Asuswrt-Merlin 376.48 Beta 2 is out

[Ronv42]

I do have a problem. Under 'VPN - Status' I have something showing

'OpenVPN Server 1 - Running Last updated: Sun Oct 19 21:01:24 2014'

I've never ever used VPN on here, I have done a full factory reset, power cycle and it still is showing.....

Go to the VPN Server settings tab and turn off the server. Coming over from stock firmware/resetting the openVPN server starts by default.

 

[XIII]

I have been successfully using DNSCrypt on previous builds, but was surprised to see that nslookup on the router (RT-AC66U) would not respect those settings (see this forumpost).

Since I noticed this in the release notes I thought I should give this beta a try:

   - FIXED: DNS queries run on the router were forwarded to upstream
            nameservers instead of dnsmasq

Now nslookup does not seem to work "at all" anymore:

nslookup forums.smallnetbuilder.com
Server:    127.0.0.1
Address 1: 127.0.0.1 localhost.localdomain

nslookup: can't resolve 'forums.smallnetbuilder.com'

In fact, my DNSCrypt setup also no longer works. I have to override the DNS servers on my PC and iOS devies to use any nodes on the internet. Also, the router is unable to connect to the asuscomm.com DDNS service.

For reference my DNSCrypt config:

/jffs/scripts/wan-start:

#!/bin/sh

logger -t $(basename $0) "started [$@]"

# Now resolve DNS name for NTP server
ntp_name=$(nvram get ntp_server0)
grep "$ntp_name" /jffs/configs/hosts.add > /dev/null 2>&1 || \
for ip in $(/jffs/bin/hostip $ntp_name)
do
    echo $ip $ntp_name >>  /jffs/configs/hosts.add
done

# restart NTP client to eliminate 4-5 mins delay
killall ntp && sleep 1
service restart_dnsmasq
service restart_ntpc

/jffs/bin/dnscrypt-proxy --local-address=127.0.0.1:65053 --daemonize --resolver-address=176.56.237.171:443 --provider-name=2.dnscrypt-cert.resolver1.dnscrypt.eu --provider-key=67C0:0F2C:21C5:5481:45DD:7CB4:6A27:1AF2:EB96:9931:40A3:09B6:2B8D:1653:1185:9C66 --loglevel=5

/jffs/configs/dnsmasq.conf.add:

no-resolv
server=127.0.0.1#65053
server=127.0.0.1#65054

/jffs/configs/hosts.add:

91.207.136.55 pool.ntp.org
194.190.168.1 pool.ntp.org
188.134.15.192 pool.ntp.org
85.21.78.8 pool.ntp.org

Any tips?

 

[SVettel]

Are you using any customized dnsmasq setting? Name resolution from the router is working fine for me, both local and public hostnames.

Yes, I only removed "no-negcache".
I reverted this and the dnsmasq setting is now default, but still no luck. Can't ping nothing from router. I'm using pppoe, I don't know if that has anything to do with this issue.
Never encountered this issue before until .48.
BTW, I think the post above from XIII has the same issue as I do?

 

[bodean]

very bad streaming / buffering over 5ghz with this beta and my fire tv / synology NAS. went back to .47

 

[RMerlin]

Hi all,

I am setting a AC87U up for a friend, I received it today from Amazon and it already has the latest stock firmware update applied, I haven't ran into any problems yet but I was wondering if I should be putting Merlin's firmware on it?

Or is it too early in the cycle and I should stick with stock for the moment?

Thanks

Not a good idea to install a beta version on a router that won't be yours (meaning it will be harder to deal with any new issue, or upgrading it to the final non-beta release).

 

[RMerlin]

very bad streaming / buffering over 5ghz with this beta and my fire tv / synology NAS. went back to .47

Do a full power cycle of your router to reset the QTN radio.

I haven't seen any complain from stock 2769 users versus the 2678 version, so I doubt it's a problem with the 2769 driver.

 

[RMerlin]

Yes, I only removed "no-negcache".
I reverted this and the dnsmasq setting is now default, but still no luck. Can't ping nothing from router. I'm using pppoe, I don't know if that has anything to do with this issue.
Never encountered this issue before until .48.
BTW, I think the post above from XIII has the same issue as I do?

