Resources for ScreenConnect on CentOS 7

[JaredBusch]

Re: Installing ScreenConnect on CentOS 7

So I finally had the time to convert my ScreenConnect install from Windows based to CentOS. That process worked jsut fine and you can look at th elink above for how to install it all.

Now my system is running and it is laggy. Giving htop a look, I do believe I need to up the memory. How much do you think would be sufficient?

[root@bnasc ~]# free -m
              total        used        free      shared  buff/cache   available
Mem:           1929        1445         365           1         118         410
Swap:          1791         871         920
[root@bnasc ~]#

The system has just over 200 machines with about 130 online.

[dafyre]

Check the output of

sysctl vm.swappiness

I usually set mine to 10 on my linux systems...

vi /etc/sysctl.d/90-swappiness.conf
content:

vm.swappiness = 10

then, reload the configuration and no reboot is required:

sysctl --system -p

(note in the command above, there's 2 dashes before system and one before the letter p).

This makes the linux kernel not use the swap quite as aggressively.

[JaredBusch]

@dafyre said in Resources for ScreenConnect on CentOS 7:

Check the output of

sysctl vm.swappiness

I set it to 10 when I built the system.

Some thread or another around here long time ago mentioned it.

[root@bnasc ~]# sysctl vm.swappiness
vm.swappiness = 10
[root@bnasc ~]#

[JaredBusch]

wow....

              total        used        free      shared  buff/cache   available
Mem:           1744        1584          65           1          94          21
Swap:          1791         116        1675
[root@bnasc ~]# /etc/init.d/screenconnect stop
Sending kill signals...
Waiting for processes to quit...
[root@bnasc ~]# free -m
              total        used        free      shared  buff/cache   available
Mem:           1744          96        1552           1          94        1508
Swap:          1791          75        1716
[root@bnasc ~]#

[JaredBusch]

boot up is busy.

[gjacobse]

From the conversations I had with SC on the matter,... it seems that the Client Agent chatter (calling home, checking updating version) can cause the system to be 'laggy'... but once done, it'll quiet down.

@scottalanmiller would know on this.

[JaredBusch]

2 minutes and 40 seconds before it stabilized.

[JaredBusch]

@gjacobse said in Resources for ScreenConnect on CentOS 7:

From the conversations I had with SC on the matter,... it seems that the Client Agent chatter (calling home, checking updating version) can cause the system to be 'laggy'... but once done, it'll quiet down.

@scottalanmiller would know on this.

The initial post and comment were made with a system that had been online for 16 hours.

And the Windows system that I migrated from never had this lag.

[Dashrender]

@gjacobse said in Resources for ScreenConnect on CentOS 7:

From the conversations I had with SC on the matter,... it seems that the Client Agent chatter (calling home, checking updating version) can cause the system to be 'laggy'... but once done, it'll quiet down.

@scottalanmiller would know on this.

How long does it take for things to quiet down?

[dafyre]

@Dashrender said in Resources for ScreenConnect on CentOS 7:

@gjacobse said in Resources for ScreenConnect on CentOS 7:

From the conversations I had with SC on the matter,... it seems that the Client Agent chatter (calling home, checking updating version) can cause the system to be 'laggy'... but once done, it'll quiet down.

@scottalanmiller would know on this.

How long does it take for things to quiet down?

@JaredBusch said in Resources for ScreenConnect on CentOS 7:

2 minutes and 40 seconds before it stabilized.

About 3 minutes.

[JaredBusch]

[JaredBusch]

After 10 minutes, it has been sitting unchanged with the memory usage. CPU spikes to 100% a few times that I noticed.

I intentionally used 4GB here to make sure nothing should feel the need to go to swap.

[root@bnasc ~]# free -m
              total        used        free      shared  buff/cache   available
Mem:           3760        1293        1959           8         506        2214
Swap:          1791           0        1791
[root@bnasc ~]#

[scottalanmiller]

At 2GB, we never see it need it all. It's decently lean.

[JaredBusch]

@scottalanmiller said in Resources for ScreenConnect on CentOS 7:

At 2GB, we never see it need it all. It's decently lean.

