[ 386.12 alpha Build(s) ] Testing available build(s)

[SomeWhereOverTheRainBow]

I suppose I spoke a bit too soon. I'm getting the following error on my AC86U that's in AP mode. Was not getting this with 386.11. The error repeats at intervals anywhere between a few minutes to ~6 hours. The "asd" keyword in the output seems to suggest that this is the now infamous ASUS asd service puking on itself. There is no obvious performance issue with the AP. Interestingly, its not happening on the main router.

Aug 14 09:03:42 kernel: pgd = ffffffc01432b000
Aug 14 09:03:42 kernel: [00000168] *pgd=0000000014c3e003, *pud=0000000014c3e003, *pmd=0000000016c57003, *pte=0000000000000000
Aug 14 09:03:42 kernel: CPU: 1 PID: 15595 Comm: asd Tainted: P           O    4.1.27 #2
Aug 14 09:03:42 kernel: Hardware name: Broadcom-v8A (DT)
Aug 14 09:03:42 kernel: task: ffffffc014c24100 ti: ffffffc01434c000 task.ti: ffffffc01434c000
Aug 14 09:03:42 kernel: PC is at 0xf6b6016e
Aug 14 09:03:42 kernel: LR is at 0xf6b2b7fc
Aug 14 09:03:42 kernel: pc : [<00000000f6b6016e>] lr : [<00000000f6b2b7fc>] pstate: 200f0030
Aug 14 09:03:42 kernel: sp : 00000000ffa3bd38
Aug 14 09:03:42 kernel: x12: 00000000ffffffff
Aug 14 09:03:42 kernel: x11: 00000000ffa3c234 x10: 000000000000001c
Aug 14 09:03:42 kernel: x9 : 00000000ffffffff x8 : 000000000000016d
Aug 14 09:03:42 kernel: x7 : 0000000000000000 x6 : 00000000ffa3c240
Aug 14 09:03:42 kernel: x5 : 0000000000000073 x4 : 0000000000000005
Aug 14 09:03:42 kernel: x3 : 0000000000000000 x2 : 0000000000000001
Aug 14 09:03:42 kernel: x1 : 0000000000000168 x0 : 000000000000016d
Aug 14 09:03:42 kernel: CPU: 1 PID: 15595 Comm: asd Tainted: P           O    4.1.27 #2
Aug 14 09:03:42 kernel: Hardware name: Broadcom-v8A (DT)
Aug 14 09:03:42 kernel: task: ffffffc014c24100 ti: ffffffc01434c000 task.ti: ffffffc01434c000
Aug 14 09:03:42 kernel: PC is at 0xf6b6016e
Aug 14 09:03:42 kernel: LR is at 0xf6b2b7fc
Aug 14 09:03:42 kernel: pc : [<00000000f6b6016e>] lr : [<00000000f6b2b7fc>] pstate: 200f0030
Aug 14 09:03:42 kernel: sp : 00000000ffa3bd38
Aug 14 09:03:42 kernel: x12: 00000000ffffffff
Aug 14 09:03:42 kernel: x11: 00000000ffa3c234 x10: 000000000000001c
Aug 14 09:03:42 kernel: x9 : 00000000ffffffff x8 : 000000000000016d
Aug 14 09:03:42 kernel: x7 : 0000000000000000 x6 : 00000000ffa3c240
Aug 14 09:03:42 kernel: x5 : 0000000000000073 x4 : 0000000000000005
Aug 14 09:03:42 kernel: x3 : 0000000000000000 x2 : 0000000000000001
Aug 14 09:03:42 kernel: x1 : 0000000000000168 x0 : 000000000000016d

 

[maxbraketorque]

Kinda puzzling what its only happening on my AC86U AP and not on my AC86U main router. No performance hit, so meh?

For kicks, I decided to delete the asd directory on the AP per this thread:

Quickest/bet fix for asus asd bug fix

Hi, I have an asus firewall that is still needing daily reboots and is buggy with the asd issue from last week. I don't want to backup the config, factory reset, and reload the config (cons eg: requires travel, might loose ssl vpn cert's). What is the latest on how to fix this easily and...

www.snbforums.com

The AP has been rebooted. I'll report back on whether deleting the ASD directory resolved the asd service crash.

 

[maxbraketorque]

Kinda puzzling what its only happening on my AC86U AP and not on my AC86U main router. No performance hit, so meh?

For kicks, I decided to delete the asd directory on the AP per this thread:

Quickest/bet fix for asus asd bug fix

Hi, I have an asus firewall that is still needing daily reboots and is buggy with the asd issue from last week. I don't want to backup the config, factory reset, and reload the config (cons eg: requires travel, might loose ssl vpn cert's). What is the latest on how to fix this easily and...

www.snbforums.com

The AP has been rebooted. I'll report back on whether deleting the ASD directory resolved the asd service crash.

Deleting the asd directory did not resolve the issue.

