[Release] Asuswrt-Merlin 384.10 is now available

[tallytr]

Starting with 384.10 both my routers have frozen several times. Never happened before. Also my son (playing games) is complaining that "internet disappears" several times / day. Uploaded 384.10.2 yesterday evening. Lets hope for the best.

Happened to me as well. I went back to 384.9 and restored settings from saved 384.9 file, then rebooted router and uploaded 384.10, unplugged power for a minute and powered up again, perfect performance. Uploaded 384.10.2 without any issues after first rebooting router....unmount USB drives before upgrading...

 

[Xentrk]

Happened to me as well. I went back to 384.9 and restored settings from saved 384.9 file, then rebooted router and uploaded 384.10, unplugged power for a minute and powered up again, perfect performance. Uploaded 384.10.2 without any issues after first rebooting router....unmount USB drives before upgrading...

A good reminder for all on the importance of taking a back up copy of the existing configuration before performing the firmware update in case Murphy shows up and you have to revert to the prior version.

 

[maxbraketorque]

Happened to me as well. I went back to 384.9 and restored settings from saved 384.9 file, then rebooted router and uploaded 384.10, unplugged power for a minute and powered up again, perfect performance. Uploaded 384.10.2 without any issues after first rebooting router....unmount USB drives before upgrading...

I was seeing this for 384.10_0 as well. So far I have not seen this with 384.10_2.

 

[TheKolaNN]

Starting with 384.10 both my routers have frozen several times. Never happened before. Also my son (playing games) is complaining that "internet disappears" several times / day. Uploaded 384.10.2 yesterday evening. Lets hope for the best.

Me too have similar issue.
I have got set up a Reboot Scheduler at 5 AM every day, and according to logs, last entry before my manual reboot I perform when I wake up is at 5 AM... Suggests my RT-AC88U is rebooting abnormally and crashes. Happened since 384.10.0 and in 384.10.2 too. What I see when I wake up is that both wi-fi radios are down and the internet icon is not white, but looks like amber color. My PC on wired connection cannot resolve IP. No internet access.
I could not reproduce the issue when I was in front of a computer changing rebooting time 2-5 minutes ahead of current time though.
Yesterday I installed 384.10.2 clean install and reconfigured router back to what is was manually (took me 90 minutes to reconfigure), but today morning I had the same issue. I guess its not dirty install issue but the software itself.

 

[L&LD]

Me too have similar issue.
I have got set up a Reboot Scheduler at 5 AM every day, and according to logs, last entry before my manual reboot I perform when I wake up is at 5 AM... Suggests my RT-AC88U is rebooting abnormally and crashes. Happened since 384.10.0 and in 384.10.2 too. What I see when I wake up is that both wi-fi radios are down and the internet icon is not white, but looks like amber color. My PC on wired connection cannot resolve IP. No internet access.
I could not reproduce the issue when I was in front of a computer changing rebooting time 2-5 minutes ahead of current time though.
Yesterday I installed 384.10.2 clean install and reconfigured router back to what is was manually (took me 90 minutes to reconfigure), but today morning I had the same issue. I guess its not dirty install issue but the software itself.

I think that a lot of the issues and symtoms being reported are because the routers have not had a full reset to factory defaults followed by a minimal and manual configuration to secure the router and connect to the ISP (or in other words; a full M&M Config, please see the links in my signature below for further details).

Why do you need a reboot schedule? Is it masking a problem that is there for a long time?

When was the last time a full M&M Config was performed on the network equipment?

Many others have seen their problems disappear when the router and the rest of the network is put into a good/known state.

Flashing firmware, multiple times over and over won't fix any of the underlying issue(s), if those issues are unknown interactions between bits of code from different firmware that may be used differently by the same-named variables over time. An M&M Config done properly and in order will ensure that the issues are not inter-firmware related. Any bugs or glitches shown will either be real or they will be hardware based (indicating an RMA, if possible).

 

[TheKolaNN]

Why do you need a reboot schedule? Is it masking a problem that is there for a long time?

When was the last time a full M&M Config was performed on the network equipment?

Many others have seen their problems disappear when the router and the rest of the network is put into a good/known state.

