn[1873]: library versions: OpenSSL 1.0.0r 19 Mar 2015, LZO 2.08
May 10 12:37:38 openvpn[1873]: NOTE: the current --script-security setting may allow this configuration to call user-defined scripts
May 10 12:37:38 openvpn[1873]: Socket Buffers: R=[122880->131072] S=[122880->131072]
May 10 12:37:53 openvpn[1873]: RESOLVE: Cannot resolve host address: del-c02.ipvanish.com: Name or service not known
May 10 12:38:08 openvpn[1879]: RESOLVE: Cannot resolve host address: del-c02.ipvanish.com: Name or service not known
May 10 12:38:28 openvpn[1879]: RESOLVE: Cannot resolve host address: del-c02.ipvanish.com: Name or service not known
May 10 12:38:48 openvpn[1879]: RESOLVE: Cannot resolve host address: del-c02.ipvanish.com: Name or service not known
May 10 12:39:08 openvpn[1879]: RESOLVE: Cannot resolve host address: del-c02.ipvanish.com: Name or service not known
May 10 12:39:28 openvpn[1879]: RESOLVE: Cannot resolve host address: del-c02.ipvanish.com: Name or service not known
May 10 12:39:48 openvpn[1879]: RESOLVE: Cannot resolve host address: del-c02.ipvanish.com: Name or service not known
May 10 12:39:48 openvpn[1879]: SIGUSR1[soft,init_instance] received, process restarting
May 10 12:39:48 openvpn[1879]: Restart pause, 2 second(s)
May 10 12:39:50 openvpn[1879]: NOTE: the current --script-security setting may allow this configuration to call user-defined scripts
May 10 12:39:50 openvpn[1879]: Socket Buffers: R=[122880->131072] S=[122880->131072]
May 10 12:40:05 openvpn[1879]: RESOLVE: Cannot resolve host address: del-c02.ipvanish.com: Name or service not known
May 10 12:40:20 openvpn[1879]: RESOLVE: Cannot resolve host address: del-c02.ipvanish.com: Name or service not known
May 10 12:40:40 openvpn[1879]: RESOLVE: Cannot resolve host address: del-c02.ipvanish.com: Name or service not known
May 10 12:41:00 openvpn[1879]: RESOLVE: Cannot resolve host address: del-c02.ipvanish.com: Name or service not known
May 10 12:41:20 openvpn[1879]: RESOLVE: Cannot resolve host address: del-c02.ipvanish.com: Name or service not known
May 10 12:41:32 rc_service: httpd 1767:notify_rc stop_vpnclient1

 

[Sanchu]

Looks like there is a bug indroduced in this area. /tmp/resolv.conf contains the vpn dns entries and it is not cleared when i disable the vpn service
nameserver 198.18.0.1
nameserver 198.18.0.2

nameserver 8.8.8.8
nameserver 8.8.4.4

First 3 lines should be deleted after terminating the vpn connection and this is not happening

 

[Sanchu]

I added a script to remove the unwanted vpn dns entries and restart the dnsmasq . If any boy sees any problem let me know
#!/bin/sh
NO_OF_LINES="$(cat resolv.conf | wc -l)"
echo "No of lines == $NO_OF_LINES"
if [ $NO_OF_LINES -eq 5 ]; then
sed -i '1,3d' resolv.conf
ps | grep dnsmasq | grep -v grep | awk '{print "kill " $1}' | sh
dnsmasq --log-async
fi

 

[Martineau]

Looks like there is a bug indroduced in this area. /tmp/resolv.conf contains the vpn dns entries and it is not cleared when i disable the vpn service
nameserver 198.18.0.1
nameserver 198.18.0.2

nameserver 8.8.8.8
nameserver 8.8.4.4

First 3 lines should be deleted after terminating the vpn connection and this is not happening

I believe RMerlin identified at least one situation where the openvpn client can no longer run its 'clean-up' script when the VPN client is terminated.

When I terminate my VPN client, I see the call to update/restart dnsmasq in my Syslog:

May 10 10:06:23 RT-AC56U daemon.notice openvpn[5133]: updown.sh tun11 1500 1541 xxx.xxx.xxx.xxx 255.255.252.0 init

May 10 10:06:23 RT-AC56U kern.emerg rc_service: service 5180:notify_rc updateresolv

Perhaps you have a similar 'invalid/incompatible' directive in the Custom Client Configuration?

http://www.snbforums.com/threads/asuswrt-merlin-378-53-is-now-available.24172/page-8#post-180828

 

[Sanchu]

I believe RMerlin identified at least one situation where the openvpn client can no longer run its 'clean-up' script when the VPN client is terminated.

When I terminate my VPN client, I see the call to update/restart dnsmasq in my Syslog:

May 10 10:06:23 RT-AC56U daemon.notice openvpn[5133]: updown.sh tun11 1500 1541 xxx.xxx.xxx.xxx 255.255.252.0 init

May 10 10:06:23 RT-AC56U kern.emerg rc_service: service 5180:notify_rc updateresolv

Perhaps you have a similar 'invalid/incompatible' directive in the Custom Client Configuration?

http://www.snbforums.com/threads/asuswrt-merlin-378-53-is-now-available.24172/page-8#post-180828

What you said is correct. My logs looks like the following when i stop vpn
May 10 14:25:35 rc_service: httpd 1767:notify_rc stop_vpnclient1
May 10 14:25:35 openvpn[3449]: event_wait : Interrupted system call (code=4)
May 10 14:25:35 openvpn[3449]: SIGTERM received, sending exit notification to peer


How to identify the invalid/incompatible configuration . I am following instructions from this page https://vpntips.com/vpn-router-install/

 

[Martineau]

What you said is correct. My logs looks like the following when i stop vpn
May 10 14:25:35 rc_service: httpd 1767:notify_rc stop_vpnclient1
May 10 14:25:35 openvpn[3449]: event_wait : Interrupted system call (code=4)
May 10 14:25:35 openvpn[3449]: SIGTERM received, sending exit notification to peer


How to identify the invalid/incompatible configuration . I am following instructions from this page https://vpntips.com/vpn-router-install/

If you're using IPVanish, and all of the directives are mandatory, then I guess you will need to attempt to implement your own down script (in lieu of the failure to run the openvpn updown.sh script.) although I don't see how you would automatically trigger it if openvpn has crashed?

However I would try to start the VPN client with just one custom directive, and tediously work through them until you find the actual one (or combination) that causes the error.

 

[john9527]

Merlin has already coded a fix for the 'IPVanish' case for the next release.....
https://github.com/RMerl/asuswrt-merlin/commit/81909a28ec050d0f45613f6e63bdea45b34672d3