 

[MysticGold04]

Can confirm reset to default and fresh flash that I cannot reach the webui, the router does respond to pings though. (AC68U and AC3100)

However, upgrades without a reset are fine. Have two media bridges and an AP working fine.

Thank you for your hard work Merlin!

 

[Howard_inGA]

Did any of you do a hard reset ?

Not me...

 

[Sicario]

I know it's common, but I just want to report that network map seems broken again on this version, was ok on 386.11.

 

[Nasua]

New build available 386.12_alpha1 (Custom icons still not working)
Onedrive Test builds

 

[Jeffrey Young]

New build available 386.12_alpha1
Onedrive Test builds

Seems the only change was

httpd: fix SSL certificate generation

 

[RMerlin]

Seems the only change was

httpd: fix SSL certificate generation

That solves httpd failing to start following a factory default reset due to missing certificate.

 

[Jeffrey Young]

That solves httpd failing to start following a factory default reset due to missing certificate.

Sweet. Kinda what I thought. Excellent work. Personally speaking, you do great work and I admire the personal effort you put into this project.

 

[Howard_inGA]

Updated the alpha today to the latest. This one just rocks, Eric. CPU efficiency is up, log entries are stable and minimal, temps are down on aggregate, and it is rock solid stable!!
Wow! Just Wow!!

 

[Squall Leonhart]

custom icons still busted.

 

[heysoundude]

Updated the alpha today to the latest. This one just rocks, Eric. CPU efficiency is up, log entries are stable and minimal, temps are down on aggregate, and it is rock solid stable!!
Wow! Just Wow!!

updates like this kinda-sorta foreshadow the end of support by Asus, and eventually Merlin, when they decide they just can't do any better on the old hardware. If things are as you say, that day may be sooner than we might think.

 

[SomeWhereOverTheRainBow]

updates like this kinda-sorta foreshadow the end of support by Asus, and eventually Merlin, when they decide they just can't do any better on the old hardware. If things are as you say, that day may be sooner than we might think.

I suspected all of this was going to happen back when I saw all the newer AX models rolling out fast and furiously. It is a clear indication that Asus is stepping up its game to newer hardware, kernels, and bloatware. Unfortunately, Asus probably had to start an exit strategy for the old AC generation. At least they have been courteous about it by doing it in phases. They have been gradually phasing out models over the last couple of years, instead of doing it all at once. This is actually a good strategy because it gives consumers time to decide their next move without it being forced onto them. However, you will always have those few who like to linger on old stuff well past its deprecation.

 

[MysticGold04]

Something I’ve seen with 386.11 at least on my RT-AC3100 is that at about 50 days of uptime, memory is almost maxed. It will be interesting to see if the same thing happens on 386.12. So far all good, upgraded a few days ago. I need to flash the alpha1 released today, but everything is smooth running so far!

Thanks Merlin and to all involved with development!

 

[Jeffrey Young]

memory is almost maxed

You do know the reported memory used by the router is misleading. It includes what is being used by the system cache. Memory used for cache will be taken back by the kernel as needed. You can use the free command for a better picture of what memory is used for what.

 

[maxbraketorque]

I suspected all of this was going to happen back when I saw all the newer AX models rolling out fast and furiously. It is a clear indication that Asus is stepping up its game to newer hardware, kernels, and bloatware. Unfortunately, Asus probably had to start an exit strategy for the old AC generation. At least they have been courteous about it by doing it in phases. They have been gradually phasing out models over the last couple of years, instead of doing it all at once. This is actually a good strategy because it gives consumers time to decide their next move without it being forced onto them. However, you will always have those few who like to linger on old stuff well past its deprecation.

I haven't seen any particular value yet to the 3006 FW. The only advantage I've found with the 3006 FW is the ability to set up an isolated guest network on an AP without having to use a script. However, ASUS took a significant step backward in OVPN functionality with the 3006 FW. I can't think of anything else that's different between 3004.386 and 3006.

But AX and 2.5 gbps networking is a nice feature of the newer routers.

 

[MysticGold04]

You do know the reported memory used by the router is misleading. It includes what is being used by the system cache. Memory used for cache will be taken back by the kernel as needed. You can use the free command for a better picture of what memory is used for what.

Yes I understand that, but this router is only in AP mode, and eventually it will stop responding…usually around the 60 day mark. I’ll most likely do a factory reset once 386.12 is released…

 

[JTnola]

I haven't seen any particular value yet to the 3006 FW. The only advantage I've found with the 3006 FW is the ability to set up an isolated guest network on an AP without having to use a script. However, ASUS took a significant step backward in OVPN functionality with the 3006 FW. I can't think of anything else that's different between 3004.386 and 3006.

But AX and 2.5 gbps networking is a nice feature of the newer routers.

Step backward how/in what ways?

 

[Clark Griswald]

Yes I understand that, but this router is only in AP mode, and eventually it will stop responding…usually around the 60 day mark. I’ll most likely do a factory reset once 386.12 is released…

AC86 AP running alpha without any issues. RAM usage currently 53% and verified custom icons working.
Both RMerlin FWs working well and stable in my environment.