XIII's setup is far from standard, so I don't want to treat it as the same case as yours as his problem might be completely different.

In your case, I can only guess it's related to PPPoE, since my own ISP is DHCP-based.

Post the output from these commands:

cat /etc/resolv.conf
cat /tmp/resolv.conf
cat /tmp/resolv.dnsmasq
cat /etc/dnsmasq.conf

 

[RMerlin]

I have been successfully using DNSCrypt on previous builds, but was surprised to see that nslookup on the router (RT-AC66U) would not respect those settings (see this forumpost).

Since I noticed this in the release notes I thought I should give this beta a try:

   - FIXED: DNS queries run on the router were forwarded to upstream
            nameservers instead of dnsmasq

Now nslookup does not seem to work "at all" anymore:

The only thing this change did was to tell the router to use 127.0.0.1 as its DNS instead of using those from the ISP, so I don't see why this would have any effect on your setup, unless something else in the whole DNSCrypt setup isn't working right.

I've never used DNSCrypt, so I can't help you there. If you remove your DNSCrypt setup do resolutions start working properly again?

 

[SVettel]

Post the output from these commands:

cat /etc/resolv.conf
cat /tmp/resolv.conf
cat /tmp/resolv.dnsmasq
cat /etc/dnsmasq.conf

/etc/resolv.conf:

nameserver 208.67.220.220
nameserver 8.8.8.8

/tmp/resolv.conf

nameserver 208.67.220.220
nameserver 8.8.8.8

/tmp/resolv.dnsmasq

server=2620:0:ccc::2
server=2620:0:ccd::2

/etc/dnsmasq.conf

pid-file=/var/run/dnsmasq.pid
user=nobody
resolv-file=/tmp/resolv.conf
servers-file=/tmp/resolv.dnsmasq
no-poll
min-port=4096
bind-dynamic
interface=br0
interface=ppp1*
no-dhcp-interface=ppp1*
domain=kIssGTR
expand-hosts
bogus-priv
local=/kIssGTR/
no-negcache
cache-size=1500
dhcp-range=lan,192.168.18.100,192.168.18.254,255.255.255.0,86400s
dhcp-option=lan,3,192.168.18.1
dhcp-option=lan,15,kIssGTR
dhcp-option=lan,44,0.0.0.0
dhcp-option=lan,252,"\n"
dhcp-lease-max=155
dhcp-authoritative
read-ethers
addn-hosts=/etc/hosts.dnsmasq
strict-order

 

[RMerlin]

/etc/resolv.conf:

nameserver 208.67.220.220
nameserver 8.8.8.8

Based on this, you are actually even overriding my change, and having your router talk directly to OpenDNS and Google's servers when locally doing a query, instead of using 127.0.0.1.

Do you use DNSFilter, or have any other configuration that would block access to either OpenDNS or Google's DNS servers?

 

[White_Knight]

Need more info about your connecting clients, how you have security configured on your shares, etc...

Hello Merlin,

Could you tell me, which information you need about clients?
SAMBA Configuration file ( smb.conf ) i post here later.
If you need more configuration file , please tell me which you need

 

[SVettel]

Based on this, you are actually even overriding my change, and having your router talk directly to OpenDNS and Google's servers when locally doing a query, instead of using 127.0.0.1.

Do you use DNSFilter, or have any other configuration that would block access to either OpenDNS or Google's DNS servers?

Sorry, the "/etc/resolv.conf" I posted was from .47. In .48 it's:

nameserver 127.0.0.1

Further more, I notice the problem is from the new /etc/resolv.conf. If I add following to "/etc/dnsmasq.conf", problem gone.

no-resolv
server=208.67.220.220

----------------Updated:-------------------
Very wield. After I've done above steps once, and then reverted them all to the default, and from that spot on, this issue never happened again. I rebooted and power-cycled it several times, everything is good so far.
Before I did the above steps, I must go back to .47 to get it work.
Have no idea why this happens.

 

[Pyropetepete]

Go to the VPN Server settings tab and turn off the server. Coming over from stock firmware/resetting the openVPN server starts by default.