There is a good reason behind that. My ISP is renewing WAN IP address every 24 hours and I do not want this to be done the exact hour I turned on the router (most likely during the day or working hours). Restart at 5 AM ensures me that the IP will not change when I am using the internet.
I can always go back before 384.10.0, where it was all good, but it makes no sense since the firmware will not become better if I do not indicate the problem, which this FW certainly has (I am not the only one with it).

EDIT:
So I managed to reproduce the issue. It was 10:31 AM and I changed the Reboot Scheduler time to 10:40. The router just changed the color or the internet icon to amber, not restarted at all. GUI is inaccessible. Last log entries before my hard restart:
Apr 5 10:38:42 ovpn-server1[1198]: 95.40.252.7:48647 SIGUSR1[soft,connection-reset] received, client-instance restarting
Apr 5 10:40:11 kernel: SysRq : Emergency Sync
Apr 5 10:40:11 kernel: Emergency Sync complete

 

[manol68]

384.10_2 / RT-AX88U

Apr  5 03:50:10 rc_service: amas_lib 1216:notify_rc restart_firewall
Apr  5 03:50:10 nat: apply nat rules (/tmp/nat_rules_eth0_eth0)
Apr  5 10:14:26 kernel: eth0 (Int switch port: 3) (Logical Port: 3) (phyId: c) Link DOWN.
Apr  5 10:14:30 kernel: eth0 (Int switch port: 3) (Logical Port: 3) (phyId: c) Link UP at 1000 mbps full duplex
Apr  5 10:15:01 kernel: eth0 (Int switch port: 3) (Logical Port: 3) (phyId: c) Link DOWN.
Apr  5 10:15:04 DualWAN: skip single wan wan_led_control - WANRED off
Apr  5 10:15:06 kernel: eth0 (Int switch port: 3) (Logical Port: 3) (phyId: c) Link UP at 1000 mbps full duplex
Apr  5 10:15:09 WAN_Connection: Ethernet link up.
Apr  5 10:15:09 rc_service: wanduck 1035:notify_rc restart_wan_if 0
Apr  5 10:15:11 kernel: eth0 (Int switch port: 3) (Logical Port: 3) (phyId: c) Link DOWN.
Apr  5 10:15:11 nat: apply nat rules (/tmp/nat_rules_eth0_eth0)
Apr  5 10:15:11 wan: finish adding multi routes
Apr  5 10:15:13 kernel: eth0 (Int switch port: 3) (Logical Port: 3) (phyId: c) Link UP at 1000 mbps full duplex
Apr  5 10:15:21 WAN_Connection: WAN was restored.

Why firewall is being restarted in the middle of the night or every few hours ? I have seen that at least few times now, but i have cleared the system logs already. It has never happened before. Also, some weird things is happening to the WAN connection ? Also never seen that in the system log before...

 

[telUK]

Have been running 384.8_2 for awhile and today updated to 384.10_2. The Network Map/View Client that has been broken for awhile (Asus issue) is working now. This version seems to be rock solid. As others have reported, my memory usage is lower (79% vs 90's% before) in this version even though I'm running more things than ever including the 2 newest scripts, ntpMerlin and spdMerlin. I do realize the way Linux handles memory differently so not sure if it's relevant.

Still on 384.8_2 myself but think I will update soon and reset after and manually input all my settings again.

But at the same time I have been happy on 384.8_2, been rock solid for a while and I believe my ac86u states an uptime of about 50+ days.

 

[Martineau]

RIP User NVRAM Save/Restore Utility
You are greatly missed.

I still use a custom version of @john9527's scripts.

For /jffs backup/restores, I've never stopped using it - useful for taking snapshots when developing scripts locally.

Spoiler: JFFS management

export OLDPWD=`pwd`;cd /mnt/`nvram get productid`/BackupScriptsJFFS;./jffs-restore.sh;cd $OLDPWD

