[Release] Load Monitor 0.1 - Your server loads in a fancy chart :)

[Richard G]

Hello.

I don't use rspamd, so it might be SQLight was busy by DA itself, I don't know. However, it's odd that it's a hour long because I these mails from cron for an hour long. First time I've seen this too, while the plugin is running already on the servers for some time.

On the other hand.... I only got them yesterday night first time, and last night I didn't get anything, so if this is run every night, then it seems is indeed has been a conflicting situation as you think.
It's just a bit odd that this conflict is happening on 2 servers on exactly the same time. But maybe some DA cron is running at that time too, which would explain it.

I will report back if it happens again, or if it will happen just once a while, then at least I know I don't need to worry. Thanks!

 

[zEitEr]

Yes, that's odd, that different servers managed by different administrators with different set of software had the same issue. So I still think it has something related with DirectAdmin processes. Servers on my end run the plugin fine now, no other emails arrived after that night as well.

 

[Richard G]

that different servers managed by different administrators with different set of software had the same issue.

Yes indeed. And in my case it was also different software, Centos 7 and Almalinux 8.

I'm just wondering if maybe we can change the plugin cron time to a bit later time or something if it would happen again in the future.

 

[Active8]

If you don t want to play with installing it on your server I can recommend https://hetrixtools.com/ which also do uptime/load/blacklist check of your servers,
Provides nice list of your servers too , who you can see realtime its free to use till 15 servers.

 

[Richard G]

Thank you, but I'm already using both Hetrixtools and freshping and betteruptime which all have free monitoring for some servers.
Freshping even has an app... great.

But they all don't do real time load monitoring like this tool (and Munin for example) does from within DA. My collegue likes that a lot.
And you need to install the 3rd party tool with cron (for example Hetrix) to get the load monitoring visible there. I rather use first party.

 

[Ohm J]

@Richard G
For now I thing you need to silent error log between first line and last line of "add.php" by just add code

try{
//all code inside add.php
}catch(\Exception $ex){
        exit 0;
}

###UPDATE info
from your error log, it state as some column not allowed "NULL", so some information might not handle correctly to work with php8.

 

[Active8]

But they all don't do real time load monitoring like this tool

Yes it does, you can even monitor several servers in one view but not in DA but why you want it see per server if you can see them all at once ?

And you need to install the 3rd party tool with cron (for example Hetrix

Nope, hetrix has one single line of code you have to run at the server via SSH and that is it ! you can even monitor services like exim.
See examples herunder each line is memory/cpu/network, and when you click on it you see more details

 

[Richard G]

For now I thing you need to silent error log between first line and last line of "add.php" by just add code

Thank you, but I don't want to silent things, I want to see it if things go wrong.
Also, it's not my script it's from @zEitEr so I won't make changes to it.
I was also thinking it might not handle correctly to work with PHP 8, but if that was the case, I would have seen these errors multiple times and I've only seen them once this year, while I'm running on php 8.0 already from beginning of the year.

Nope, hetrix has one single line of code you have to run at the server via SSH

Oh I thought on their website I seen a cron which was needed for that, but I was just fastly looking.
I will contact you by pm because I don't want to go off-topic here.

 

[Richard G]

@zEitEr just to inform you, it just happened again. Curious as it also happened on your server like previous time.
Got 4 notices from cron via mail and that was it, so probably fixed itself again.

 

[zEitEr]

I don't recall any recent reports. I guess the error occurs when the plugin tries to INSERT two or more records with the same time/date. There is an unique primary key "created", which is date and time in the following format "2024-01-24 20:03:02". So if for any reason the plugin tries to INSERT a record with the same time/date the error will occur. The question is why it tries to INSERT two or more records with the same primary key. I don't have an answer here. Probably a server was overloaded in certain way. I'm not too sure about the reasons. The plugin does not write logs. So there are almost no chances to find a real cause. But I'd rather ignore the error, unless emails troubles you.