Odd problem of all accounts offline -- rebuilt PHP and EasyApache but error persists

[Guest]

I started noticing a problem about 18 hours ago. I can SSH/FTP to the entire CloudLinux/WHM server, plus I can SSH/FTP to any of the accounts on the server.

But I cannot visit any of the accounts websites.

The error_log for each account do not show much in most accounts, but some do show this:

[13-Oct-2016 22:47:05 UTC] PHP Warning:  PHP Startup: Unable to load dynamic library /usr/local/lib/php/extensions/no-debug-non-zts-20131226/homeloader.so - /usr/local/lib/php/extensions/no-debug-non-zts-20131226/homeloader.so: failed to map segment from shared object: Cannot allocate memory in Unknown on line 0

I re-built EasyApache but nothing seemed to change and all sites are all offline. I also ran this aftre the build:

cagefsctl --force-update

Any other tips or ideas on how to troubleshoot this problem? I didnt do anything manually on the server in the past few weeks -- just seemed to happen all of a sudden.

Thank you!

[Guest]

I should also add that I use Uptime Robot for each accounts website on the server and for the past 18 hours it shows downtime for about 20min and then uptime for 5min and then downtime again for 15-20min, and so on. When it shows uptime, I visit a website and get a 500 Internal Server Error

[admin]

What does free -m show?

[Guest]

Thanks for your help

# free -m
total used free
shared buffers cached
Mem: 3801 3766 34
0 36 471
-/+ buffers/cache: 3257 543
Swap: 1023 27 996

[Guest]

Is there useful info there hat can show me where to start seeking a solution? Thank you

Is this a RAM issue or something nothing else? Server load is generally low.

[iseletsk]

Your best course of action is to contact https://helpdesk.cloudlinux.com
Our support is free.

[Guest]

Thank you, I will contact support!

I just doubled my RAM and size on the VPS but the problem persists. I see this:

# df -h

Filesystem      Size  Used Avail Use% Mounted on

/dev/vda1       197G   51G  137G  27% /

tmpfs           3.9G     0  3.9G   0% /dev/shm

/usr/tmpDSK     4.0G  137M  3.7G   4% /tmp

# uptime

10:13:19 up 1 min,  1 user,  load average: 1.66, 0.50, 0.17

[bogdan.sh]

Just to update this thread, according to ticket I see there were two major changes done to resolve this issue:
- disable virtual memory limits for all accounts
- disable shell fork bomb protection