jffs-restore.sh: JFFS Restore Utility - Version 26.1

1. jffs-201709121645_RT-AC68U-E4A0/
2. jffs-201709181925_RT-N66U-C978/
3. jffs-201809101938_RT-AC68U-E4A0/
4. jffs-201903281231_RT-AC68U-E4A0/
5. jffs-201903281529_RT-AC68U-E4A0/
6. jffs-201903281714_RT-AC68U-E4A0/
7. jffs-201903281717_RT-AC68U-E4A0/
8. jffs-201903281750_RT-AC68U-E4A0/
9. jffs-201903281925_RT-AC68U-E4A0/
10. jffs-201903282247_RT-AC68U-E4A0/
11. jffs-201903290928_RT-AC68U-E4A0/
12. jffs-201903290956_RT-AC68U-E4A0/
13. jffs-201903301256_RT-AC68U-E4A0/
14. jffs-201903301302_RT-AC68U-E4A0/
15. jffs-201903302205_RT-AC68U-E4A0/
16. jffs-201903310749_RT-AC68U-E4A0/
17. jffs-201904021101_RT-AC68U-E4A0/
18. jffs-201904021105_RT-AC68U-E4A0/
19. jffs-201904021444_RT-AC68U-E4A0/
20. jffs-201904021550_RT-AC68U-E4A0/
21. jffs-201904022039_RT-AC68U-E4A0/
22. jffs-201904031110_RT-AC68U-E4A0/
23. jffs-201904031743_RT-AC68U-E4A0/

 Restore last saved /jffs/ ? (jffs-201904031743_RT-AC68U-E4A0) { n | Y } (or press ENTER to ABORT) >

 Restore aborted

For NVRAM variables I use custom .ini files, that have had to be altered since v382/4.xx - present

e.g. OpenVPN

Spoiler: Backup OpenVPN NVRAM variables on HND RT-AC86U

./nvram-save.sh -i ovpn.ini -nojffs -nouser

nvram-save.sh: NVRAM User Save Utility - Version 26.2 EIC Hacked for v384.xx+!!
nvram-save.sh: Saving RT-AC86U settings from firmware 384.10_2
nvram-save.sh: Runtime options: -i ovpn.ini -nojffs -nouser
nvram-save.sh: Using custom NVRAM variable file: ovpn.ini
nvram-save.sh: Running in Backup Mode

 Saving [System - Basic]                  0 secs (0   variables, Bytes=0    Total=0    )
 Saving [VPN Client 1]                    0 secs (17  variables, Bytes=743  Total=743  )
 Saving [VPN Client 2]                    0 secs (17  variables, Bytes=736  Total=1479 )
 Saving [VPN Client 3]                    0 secs (17  variables, Bytes=169  Total=1648 )
 Saving [VPN Client 4]                    0 secs (17  variables, Bytes=32   Total=1680 )
 Saving [VPN Client 5]                    1 secs (17  variables, Bytes=32   Total=1712 )
 Saving [VPN Server Users]                0 secs (1   variables, Bytes=107  Total=1819 )
 Saving [VPN Server 1]                    0 secs (13  variables, Bytes=642  Total=2461 )
 Saving [VPN Server 2]                    0 secs (13  variables, Bytes=281  Total=2742 )

nvram-save.sh: Complete: User NVRAM saved to /tmp/mnt/RT-AC86U/BackupScriptsJFFS/nvram-restore-201904051207_RT-AC86U-6160.sh


jffs-save           201904051158_RT-AC86U-6160    Fri Apr  5 11:59:18 BST 2019
nvram-save.sh       201904051207_RT-AC86U-6160    Fri Apr  5 12:07:26 BST 2019    384.10_2                  ovpn.ini 0 minutes and 2 seconds elapsed, Total Bytes=2742

=======> ovpn.ini
# Hack to dynamically update the GUI
#
# see /jffs/scripts/openvpnclient1.postconf
# Usage ./nvram-save.sh -i ovpn.ini -nojffs -nouser

# The following line must ALWAYS be used
[System - Basic]

# Add User required NVRAM variables here.....

[VPN Client 1]
vpn_client1_desc
vpn_client1_addr
vpn_client1_proto
vpn_client1_port
vpn_client1_if
vpn_client1_userauth
vpn_client1_username
vpn_client1_password
vpn_client1_adns
vpn_client1_bridge
vpn_client1_cipher
vpn_client1_clientlist
vpn_client1_comp
vpn_client1_crypt
# Now in base64 format!! and was previously called vpn_clientX_custom2 but HND limited to 510 chars in 3 variables
vpn_client1_cust2
vpn_client1_cust21
vpn_client1_cust22

<snip>

[VPN Server Users]
vpn_serverx_clientlist

[VPN Server 1]
vpn_server1_port
vpn_server1_local
vpn_server1_sn
vpn_server1_cipher
vpn_server1_comp
vpn_server1_crypt
vpn_server1_c2c
vpn_server1_ccd
vpn_server1_userpass_auth
vpn_server1_rgw
# Now in base64 format!! and was previously called vpn_serverX_custom2 but HND limited to 510 chars in 3 variables
vpn_server1_cust2
vpn_client1_cust21
vpn_client1_cust22

@john9527's original design stands the test of time, and with care, can be used for 'Backup/Restore' of simple User fields once on the latest platforms.

