RT-AC88U with 386.9 - SSH and HTTPS not working

[popxunga]

Hi

My RT-AC88U is running 386.9 firmware since it was published.

Since then I lost internet connectivity a couple of times; the last one I had the change to connect a cable to the internet provider router and checked that it was working fine.
Now I do have internet access, I can ping the router ip address but I cannot access the router either via SSH or HTTPS.
Currently I do not have the router syslog redirected to anywhere.

Any HTTPS connection attempt ends with timeout.

An SSH connection attempt returns the following message:
kex_exchange_identification: Connection closed by remote host
Connection closed by 192.168.131.1 port 22

Found some articles suggests that the message above might be returned if the system running the SSHD daemon suffers some lack of resources.

Any ideas ?

Is there any other recommended firmware level for this router that doesn't cause such type of failures ?
I really need to have the most reliable solution.

Thanks in advance.

PS: I never seen this happen while running some Merlin firmware fork based on 380.70 ... and I ran it during years.

 

[ColinTaylor]

Reboot the router and when it comes back up log in to the GUI and look back in the System Log for error messages. You may be running out of memory or jffs space, in which case there will be errors indicating that.

 

[popxunga]

Ok ... currently I can't, but will do it as soon as possible ...

 

[popxunga]

Meanwhile router got inoperative with the LAN 1~8 LED off.

From a first look I just see lots of this:

Mar 7 12:09:34 kernel: dcd/5003: potentially unexpected fatal signal 11.
Mar 7 12:09:34 kernel: Pid: 5003, comm: dcd
Mar 7 12:09:34 kernel: CPU: 0 Tainted: P (2.6.36.4brcmarm #1)
Mar 7 12:09:34 kernel: PC is at 0x2ae26444
Mar 7 12:09:34 kernel: LR is at 0x1c310
Mar 7 12:09:34 kernel: pc : [<2ae26444>] lr : [<0001c310>] psr: 20000010
Mar 7 12:09:34 kernel: sp : 7ec779a0 ip : 000a26b4 fp : 7ec77b04
Mar 7 12:09:34 kernel: r10: 00086a00 r9 : 0000dafc r8 : 00000068
Mar 7 12:09:34 kernel: r7 : 00086a00 r6 : 00000000 r5 : 7ec77c54 r4 : 00000000
Mar 7 12:09:34 kernel: r3 : 00000000 r2 : 00000000 r1 : 00086a00 r0 : 00000000
Mar 7 12:09:34 kernel: Flags: nzCv IRQs on FIQs on Mode USER_32 ISA ARM Segment user
Mar 7 12:09:34 kernel: Control: 10c53c7d Table: 5e0e004a DAC: 00000015

That message was the last one written into the log a bit more than 6 hours ago (currently 18h:41m local time).
I'll have a better look into the log tomorrow.

 

[popxunga]

Just decided to:

• Turn OFF AI Protection, Adaptative QOS, Web history and Traffic Analyzer;
• Kept one OpenVPN Server instance and one VPN Client instance running;
• Reboot

Let's see how it goes.

 

[popxunga]

Dirty upgraded to 386.10 and the router has been running fine:

Meanwhile I noticed the following set of error messages:

Mar 12 18:08:40 rtl_fail: rtkswitch fail access, restart.
Mar 12 18:08:40 kernel: rtk_port_linkStatus_get() fail, err: 10
Mar 12 18:08:42 kernel: rtl8365mbrtl8365mb initialized(0)(retry:0)
Mar 12 18:08:42 kernel: rtk port_phyEnableAll ok
Mar 12 18:08:43 kernel: txDelay_user: 1
Mar 12 18:08:43 kernel: # txDelay - TX delay value, 1 for delay 2ns and 0 for no-delay
Mar 12 18:08:43 kernel: EXT_PORT:16 new txDelay: 1, rxDelay: 1
Mar 12 18:08:43 kernel: current EXT_PORT:16 txDelay: 1, rxDelay: 1
Mar 12 18:08:43 kernel: rxDelay_user: 4
Mar 12 18:08:43 kernel: # rxDelay - RX delay value, 0~7 for delay setupEXT_PORT:16 new txDelay: 1, rxDelay: 4
Mar 12 18:08:43 kernel: current EXT_PORT:16 txDelay: 1, rxDelay: 4

Are those messages related with the Realtek switch (ports 5 to 8) ?
Can someone decode (or let me know where I can do it myself) what the rtk_port_linkStatus_get() fail, err: 10 means for example ?

Thanks,