Hello everyone,
I'm a fairly long time user, but it's the first time I've posted on here. I really appreciate the firmware stability and I hope to be able to support it more going forward.
I could use some advice.
I'm having a unique problem where I'm losing WAN connectivity. Here are my specs:
- RT-AC66R
- Firmware: 374.43_39E39527
This problem first appeared on 39E1 (but that could be a coincidence). Originally, it appeared I was losing only wireless connectivity to the WAN because when I went to the wired desktop it was connected. However, I now think that the WAN access just returned by the time I got to the desktop.
I performed the factory reset / nvram erase guide followed by flashing the latest 374.43 version multiple times and it doesn't seem to make a difference. My last attempt, I went through the reset and went with most of the defaults (rather than restoring my configuration). The only thing I changed was the basics (ssid, password, etc.). I'm still experiencing the same issue.
I've upped the logging and I'm not seeing anything significant. I would say that I *generally* see these messages follow a loss of connectivity:
May 4 12:06:58 (none) user.warn kernel: wlc_dump_aggfifo:
May 4 12:06:58 (none) user.warn kernel: framerdy 0x10 bmccmd 4 framecnt 4
May 4 12:06:58 (none) user.warn kernel: AQM agg params 0xfc0 maxlen hi/lo 0x0 0xffff minlen 0x0 adjlen 0x8
May 4 12:06:58 (none) user.warn kernel: AQM agg results 0x8000 len hi/lo: 0x0 0x0 BAbitmap(0-3) 0 0 0 0
I will then see some DROPs (anywhere from 5-20), but they are always different ip addresses, ports, etc. and I didn't see anything that correlates so I'm not sure it's significant. I should also point out the events listed above + the drops sometimes appear without an apparent loss of connectivity, so I'm not sure it correlates 100%.
Other random tidbits:
- WAN uptime doesn't always reset, so I'm guessing that depends on the outage time.
- I have tried different DNS servers/providers including my ISP's and it doesn't seem to make a difference. FWIW - I've mostly been using Adguards lately.
- I don't seem to lose wireless AP access. I can get to the router and see the logs, etc.
- Outages can occur once or twice a day or it may not happen for 2 or 3 days.
- Utilization appears to be extremely low <.15, but perhaps I'm just not seeing the spike. Memory usage also seems fine 165MB free out of 233.92 MB
- I see temperature about 52 degrees celsius (according to the router), but the ambient temperature is 68-72 degrees Farenheight. I've made sure not to adjust the wireless power output anymore to avoid any potential overheating scenarios.
- The last message I just saw that looks concerning was this: May 4 12:30:44 (none) user.info rc_service: httpd 7923:notify_rc restart_wlcscan. However, this wasn't associated with a loss of connectivity nor do I recall seeing it before.
In one of my past jobs, I had to support some SonicWALL devices and every once in awhile I'd run across a unique problem where I had to call support. The few times I had to call they would always suggest a reset and reload, but most of the time that would not resolve the problem and the issue would subsequently be fixed in a future firmware release. The only reason I bring this up is I'd like to avoid the endless reset / re-flash solution if possible. I've already gone through the procedure 3 times and the family is getting a little restless. ;-)
At this point, I'm considering a "new" ASUS router, a new mesh solution (although not really loving a lot of the customization options for current solutions), swapping out my cable modem, installing an ASA at home, getting my syslog server back up and running, possibly the reset with the utility / instead of following the procedure manually. Are there any other suggestions I'm not seeing since I can't see the forest through the trees? ;-)
This is one of those situations where just bouncing ideas off someone else can be helpful; Thanks in advance for your time.