Been watching things and it seems to need another CPU. Everytime I click on something on the webpage, the CPU spikes to 100%.

[scottalanmiller]

@JaredBusch said in Resources for ScreenConnect on CentOS 7:

@scottalanmiller said in Resources for ScreenConnect on CentOS 7:

At 2GB, we never see it need it all. It's decently lean.

Been watching things and it seems to need another CPU. Everytime I click on something on the webpage, the CPU spikes to 100%.

I think that we use two.

[gjacobse]

@scottalanmiller said in Resources for ScreenConnect on CentOS 7:

@JaredBusch said in Resources for ScreenConnect on CentOS 7:

@scottalanmiller said in Resources for ScreenConnect on CentOS 7:

At 2GB, we never see it need it all. It's decently lean.

Been watching things and it seems to need another CPU. Everytime I click on something on the webpage, the CPU spikes to 100%.

I think that we use two.

Since we are having issue with our system

$ sysctl vm.swappiness
vm.swappiness = 30

[dafyre]

@gjacobse said in Resources for ScreenConnect on CentOS 7:

@scottalanmiller said in Resources for ScreenConnect on CentOS 7:

@JaredBusch said in Resources for ScreenConnect on CentOS 7:

@scottalanmiller said in Resources for ScreenConnect on CentOS 7:

At 2GB, we never see it need it all. It's decently lean.

Been watching things and it seems to need another CPU. Everytime I click on something on the webpage, the CPU spikes to 100%.

I think that we use two.

Since we are having issue with our system

$ sysctl vm.swappiness
vm.swappiness = 30

Why not go lower to say... 10 ?

[gjacobse]

@dafyre said in Resources for ScreenConnect on CentOS 7:

@gjacobse said in Resources for ScreenConnect on CentOS 7:

@scottalanmiller said in Resources for ScreenConnect on CentOS 7:

@JaredBusch said in Resources for ScreenConnect on CentOS 7:

@scottalanmiller said in Resources for ScreenConnect on CentOS 7:

At 2GB, we never see it need it all. It's decently lean.

Been watching things and it seems to need another CPU. Everytime I click on something on the webpage, the CPU spikes to 100%.

I think that we use two.

Since we are having issue with our system

$ sysctl vm.swappiness
vm.swappiness = 30

Why not go lower to say... 10 ?

@scottalanmiller -

[scottalanmiller]

@dafyre said in Resources for ScreenConnect on CentOS 7:

@gjacobse said in Resources for ScreenConnect on CentOS 7:

@scottalanmiller said in Resources for ScreenConnect on CentOS 7:

@JaredBusch said in Resources for ScreenConnect on CentOS 7:

@scottalanmiller said in Resources for ScreenConnect on CentOS 7:

At 2GB, we never see it need it all. It's decently lean.

Been watching things and it seems to need another CPU. Everytime I click on something on the webpage, the CPU spikes to 100%.

I think that we use two.

Since we are having issue with our system

$ sysctl vm.swappiness
vm.swappiness = 30

Why not go lower to say... 10 ?

That's what I set them to

[gjacobse]

@scottalanmiller said in Resources for ScreenConnect on CentOS 7:

@dafyre said in Resources for ScreenConnect on CentOS 7:

@gjacobse said in Resources for ScreenConnect on CentOS 7:

@scottalanmiller said in Resources for ScreenConnect on CentOS 7:

@JaredBusch said in Resources for ScreenConnect on CentOS 7:

@scottalanmiller said in Resources for ScreenConnect on CentOS 7:

At 2GB, we never see it need it all. It's decently lean.

Been watching things and it seems to need another CPU. Everytime I click on something on the webpage, the CPU spikes to 100%.

I think that we use two.

Since we are having issue with our system

$ sysctl vm.swappiness
vm.swappiness = 30

Why not go lower to say... 10 ?

That's what I set them to

Just checked - It's 30...

$ uptime
 17:48:06 up 1 day,  3:02,  2 users,  load average: 0.00, 0.04, 0.07

$ sysctl vm.swappiness
vm.swappiness = 30
~~