This didn't work. I did a factory default via the restore button, then it worked.
It is a lot faster now, not sure if I had a problem.

 

[oversim]

Just flashed on my AC68U. At boot time I mount a NAS partition using this command:

mount \\\\192.168.1.100\\NAS /tmp/mnt/NAS -t cifs -o "username=username,password=password"

Performance of transfers between AC68U and NAS dropped about 10/15%.
Really don't know if CPU is the main bottleneck. During transfer Asus's cores (BTW, I clocked them @1Ghz) never have idle percent below ~50%

 

[RMerlin]

Hello Merlin,

Could you tell me, which information you need about clients?
SAMBA Configuration file ( smb.conf ) i post here later.
If you need more configuration file , please tell me which you need

The operating system of your clients (Windows 7, Windows 8, WDTV, Roku, etc...)
Are you using user-based authentication, or Guest access?
What are the actual result when you try to access the router shares: error message when accessing the disk itself, or only when trying to access a shared folder?

 

[RMerlin]

Just flashed on my AC68U. At boot time I mount a NAS partition using this command:



Performance of transfers between AC68U and NAS dropped about 10/15%.
Really don't know if CPU is the main bottleneck. During transfer Asus's cores (BTW, I clocked them @1Ghz) never have idle percent below ~50%

50% = one core at full time usage.

The performance drop is expected with the Samba upgrade.

 

[RMerlin]

Sorry, the "/etc/resolv.conf" I posted was from .47. In .48 it's:

nameserver 127.0.0.1

Further more, I notice the problem is from the new /etc/resolv.conf. If I add following to "/etc/dnsmasq.conf", problem gone.

no-resolv
server=208.67.220.220

----------------Updated:-------------------
Very wield. After I've done above steps once, and then reverted them all to the default, and from that spot on, this issue never happened again. I rebooted and power-cycled it several times, everything is good so far.
Before I did the above steps, I must go back to .47 to get it work.
Have no idea why this happens.

Can you reproduce the problem if you reboot both the modem and the router? If you do, then it could be a timing issue of some sort.

 

[White_Knight]

The operating system of your clients (Windows 7, Windows 8, WDTV, Roku, etc...)
Are you using user-based authentication, or Guest access?
What are the actual result when you try to access the router shares: error message when accessing the disk itself, or only when trying to access a shared folder?

First of All
I can't see RouterName in Network places, i can't get access when try to enter from command line by \\RouterName or \\RouterName\Share

I use Win7 clients and use user-based authentification in router shares settings

And if i try to chyange something on SAMBA config page i see in log ( this logs you see on my past post ) :

Oct 20 19:44:08 smbd[5240]: [2014/10/20 19:44:08.050488,  0] auth/auth_util.c:841(make_system_session_info_from_pw)
Oct 20 19:44:08 smbd[5240]:   make_server_info_info3 failed with NT code 0xc0000064

 

[SVettel]

Can you reproduce the problem if you reboot both the modem and the router? If you do, then it could be a timing issue of some sort.

No, I can't reproduce it anymore even if downgraded to .47 then back to .48.
The only thing I did was added these two lines in "dnsmasq.conf", and then restarted the dnsmasq. Then everything was up. Before I did that, just can't query dns from router, but can from pc whose dns is the router. And I tried several time yesterday and today, from .47 to .48, with same issue present in .48. I didn't touch the modem during that period.
So absolutely have no idea what happened.

 

[RMerlin]

First of All
I can't see RouterName in Network places, i can't get access when try to enter from command line by \\RouterName or \\RouterName\Share

I use Win7 clients and use user-based authentification in router shares settings

And if i try to chyange something on SAMBA config page i see in log ( this logs you see on my past post ) :

Oct 20 19:44:08 smbd[5240]: [2014/10/20 19:44:08.050488,  0] auth/auth_util.c:841(make_system_session_info_from_pw)
Oct 20 19:44:08 smbd[5240]:   make_server_info_info3 failed with NT code 0xc0000064

These seem to be two separate issues then, one having to deal with the visibility (it works fine for me under 8.1, I'll have to test my Win7 laptop when I get back home), and another having to deal with management of the user/password database.