uiDivStats uiDivStats - WebUI for Diversion statistics

[Jack Yaz]

For those with CPU spikes, are you going by the Network Map chart? I've found with Network Map as the active page, it uses a chunk of CPU (confirmed in top/htop). Navigating to a static page, CPU sits around 2-3%.

 

[root]

I'm getting stuck at "uiDivStats: Creating database table and enabling write-ahead logging...".

I've tried to uninstall and reinstall, but still stuck at the same step.

uiDivStats: Welcome to uiDivStats v2.0.0, a script by JackYaz

uiDivStats: Checking your router meets the requirements for uiDivStats

Downloading http://bin.entware.net/aarch64-k3.10/Packages.gz
Updated list of available packages in /opt/var/opkg-lists/entware
Package grep (3.4-1) installed in root is up to date.
Package sqlite3-cli (3310100-1) installed in root is up to date.
Package procps-ng-pkill (3.3.15-4) installed in root is up to date.
uiDivStats: New version of uidivstats_www.asp downloaded

uiDivStats: Mounting uiDivStats WebUI page as user1.asp

uiDivStats: New version of taildns.tar.gz downloaded

Downloading http://bin.entware.net/aarch64-k3.10/Packages.gz
Updated list of available packages in /opt/var/opkg-lists/entware
Package grep (3.4-1) installed in root is up to date.
Package sqlite3-cli (3310100-1) installed in root is up to date.
Package procps-ng-pkill (3.3.15-4) installed in root is up to date.
uiDivStats: Creating database table and enabling write-ahead logging...

 

[JemTheWire]

For those with CPU spikes, are you going by the Network Map chart? I've found with Network Map as the active page, it uses a chunk of CPU (confirmed in top/htop). Navigating to a static page, CPU sits around 2-3%.

Yes I am. As I might have said previously, when I check using TOP core usage is reported normal. If I use the GUI memory usage checker, cores spike.

I do not understand this disparity.

I will say again all cores drop back to normal after I uninstall uiDivStats v2.

Not only do the cores spike, the whole router GUI even when using the CLI is very sluggish when uiDivStats is installed.

As much as I like the script I am going to have to remove it for now.

 

[HorseCalledHorse]

Yes I am. As I might have said previously, when I check using TOP core usage is reported normal. If I use the GUI memory usage checker, cores spike.

I do not understand this disparity.

I will say again all cores drop back to normal after I uninstall uiDivStats v2.

Not only do the cores spike, the whole router GUI even when using the CLI is very sluggish when uiDivStats is installed.

As much as I like the script I am going to have to remove it for now.

Same experience here. And I didn't see this behaviour with any previous versions.

 

[Jack Yaz]

Yes I am. As I might have said previously, when I check using TOP core usage is reported normal. If I use the GUI memory usage checker, cores spike.

I do not understand this disparity.

I will say again all cores drop back to normal after I uninstall uiDivStats v2.

Not only do the cores spike, the whole router GUI even when using the CLI is very sluggish when uiDivStats is installed.

As much as I like the script I am going to have to remove it for now.

Have you tried restarting httpd?

service restart_httpd

Other than the WebUI seeming slow (which is par for the course), are you experiencing any other issues with the perceived higher CPU usage? These forums do seem to have people worrying when their RAM or CPU is actually being used

 

[martinr]

It’s been quite a while since I looked at the uiDivStats page in the webUI, so I got a shock at just how good the new version looks and just how much information is on display. No problems with it at all (though my router (AC68U) got stuck updating, so I merely rebooted, uninstalled and reinstalled).

Many thanks, Jack; it’s superb.

 

[pr0jects]

Same experience here. And I didn't see this behaviour with any previous versions.

Same behaviour here - AC88U

 

[JemTheWire]

Have you tried restarting httpd?

service restart_httpd

Other than the WebUI seeming slow (which is par for the course), are you experiencing any other issues with the perceived higher CPU usage? These forums do seem to have people worrying when their RAM or CPU is actually being used

I have restarted httpd several times and I have also done a few routers power cycles. Not just a reboot, I mean full power off, wait then power on. Results is always the same when uiDivStats v2 is installed.

Yes I an noticing a complete slowdown on the cli.

For example, when I SSH into the router and type AMTM, it literally takes a couple of seconds for ANYTHING to happen with uIDivStats installed. With uiDIvStats removed, typing AMTM results in an immediate response.

I appreciate that when someone says ‘slowdown’ it is all relative and slowdown for one person is perhaps not as noticeable to someone else, but I am experiencing a real slowdown.

I don’t know how else to describe it.

