Virtualmin 3.96 [SECURITY] _ fcgid disabled


For virtual servers using CGI or fcgid mode for executing PHP, mod_php mode is now forcibly disabled to prevent potential security issues. This is also done for all domains at installation time.

  1. why fcgid changet to mod_php ?
  2. did it is not secure to move back to fcgid ?
  3. also last 20 days i need to restart apache every day,
    becouse of swap file is become hight and ram is out
    (most of all websites were as fcgid )
    2GB RAM, and this wasnt before …

Howdy,

The PHP Execution Mode shouldn’t be changed. If it was using FCGID, it should still be using FCGID.

All the Virtualmin change does is prevent Apache from executing scripts as mod_php when FCGID is being used on that Virtual Server… and it does that by adding this single line into the Apache config for that VirtualHost:

php_admin_value engine Off

The problem you’re seeing – did that occur with just one Virtual Server, or with all of them?

-Eric

  1. and 2) all is ok it is my mistake

    • lot of servers with fcgid ( approx 40 of 50 )

for now in procecess

8014 www-data 274060 kB /usr/sbin/apache2 -k start

ATOP - vcp 2012/11/28 16:29:02 10 seconds elapsed
PRC | sys 0.27s | user 1.77s | #proc 321 | #zombie 0 | #exit 0 |
CPU | sys 2% | user 18% | irq 0% | idle 170% | wait 10% |
cpu | sys 1% | user 10% | irq 0% | idle 85% | cpu000 w 3% |
cpu | sys 1% | user 8% | irq 0% | idle 85% | cpu001 w 7% |
CPL | avg1 0.35 | avg5 0.42 | avg15 0.43 | csw 3320 | intr 3672 |
MEM | tot 1.9G | free 94.8M | cache 284.7M | buff 28.9M | slab 56.3M |
SWP | tot 2.9G | free 2.4G | | vmcom 3.1G | vmlim 3.9G |
DSK | sdb | busy 8% | read 10 | write 83 | avio 8 ms |
DSK | sda | busy 6% | read 19 | write 83 | avio 6 ms |
NET | transport | tcpi 711 | tcpo 808 | udpi 1 | udpo 1 |
NET | network | ipi 712 | ipo 815 | ipfrw 0 | deliv 712 |
NET | eth0 0% | pcki 433 | pcko 533 | si 50 Kbps | so 455 Kbps |
NET | lo ---- | pcki 282 | pcko 282 | si 473 Kbps | so 473 Kbps |

PID MINFLT MAJFLT VSTEXT VSIZE RSIZE VGROW RGROW MEM CMD 1/1
1589 3 0 8982K 1.1G 65908K 0K 0K 3% mysqld
7741 779 0 7140K 205.3M 54128K 0K 0K 3% php5-cgi
3977 324 0 7140K 178.5M 25376K 0K 28K 1% php5-cgi
8014 8 0 440K 267.6M 14360K 0K 0K 1% apache2
29486 1613 0 7140K 169.3M 13844K 0K 0K 1% php5-cgi
7512 5 0 440K 267.6M 13808K 0K 12K 1% apache2
6755 31 0 440K 266.6M 13188K 0K 0K 1% apache2
5320 46 0 440K 266.6M 13180K 0K 0K 1% apache2
8646 2 0 440K 266.8M 13132K 0K 0K 1% apache2
4698 1 0 440K 266.7M 12860K 0K 0K 1% apache2
4690 35 0 440K 266.8M 12804K 0K 0K 1% apache2
5967 1355 0 7140K 167.0M 10464K 288K 288K 1% php5-cgi
6192 1781 0 7140K 167.0M 10464K 40K 24K 1% php5-cgi
9163 17 0 440K 263.5M 7484K 0K 0K 0% apache2
9168 4 0 440K 263.4M 7336K 0K 0K 0% apache2
5972 6 0 7140K 162.6M 7156K 0K 0K 0% php5-cgi
13633 0 0 586K 77496K 3880K 0K 0K 0% nginx
10341 1320 0 111K 13808K 3576K 0K 0K 0% atop
10283 0 0 451K 72588K 3348K 0K 0K 0% sshd
13154 0 0 4K 77160K 2644K 0K 0K 0% miniserv.pl
1398 3 0 35K 23568K 432K 0K 0K 0% cron
26 0 0 0K 0K 0K 0K 0K 0% kondemand/0
338 0 0 0K 0K 0K 0K 0K 0% md2_raid1
358 0 0 0K 0K 0K 0K 0K 0% jbd2/md2-8

Howdy,

So just to clarify – are you saying that everything is okay now, except that over the last month or so you need to restart Apache once a day in order to keep RAM usage low?

What Linux distribution/version are you using?

And how much RAM/swap do you have? You can determine that by running the command “free -m”.

-Eric

yes

debian 6

free -m
total used free shared buffers cached
Mem: 1980 1917 63 0 19 231
-/+ buffers/cache: 1666 314
Swap: 2980 647 2332

also very long response from apache when website opens onces in 10 hours (approx)
(1-2 users on website in day)
and apache error logs get this
(think this problems are connected together )

[Thu Nov 29 12:46:41 2012] [warn] [client 37.55.41.22] mod_fcgid: read data timeout in 31 seconds
[Thu Nov 29 12:46:41 2012] [error] [client 37.55.41.22] Premature end of script headers: index.php