Can't ever recall if I ever took advantage of the migrate feature, so I too had to manually reconfig by scratch … as solidly preached by @L&LD - wise words

 

[RMerlin]

384.10_2 / RT-AX88U

Apr  5 03:50:10 rc_service: amas_lib 1216:notify_rc restart_firewall

This is the AiMesh service. If you ever used AiMesh on this router, best to do a factory default reset to remove any leftover configuration.

That library is closed source, so I have no idea what triggers a firewall restart from it.

Apr  5 03:50:10 nat: apply nat rules (/tmp/nat_rules_eth0_eth0)
Apr  5 10:14:26 kernel: eth0 (Int switch port: 3) (Logical Port: 3) (phyId: c) Link DOWN.
Apr  5 10:14:30 kernel: eth0 (Int switch port: 3) (Logical Port: 3) (phyId: c) Link UP at 1000 mbps full duplex
Apr  5 10:15:01 kernel: eth0 (Int switch port: 3) (Logical Port: 3) (phyId: c) Link DOWN.
Apr  5 10:15:04 DualWAN: skip single wan wan_led_control - WANRED off
Apr  5 10:15:06 kernel: eth0 (Int switch port: 3) (Logical Port: 3) (phyId: c) Link UP at 1000 mbps full duplex
Apr  5 10:15:09 WAN_Connection: Ethernet link up.
Apr  5 10:15:09 rc_service: wanduck 1035:notify_rc restart_wan_if 0
Apr  5 10:15:11 kernel: eth0 (Int switch port: 3) (Logical Port: 3) (phyId: c) Link DOWN.
Apr  5 10:15:11 nat: apply nat rules (/tmp/nat_rules_eth0_eth0)
Apr  5 10:15:11 wan: finish adding multi routes
Apr  5 10:15:13 kernel: eth0 (Int switch port: 3) (Logical Port: 3) (phyId: c) Link UP at 1000 mbps full duplex
Apr  5 10:15:21 WAN_Connection: WAN was restored.

Why firewall is being restarted in the middle of the night or every few hours ? I have seen that at least few times now, but i have cleared the system logs already. It has never happened before. Also, some weird things is happening to the WAN connection ? Also never seen that in the system log before...

Your Internet went down. Could be ISP maintenance, or modem crashing/losing sync. Unrelated to the firmware, it simply reported that the Ethernet connection completely went down, then back up.

 

[Butterfly Bones]

I recently started dreading firmware upgrades because of the potential need to perform a factory reset and manual restore. I have three Asus routers to support in addition to my sand box Asus router. Since I've fallen behind several versions on all four, the factory reset and manual restore step will definitely be required now.

RIP User NVRAM Save/Restore Utility
You are greatly missed. Maybe one of the script writers can resurrect it from the dead?

View attachment 16882

I used it too. I know there were changes by ASUS to nvram that made maintaining the script very tedious, if not impossible. I think each field now has varying links.

The only real complexity to full reset and reconfigure for me is that I assign a static IP to all the devices on my network. However, thanks to some helpful users here, there is a simple workaround. It should only be used on the same router version, do not migrate. Look at both post #74 and well as this post #75 in the link.
https://www.snbforums.com/threads/backup-manual-dhcp-list.12876/page-4#post-469462

I have ovpn files from my VPN host, and keep most other settings in a text file linked off my desktop that syncs with my phone and tablet for handy reference. Makes updates much less time consuming. As stated many times, backup both router and jffs using the ASUSwrt-Merlin webGUI tools. Another step I make before upgrade is backup my jffs partition and my entire USB stick to my computer using and FTP client. I am a firm believer in repetitive redundancy. Easy backup and, if necessary, restore since I am an long time geek and primarily use Linux, all permissions are preserved. (I have an old Win7 laptop for those few items I need that only work with 'Doze)

 

[skeal]

I used it too. I know there were changes by ASUS to nvram that made maintaining the script very tedious, if not impossible. I think each field now has varying links.

