Question Plesk Panel Extremely Slow and Shows White Screen Despite Server Resources Being Idle

[givemesomoney]

Server operating system version

Ubuntu 22

Plesk version and microupdate number

Latest

Plesk Panel Extremely Slow and Shows White Screen Despite Server Resources Being Idle​

Environment​

• OS: Ubuntu 22 LTS
• Web Server: LiteSpeed
• Server Specs: 10 CPU cores
• Plesk Version: Latest Version

Problem Description​

My Plesk panel interface is extremely slow and often shows a white/blank screen after loading for several minutes. However:

• Server resources are completely idle (CPU at 0-1%, 99-100% idle)
• Websites hosted on the server work perfectly fine and load very fast
• Only Plesk panel interface is affected (all Plesk-related screens)
• When I visit a domain already installed on Plesk, it works very fast
• In short: everything related to Plesk interface is extremely slow, but hosted websites are fine

Additional Findings​

• Found 6190 files in /var/lib/plesk/task-manager/runtime/ directory
• 219 failed login attempts in Plesk action log

Server Resource Status (from dstat):​

usr sys idl wai stl| read writ| recv send| in out | int csw
0 0 100 0 0| 0 0 | 20k 97k| 0 0 |1430 1374
0 0 99 0 0| 0 812k| 208k 4325k| 0 0 |3538 2581

Symptoms​

1. Accessing Plesk at https://domain.com:8443 takes several minutes
2. Browser shows loading spinner, then displays blank white page
3. Network tab shows net::ERR_HTTP2_PROTOCOL_ERROR on some requests
4. Some requests show Status 200 but still don't load completely

Temporary Solution Found​

The only thing that temporarily fixes the issue is killing all sw-engine processes and restarting:

pkill -f sw-engine
service sw-engine start
service plesk-web-socket restart
service sw-cp-server restart


However, the problem returns after some time and I have to repeat this process.

What I've Already Tried​

1. Cleared Plesk cache and sessions:
   • plesk bin cache --clear (command not found in my version)
   • plesk db "TRUNCATE TABLE sessions;"
   • Deleted old log entries
2. Restarted all Plesk services multiple times:
   • service sw-engine restart
   • service sw-cp-server restart
   • systemctl restart psa
3. Killed and restarted sw-engine processes:
   • pkill -f sw-engine
   • This temporarily fixes the issue but it returns after some time
4. Increased PHP/Nginx timeout values:
   • Set max_execution_time to 300
   • Set proxy timeouts to 600
   • Added timeout configurations
5. Checked for resource-heavy processes:
   • Found backup/migration processes running
   • Killed them but issue persists
6. Attempted to optimize database:
   • Optimized Plesk tables
   • Checked for slow queries
7. Security measures:
   • Installed fail2ban
   • Checked SSH logs for brute force attacks
   • Found significant attack attempts

Logs Checked​

• /var/log/plesk/panel.log
• /var/log/sw-cp-server/error_log
• MySQL process list

Questions​

1. What could cause sw-engine processes to hang/freeze regularly?
2. Is there a known issue with sw-engine-fpm becoming unresponsive?
3. How can I permanently fix this without having to manually restart services?
4. Are there any specific Plesk settings I should check for Ubuntu 22 + LiteSpeed combination?

Any help would be greatly appreciated as this is seriously affecting my ability to manage the server.


Thank you!

 

[Sebahat.hadzhi]

Hello, @givemesomoney . Could you please check /var/log/httpd/stderr.log and confirm if there are any entries suggesting you hit LSAPI_CHILDREN?

 

[HHawk]

I have experienced this issue many times myself. Plesk interface being slow, despite a low count of hosting accounts (customers, domains, subscriptions). Load is generally low, at least not high. Memory usage also normal. And it also doesn't matter is on SSDs or NVMe.

Plesk interface is simply, for no apparent reason, slow/sluggish. My guess is, is that it has to do with some internal (Cron) jobs which are running within Plesk. Created a few support tickets about this in the past, but support couldn't find any issues obviously as it's random. Though our server customers do complain about it from time to time.

It's one of those things you should simply accept, I guess. I personally think it's caused by all the interface (UI/UX changes during the years. Too many calls perhaps? Too many queries? Who knows. Sometimes we nowadays get a blank/white screen and even errors like "We run into a problem" (or something like that).

Ah well. It doesn't happen all the time luckily. So it's something you have to accept. Similar to Windows 11 giving a crash/BSOD or something. I can only hope some of the excessive price hikes are not going to the investment company, but used in improving the general Plesk experience and performance, instead of adding useless gimmicks who nobody uses...

Sidenote; I also changed and tweaked a ton of settings (everywhere), but no results and kinda difficult to test because it happens randomly. Heck, it even can happen on a fully clean install of Plesk. So very difficult to troubleshoot to be honest.