The only true test I guess is to factory reset the router and load everything back, manually (not restore from a backup) but at this time I don’t want to do that for one script.

It’s not as if I have an old underpowered router, it’s an AX88U.

 

[Jack Yaz]

I have restarted httpd several times and I have also done a few routers power cycles. Not just a reboot, I mean full power off, wait then power on. Results is always the same when uiDivStats v2 is installed.

Yes I an noticing a complete slowdown on the cli.

For example, when I SSH into the router and type AMTM, it literally takes a couple of seconds for ANYTHING to happen with uIDivStats installed. With uiDIvStats removed, typing AMTM results in an immediate response.

I appreciate that when someone says ‘slowdown’ it is all relative and slowdown for one person is perhaps not as noticeable to someone else, but I am experiencing a real slowdown.

I don’t know how else to describe it.

The only true test I guess is to factory reset the router and load everything back, manually (not restore from a backup) but at this time I don’t want to do that for one script.

It’s not as if I have an old underpowered router, it’s an AX88U.

Try

top -bn 1 | head -25

and a screenshot of htop after sorting by CPU (press f6 to check its sorted by percent_cpu) - it may not be uiDivStats directly, but something else not playing nice. FWIW I've been running v2 on my 86U upwards of 2 months without such issues!

 

[JemTheWire]

Here is a screen shot of the TOP command:-

Nothing too untoward.

 

[dave14305]

Here is a screen shot of the TOP command:-

View attachment 23979
Nothing too untoward.

What about running:

ps w | grep " D " | grep -v grep

I/O seems to be hitting one CPU hard. See if any process is waiting on Disk access.

 

[CriticJay]

Yes I am. As I might have said previously, when I check using TOP core usage is reported normal. If I use the GUI memory usage checker, cores spike.

I do not understand this disparity.

I will say again all cores drop back to normal after I uninstall uiDivStats v2.

Not only do the cores spike, the whole router GUI even when using the CLI is very sluggish when uiDivStats is installed.

As much as I like the script I am going to have to remove it for now.

I had the same experience yesterday. Although I'm unable to troubleshoot right now (at work for the whole day) - I also remember the system uptime "load average" counters were at 4+ which I've never seen before.

i mean the 3 numbers you get when you run the uptime command. usually I see like: 1.71, 1.85, 1.69
sometimes when it's busy i'd see: 2.56, 2.32, 2.29

during my experience yesterday with uiDivStats, I saw the router maintain itself at: 4.15, 4.09, 4.05

that was the last thing i saw before i uninstalled the script

 

[CriticJay]

Yes I an noticing a complete slowdown on the cli.

For example, when I SSH into the router and type AMTM, it literally takes a couple of seconds for ANYTHING to happen with uIDivStats installed. With uiDIvStats removed, typing AMTM results in an immediate response.

just want to echo that my experience on my 86U is identical to Jem above

hopefully whatever fixes the issue for Jem will also fix it for me

 

[charlie2alpha]

What about running:

ps w | grep " D " | grep -v grep

I/O seems to be hitting one CPU hard. See if any process is waiting on Disk access.

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.

 

[JemTheWire]

What about running:

ps w | grep " D " | grep -v grep

I/O seems to be hitting one CPU hard. See if any process is waiting on Disk access.

What is the command supposed to do? When i run it it just drops back to the :> prompt


Thanks

 

[dave14305]

What is the command supposed to do? When i run it it just drops back to the :> prompt

It might have shown any processes blocked on disk access. Just a stab in the dark since top showed io at 25% (one core).

 

[Jack Yaz]

Here is a screen shot of the TOP command:-

View attachment 23979
Nothing too untoward.

Do you have unbound's logging on? I wonder if that + uiDivStats logging is causing an IO overload to your USB.

 

[Jack Yaz]

The entware utility

opkg install atop

DSK row will show if your USB is busy.

How many devices on the network? I've just tried hammering an infinite loop worth of DNS requests from 1 of my machines and I get the below:

Do you have a particularly busy network?
How fast updates do you get if you:

tail -F /opt/var/log/dnsmasq.log /opt/var/log/dnsmasq.log3

 

[JemTheWire]

Something is hammering my USB regularly.
No, i don't have Unbound logging enabled.

But why only with uiDivStats installed?

 

[Jack Yaz]

Something is hammering my USB regularly.
No, i don't have Unbound logging enabled.

But why only with uiDivStats installed?

uiDivStats 2.0 does tail the dnsmasq log files to a sqlite db on the USB, but it shouldn't be excessive. Do you have a lot of clients?