The only real complexity to full reset and reconfigure for me is that I assign a static IP to all the devices on my network. However, thanks to some helpful users here, there is a simple workaround. It should only be used on the same router version, do not migrate. Look at both post #74 and well as this post #75 in the link.
https://www.snbforums.com/threads/backup-manual-dhcp-list.12876/page-4#post-469462

I have ovpn files from my VPN host, and keep most other settings in a text file linked off my desktop that syncs with my phone and tablet for handy reference. Makes updates much less time consuming. As stated many times, backup both router and jffs using the ASUSwrt-Merlin webGUI tools. Another step I make before upgrade is backup my jffs partition and my entire USB stick to my computer using and FTP client. I am a firm believer in repetitive redundancy. Easy backup and, if necessary, restore since I am an long time geek and primarily use Linux, all permissions are preserved. (I have an old Win7 laptop for those few items I need that only work with 'Doze)

You operate like I do.

 

[L&LD]

There is a good reason behind that. My ISP is renewing WAN IP address every 24 hours and I do not want this to be done the exact hour I turned on the router (most likely during the day or working hours). Restart at 5 AM ensures me that the IP will not change when I am using the internet.
I can always go back before 384.10.0, where it was all good, but it makes no sense since the firmware will not become better if I do not indicate the problem, which this FW certainly has (I am not the only one with it).

EDIT:
So I managed to reproduce the issue. It was 10:31 AM and I changed the Reboot Scheduler time to 10:40. The router just changed the color or the internet icon to amber, not restarted at all. GUI is inaccessible. Last log entries before my hard restart:
Apr 5 10:38:42 ovpn-server1[1198]: 95.40.252.7:48647 SIGUSR1[soft,connection-reset] received, client-instance restarting
Apr 5 10:40:11 kernel: SysRq : Emergency Sync
Apr 5 10:40:11 kernel: Emergency Sync complete

I don't think you understood my post completely? The issue doesn't seem to be 384.10 (or the newer one 384.10_2), it seems to be an issue on your ISP's side. Including the fact that you don't think that a full M&M Config may even fix or at least minimize these issues fully for you too.

Do I understand correctly that you turn your router off at night?

 

[HawkInOz]

Unrelated but curious.

Someone send me this YouTube video regarding online privacy in 2019. This guy goes way above and beyond anything we're doing here. Is there any validity to all this or is this simply overkill?

Start a new thread for stuff like this. It has nothing to do with the firmware version being discussed in this thread... Thanks.

 

[eastavin]

DNS question. Is there any known reason not to use DNS 1.1.1.1 or 1.0.0.1 in this f/w? I heard that it causes problems in some brands of routers. Many thanks.

 

[tallytr]

DNS question. Is there any known reason not to use DNS 1.1.1.1 or 1.0.0.1 in this f/w? I heard that it causes problems in some brands of routers. Many thanks.

?
I've always used Google 8.8.8.8 for DNS Server1 and 8.8.4.4 for DNS Server2
Works great.

 

[DAVID LONG]

DNS question. Is there any known reason not to use DNS 1.1.1.1 or 1.0.0.1 in this f/w? I heard that it causes problems in some brands of routers. Many thanks.

Those servers are all I have ever used on Asus-Merlin

 

[QuikSilver]

DNS question. Is there any known reason not to use DNS 1.1.1.1 or 1.0.0.1 in this f/w? I heard that it causes problems in some brands of routers. Many thanks.

I use those servers for most PCs on my network but stick with google DNS for video devices (ie firetv, appletv, etc...)

 

[MoonPie2000]

Not sure if this is due to the firmware update but after the update I changed my VPN server and when I started the VPN back up the local address changed from 10.x.x.x to 172.x.x.x. What would make this happen? Is it something I can / or got configured somewhere with the update or none of the above. Everything seems to be working just the local address changed.

 

[L&LD]

Not sure if this is due to the firmware update but after the update I changed my VPN server and when I started the VPN back up the local address changed from 10.x.x.x to 172.x.x.x. What would make this happen? Is it something I can / or got configured somewhere with the update or none of the above. Everything seems to be working just the local address changed.

I can't see how the firmware would affect this? If it is working otherwise, safe to ignore for now.

But keep an out out for more/other random bugs/glitches. It may indicate a full reset is in order (M&M Config, see my signature for link and more details below) to get your router back to a good/known state.