What's new

ASUSWRT - scheduled reboots and power-downs

  • SNBForums Code of Conduct

    SNBForums is a community for everyone, no matter what their level of experience.

    Please be tolerant and patient of others, especially newcomers. We are all here to share and learn!

    The rules are simple: Be patient, be nice, be helpful or be gone!

ForkWNY

Senior Member
I like to power down/reboot my main GT-AC5300 router and Aimesh nodes at least once per week. Right now I have the router, cable modem, and Aimesh nodes plugged into Wifi outlets, so I schedule power-downs in the middle of the night once weekly, powering them all down for about 10 minutes, then they're powered back on in sequence automatically. This seems to keep things running optimally and automation is obviously convenient.

I've noticed that the "cog wheel" reboot button within the ASUS Rapture UI doesn't really do a hard reboot of the GT-AC5300 or any of my Aimesh nodes - when I SSH into any of the Aimesh nodes after "rebooting" from the Rapture UI, the uptime doesn't indicate that any of the nodes were actually rebooted (on ASUSWRT official). If the uptime is 3 days, and I run a "reboot" from the GT-AC5300 UI, the uptime still says 3 days after the reboot has completed. Odd IMO.

For those using the official ASUSWRT firmware releases, is there a better way to handle rebooting, aside from SSH and typing "reboot" at the command prompt within each Aimesh node, or powering things down? I'm presuming crontab can't be used for ASUSWRT official firmwares, it's wiped whenever the router is rebooted. I've used Merlin in the past but tend to stick with official builds for newer hardware.
 
The reboot time should be reset. Is it maybe a browser caching issue? Does using another browser show the same? Does a second reboot (back to back) help? :)

If you don't reboot, are you experiencing any issues? If you aren't, then you are putting extra stress on your network equipment for no good reason.

The problem with doing the reboots as indicated in your post above is that the various resources within the router (i.e. a USB drive, for example) aren't being 'safely' ejected first. You may be causing more harm than good by doing this 52 times a year, without reason, to your equipment. ;)
 
The reboot time should be reset. Is it maybe a browser caching issue? Does using another browser show the same? Does a second reboot (back to back) help? :)

If you don't reboot, are you experiencing any issues? If you aren't, then you are putting extra stress on your network equipment for no good reason.

The problem with doing the reboots as indicated in your post above is that the various resources within the router (i.e. a USB drive, for example) aren't being 'safely' ejected first. You may be causing more harm than good by doing this 52 times a year, without reason, to your equipment. ;)

The reboot time is reset within the router UI. However, if you SSH into any of the devices (router or Aimesh nodes), the uptime does not indicate a reboot occurred at the OS level. It appears that OS-level reboots only occur when firmware updates are applied. If there are memory leaks or other things occurring that only an OS-level reboot would clear out, those problems would potentially persist after running a "soft reboot" from the router's UI.

I don't use USB drives in conjunction with the router, so no issues with properly dismounting USB drives prior to reboot. As far as having good reason, I guess my thought is when you get multiple devices in the mix, if one of them decides to go out of whack, especially something more subtle, an OS-level reboot (or power cycle) clears volatile memory and often will resolve the more mundane issues that can arise. It's especially useful if I'm out of town for several days and one of the devices (or the router itself) decides to take a dump. I have security cams, etc. at home and it's good to know that in the event something goes down, a power cycle will have it back up and running at a sooner time, rather than waiting to come home and power cycle things afterward.
 
Last edited:
I too am a believer in both soft-reboots and hard-reboots (via pulling the power plug, when warranted). But I don't know why ssh doesn't show the router has rebooted though? How are you seeing this difference vs. what the GUI reports?

How are you rebooting the nodes, btw? I don't recall that as an available option via the GUI.

It doesn't seem that you're having any issues though on your network equipment. Still, I don't like the 'reboot, just in case' clause. :)

Do you have quality UPS devices backing up your core network devices (modem/ONT and router/switches)?

How long are you (and everyone else) gone from your home that even a weekly reboot is needed? :)

Again, I am agreeing with the benefits a 'hard' reboot makes to the performance of a network. What I'm less convinced about is the requirement to do so in a properly configured and setup router (including being 'protected' from a quality UPS).

What is a properly configured and setup router to me? The post below offers an overview of a procedure to get your router to a good/known state. (And part of that state includes being able to run indefinitely).

https://www.snbforums.com/threads/ax88-packet-loss.62891/#post-563326

HTH.
 
After checking things a bit further with rebooting using the Rapture UI - here's what I think is going on...

A "scheduled" reboot within the Rapture UI (my router is a GT-AC5300), seems to reboot both the router and all connected AiMesh nodes (works as expected). This is a good thing.

When manually clicking the reboot icon within the Rapture UI (shown in screenshot below), none of the Aimesh nodes reboot - they stay up, and the main router reconnects them after rebooting. If you SSH into any of the AiMesh nodes after a UI reboot, their uptime remains unchanged. If I'm rebooting the router from the UI, the connected AiMesh nodes should reboot as well, just my $0.02 on that.

rapture-reboot.jpg


Sometimes I'm gone on business trips for 2 weeks or more (when the US isn't on lock-down thanks to a pandemic).

I don't have my router connected to a UPS, though I probably should (I use one for my desktop PC). When I lose power, the cable connection goes down, so the router essentially becomes useless anyway...a generator does not restore WAN connectivity. Unfortunately, I need grid power for WAN to work from my cable provider. Internal devices are fine with generator power because everything is home-run down to a 24-port gig switch, as long as that is up LAN comms are good.

Don't have any issues with the router (it's brand new, only 2 weeks old, and I configured it from scratch), runs like a champ. I like to power cycle it along with my cable modem. The cable modem definitely benefits from a power cycle, and I've run into reconnect issues with the cable modem when it's power cycled and the router is left as-is. Otherwise the router is fine, and so are the AiMesh nodes. There are many different schools of thought on power cycling/rebooting network devices, ask 10 people they'll give 10 different answers.
 
Last edited:
It would be great if you could issue a 'reboot all routers/nodes' command, but there isn't one, it seems. :)

Rebooting just the router seems correct to me in this case. Even if it is not the behavior expected by yourself. ;)

You do have some good reasons for rebooting it weekly. But now, with lockdown, you could disable it and see if it is more stable than you are currently giving it credit for. ;)

But, get a UPS for your core network equipment 'pronto'. :)
 
It would be great if you could issue a 'reboot all routers/nodes' command, but there isn't one, it seems. :)
...

Seems the only way to do that is to set up a scheduled reboot within the Rapture UI - Administration > System. The scheduled reboot definitely reboots the router and AiMesh nodes, the same should be true when manually rebooting from the UI (shown in the screenshot), but that's something ASUS would need to change, and who knows if that'll ever happen.

I have disabled the scheduled "power cycles" on the wifi outlets powering the Aimesh nodes and router. I'll let 'em stay up and powered on for the next month to find out if any instability issues creep up. I suspect the cable modem will be the main concern...I'm using my own Motorola 8600 which is supposed to be one of the better modems (Broadcom chipset). I'll let them stay on for awhile.

On a tangent, are you aware if the AiCloud service automatically backs up the router settings? I'm not familiar with it and am curious as to what benefits it might provide. I don't need any more backup services for data, have plenty of TB available at home. ;)
 
AiCloud shouldn't have access to your router settings! :eek:

I don't think that is even a possibility. :)

AiCloud shouldn't even be enabled if you have another solution. :)
 

Sign Up For SNBForums Daily Digest

Get an update of what's new every day delivered to your mailbox. Sign up here!
Top