What's new

How often do you reboot rt-ac88u?

  • 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!

rkalinka

Occasional Visitor
Seems that I cannot make it beyond 40 days without having to restart my router. Everything works great - but then it completely locks up - and I have to unplug/re-plug in the main router for the system to work. Anyone else have this issue? I know there is a reboot scheduler, but it only allows for weekly/daily reboots. I'd like to have an automatic reboot once a month. Any takers? Current configuration is 3-node AiMesh setup - all 3 nodes are Asus rt-ac88u running latest STOCK firmware.
 
I know there is a reboot scheduler, but it only allows for weekly/daily reboots. I'd like to have an automatic reboot once a month.
Since you're running Merlin on the main router you can setup any reboot schedule you want using cron and a user script.

 
I guess I shoudl've been more clear. I am NOT running merlin (yet). Is anyone else making it beyond 40 days before having to reboot the main router in an AiMesh setup? @ColinTaylor - I think I may end up going your route and installing Merlin to accomodate my issue - is crontab persistent through reboots?
 
Possibly a stupid question but would a weekly reboot at some ungodly hour of the morning when everyone is asleep do the job? I'm curious on why the ask for monthly.

Appreciate it doesn't get the fancy uptime award but if it does the job on stability for users with the result of 100% availability when it's needed...
 
@eightiescalling - not a stupid question whatsoever. I am trying to avoid downtime where possible as we use the WIFI for Occasional Babycam monitoring. IE - there is an app on my iPad that allows me to monitor my kids. THat being said, when the internt goes offline - bad things happen with the app and releveant monitoring which makes for overnight monitoring problems. It's not a HUGE problem if it's like once a month - but if it's once a week - it becomes more cumbersome. On top of that, sometimes other home devices/monitoring that also gather information gets sometimes a little broken. Overall, the system is stable - but like clockwork - at about 41 days - it just freezes. I have it mounted on a wall up high, to power cycling is a pain (but again, doable). So - weighing the value of a weekly reboot on STOCK firwmare, vs Monthly on Merlin at this point.
 
Having a reboot schedule is not fixing anything.

You should continue to look for the root cause of the issue.

See my responses in the link below.

 
@eightiescalling - not a stupid question whatsoever. I am trying to avoid downtime where possible as we use the WIFI for Occasional Babycam monitoring. IE - there is an app on my iPad that allows me to monitor my kids. THat being said, when the internt goes offline - bad things happen with the app and releveant monitoring which makes for overnight monitoring problems. It's not a HUGE problem if it's like once a month - but if it's once a week - it becomes more cumbersome. On top of that, sometimes other home devices/monitoring that also gather information gets sometimes a little broken. Overall, the system is stable - but like clockwork - at about 41 days - it just freezes. I have it mounted on a wall up high, to power cycling is a pain (but again, doable). So - weighing the value of a weekly reboot on STOCK firwmare, vs Monthly on Merlin at this point.
Depends where the bar for availability is set I guess.

I figure a 10 minute outage including WiFi recovery for IoT devices at 4am on a Sunday doesn't bother anything critical and it's a regular predictable pattern on any data gathering. (The fact anything wired is on a separate switch to the router helps.)

Sounds like you have an app that would react badly though (bad app design if it reacts that way I'd say but that's just me) so I can see that especially if it results in you all being woken up!

@L&LD - I don't disagree with the sentiment but in the case of an issue that infrequently I'd be for the quick and simple workaround. (Though I admit it feels wrong to treat something on a Linux kernel the way an old Windows 95 install needed to be treated!)
 
Your choice, @eightiescalling.

But this will grow worse if left alone. And even now, what the router is doing doesn't make much sense (i.e. it is not in a good/known state as it is today).
 
It's a bit chicken and egg - you don't know if it will get worse without knowing the root cause.

Completely agree it's possible and doesn't make sense but when it comes to software in particular it could be anything. The longer you leave something running the higher the chance you trip over issues that only surface through long running.

Would I be happy with weekly reboots with something running in my companies DC? No chance! For a home router though, and assuming reboots don't significantly affect the long term running of the device (happy to be corrected on that assumption), a regular reboot is a way to avoid issues that only surface in the long term - a badly handled overflow or a memory leak that chews up a small % of memory a week for example. It's also quick and easy where the user is less technically able (or inclined).

Sometimes the perception of 100% available can be as good as actually being 100% available.
 
Having a reboot schedule is not fixing anything.

You should continue to look for the root cause of the issue.

See my responses in the link below.

@L&LD - I looked at this thread (and many others) with respect to ports 5-8. Already ran into this problem on 1 of my 3 nodes pretty regularly - it was fixed via replacement. There is a real possibility that it's the onboard switch OR the power supply (from other threads). It just struck me odd that this problem would happen about 40 days into operation when all other performance and related items are stable otherwise. There is nothing unusual showing up in logs either which is was struck me a bit odd. For example, when ports 5-8 died on the old (since replaced) node, it would log an error in syslog. In this case, I am not seeing anything beyond normal operation and performance but 3 times now - on "day 41" the master node simply becomes unresponsive. I still have wifi being broadcast (could be from the other nodes though)- just no routing of any kind and at least one LAN port does not respond. The simple fix is a powercycle - and no more problems for the next 40 days.

I supposed I could try swapping out the main node with one of the other AiMesh nodes and see if the problem follows that device or not. Or try a different power supply - but this device is in a difficult spot to reach and reliable otherwise - so was thinking a simple monthly reboot would give me the best of both worlds. OR wonderig if going to Merlin firmware would yield any benefits.

I guess I was wondering if others out there have to restart this Asus routers every 40 days - or if you never reboot them.
 
Rebooting a router, for peace of mind, daily, weekly, monthly, or 40 days indicates an issue to be fixed. Either in the hardware, the features/options and/or scripts used, or the firmware itself.

Rebooting 'never' is what a router should be able to do. If your experience is otherwise, there is an issue somewhere that should be sought out.

The Asus 8 Port routers and their random, goose-chasing issues never seemed to be fixed with a new power supply. Not using Ports 5-8 is the best solution/workaround if they must be kept in use as the primary router in a network.
 

Similar threads

Support SNBForums w/ Amazon

If you'd like to support SNBForums, just use this link and buy anything on Amazon. Thanks!

Sign Up For SNBForums Daily Digest

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