logo
Welcome Guest! To enable all features please Login or Register.

Notification

Icon
Error

Options
Go to last post Go to first unread
tylerwight  
#1 Posted : Wednesday, May 24, 2017 7:44:07 AM(UTC)
tylerwight


Rank: Guest

Joined: 5/24/2017(UTC)
Posts: 1
United States

Hello guys,

Hope this is the right place. I have done a lot of googling and could only find a few references to this issue:

http://forum.screenconne...2903_High-CPU-usage.aspx

This one was from a while ago with a much lower version. Is this still an issue with Linux?


Every time I try and migrate our ScreenConnect server from Windows to Linux, once I start it up Mono CPU usage shoots to %150~. We have around 3700 clients on our machine. These are running in EC2 instances, large size (8gb ram 2cpus). CPU usage stays below 50 percent on Windows instance at all times.

Windows is running versions 6.0, after the linux install it's running 6.2 because the upgrade script that converts the web.config, does this automatically.

I have tried following this guide:
https://docs.connectwise...allation_to_a_new_server

as well as copying the whole folder and running the install. I have everything "working" (ssl, sever, etc.) but it is so dead slow it is unusable. I have left it running for over 3 hours to see if everything just needed to 'check in' but still nothing. I can login but each page takes 5+ minutes to open (even on meetings page which we don't use).

I have tried this on Ubuntu 14.04, 16.04, centos 7.3, all of them fully updated and all the same issue. Using IOtop I see no disk write, it uses very little memory and I have no iowait. All I can see is massive cpu usage. All ports are opened correctly using tcpdump I can see traffic going both in and out.

After all this, I'm wondering. Is this still a bug with linux?

Code:
top - 07:38:53 up 23 min,  2 users,  load average: 1.23, 1.77, 1.50
Tasks: 133 total,   2 running, 131 sleeping,   0 stopped,   0 zombie
%Cpu(s): 69.5 us,  0.5 sy,  0.0 ni, 30.0 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st
KiB Mem :  8173748 total,  7163884 free,   510812 used,   499052 buff/cache
KiB Swap:        0 total,        0 free,        0 used.  7391924 avail Mem 

  PID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND                                                                                                     
 2249 root      20   0 2091320 700532 289252 S 140.3  8.6  27:18.12 mono                                                                                                        
    1 root      20   0   37908   5872   3900 S   0.0  0.1   0:01.95 systemd     



iotop shows almost no disk write after running for more than a minute


Code:
Total DISK READ :       0.00 B/s | Total DISK WRITE :       0.00 B/s
Actual DISK READ:       0.00 B/s | Actual DISK WRITE:       0.00 B/s
  PID  PRIO  USER     DISK READ  DISK WRITE  SWAPIN     IO>    COMMAND                                                                                                          
  316 be/3 root          0.00 B      8.00 K  0.00 %  0.01 % [jbd2/xvda1-8]



using strace I see the same thing over and over again
Code:

strace: Process 2249 attached
rt_sigsuspend(~[INT QUIT ABRT TERM XCPU RTMIN RT_1], 8) = ? ERESTARTNOHAND (To be restarted if no handler)
--- SIGXCPU {si_signo=SIGXCPU, si_code=SI_TKILL, si_pid=2249, si_uid=0} ---
rt_sigreturn({mask=~[INT QUIT ABRT KILL STOP RTMIN RT_1]}) = -1 EINTR (Interrupted system call)
futex(0x965cc0, FUTEX_WAKE_PRIVATE, 1)  = 1
rt_sigreturn({mask=[TERM]})             = -1 EINTR (Interrupted system call)
rt_sigtimedwait([TERM], NULL, NULL, 8)  = -1 EINTR (Interrupted system call)
--- SIGPWR {si_signo=SIGPWR, si_code=SI_TKILL, si_pid=2249, si_uid=0} ---
rt_sigsuspend(~[INT QUIT ABRT TERM XCPU RTMIN RT_1], 8) = ? ERESTARTNOHAND (To be restarted if no handler)
--- SIGXCPU {si_signo=SIGXCPU, si_code=SI_TKILL, si_pid=2249, si_uid=0} ---
rt_sigreturn({mask=~[INT QUIT ABRT KILL STOP RTMIN RT_1]}) = -1 EINTR (Interrupted system call)
rt_sigreturn({mask=[TERM]})             = -1 EINTR (Interrupted system call)
rt_sigtimedwait([TERM], NULL, NULL, 8)  = -1 EINTR (Interrupted system call)
--- SIGPWR {si_signo=SIGPWR, si_code=SI_TKILL, si_pid=2249, si_uid=0} ---
rt_sigsuspend(~[INT QUIT ABRT TERM XCPU RTMIN RT_1], 8) = ? ERESTARTNOHAND (To be restarted if no handler)
--- SIGXCPU {si_signo=SIGXCPU, si_code=SI_TKILL, si_pid=2249, si_uid=0} ---
rt_sigreturn({mask=~[INT QUIT ABRT KILL STOP RTMIN RT_1]}) = -1 EINTR (Interrupted system call)
rt_sigreturn({mask=[TERM]})             = -1 EINTR (Interrupted system call)
rt_sigtimedwait([TERM], NULL, NULL, 8)  = -1 EINTR (Interrupted system call)


Scott  
#2 Posted : Wednesday, May 24, 2017 12:59:32 PM(UTC)
Scott


Rank: Administration

Medals: Level 4: Wise Old Owl! Received 100 Thanks!

Joined: 3/28/2014(UTC)
Posts: 2,417
United States

Thanks: 3 times
Was thanked: 298 time(s) in 256 post(s)
ScreenConnect Team
Django  
#3 Posted : Tuesday, June 20, 2017 7:57:31 AM(UTC)
Django


Rank: Newbie

Joined: 2/24/2015(UTC)
Posts: 5
France
Location: Lyon

yep, go back to windows, i have 2000 clients and switching from linux to windows go run clearly more smooth
Users browsing this topic
Forum Jump  
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.