What's new

uiDivStats uiDivStats - WebUI for Diversion statistics

  • 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!

I do have a lot of IoT devices.

37 devices connected at the moment.
 
With uiDivStats installed DSK was hitting 101% on occasion.

I have uninstalled it again, and now DSK ranged from 0 > 52% but very infrequently.

Never had this problem with v1.x
 
I do have a lot of IoT devices.

37 devices connected at the moment.
Can you run the tail command I provided and let me know how many rows you see appear? Perhaps I should look into a way of buffering the entries somewhere in memory/jffs and periodically flush to USB.

The problem I have is that I can't tamper with the original dnsmasq.log file without impacting on Diversion's functions - otherwise I would simply vacuum out log lines on a regular basis
 
With uiDivStats installed DSK was hitting 101% on occasion.

I have uninstalled it again, and now DSK ranged from 0 > 52% but very infrequently.

Never had this problem with v1.x
1.0 would quite often lock up the CPU for long periods of time running grep queries on large log files.
2.0 runs a constant parser of dns logs into a db, as I said a couple of posts back.
 
I have run the tail command.

The output scrolls off the top of the screen. Not sure how to screen grab as it just scrolls up.
 
It comes in bursts.

I have a split screen so I can watch the CPU usage as the tail command runs.

There does not appear to be a connection. When the tail scrolls, the CPU does not always follow, if you see what i mean?
 
Tail 1.jpg Tail 2.jpg Tail 3.jpg
Tail 1.jpg
Tail 2.jpg
Tail 3.jpg
Tail 4.jpg
 
I had a similar experience with my AC86U, CPU pegged at 100% io. But then my USB stick connected to the router failed and I replaced it with a newer and faster one, that made the problem disappear for me.
Coincidence or not, but my Entware usb flash drive just died the night after I had installed uiDivStats 2.0.
 
It comes in bursts.

I have a split screen so I can watch the CPU usage as the tail command runs.

There does not appear to be a connection. When the tail scrolls, the CPU does not always follow, if you see what i mean?
uiDivStats logging to sqlite may run a little behind the tail, so i would expect it to be a little staggered. I suspect you're seeing the issues due to the number of clients on your network. Are you able to bring down the number of clients temporarily to confirm or deny this theory?
 
I have reduced my devices count temporarily down to 7. I have just rebooted the router and am just waiting for things to settle before i do anything else...
 
OK, after 15 minutes since reboot, things have settled.

CPU does seem calmer. Occasionally cores hit 100%, but doesn't appear to be as often. DSK (aTop). occasionally hits 101% too, but again it has calmed down considerably.

So, do I take it that it's connected (no pun intended) to the number of devices on my network?

Tail command is also very light now to, not too much at all.
 
Apart for the obvious. What can I do?
 
You seemed to have a lot of PTR (reverse lookup) queries from your devices. Would it break anyone’s heart it Jack only shows query[A] or [AAAA]? It might lighten the load.
 
I have a program called GlassWire installed. That keeps a eye on my network when my PC is switched on. Could that have anything to do with the extra traffic?
 
Last edited:
@JemTheWire, Ding! Ding! Ding! We may have a winner?! :)
 
I've just disabled GlassWire and rebooted my PC, no difference to anything. Having connected all my devices back, I am seeing the same spikes, and USB being thrashed.
 
Quite frankly, my head is spinning with it all. I'm going to remove uiDivStats for now.
 

Latest threads

Sign Up For SNBForums Daily Digest

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