Search found 11 matches
- 10 Jul 2017, 21:46
- Forum: Report Bugs (csf)
- Topic: LFD HTTPS messenger used all memory and cpu
- Replies: 3
- Views: 5156
Re: LFD HTTPS messenger used all memory and cpu
1. What OS, build and kernel version is the server running? CloudLinux 7.3, kernel 3.10.0-614.10.2.lve1.4.55.el7.x86_64 2. What version of Apache is being used? Server version: Apache/2.4.25 (cPanel) Server built: Apr 14 2017 12:43:52 3. How many domains are being served? ~ 600 4. Where is MESSENGER...
- 10 Jul 2017, 20:33
- Forum: Report Bugs (csf)
- Topic: LFD HTTPS messenger used all memory and cpu
- Replies: 3
- Views: 5156
Re: LFD HTTPS messenger used all memory and cpu
And one more think I would add...
MESSENGER_CHILDREN is set to "10" and here is 36 child processes...
MESSENGER_CHILDREN is set to "10" and here is 36 child processes...
- 10 Jul 2017, 20:19
- Forum: Report Bugs (csf)
- Topic: LFD HTTPS messenger used all memory and cpu
- Replies: 3
- Views: 5156
LFD HTTPS messenger used all memory and cpu
Helo, We just had a case that server crashed because of messenger memory usage (160+ GBs or memory)... Here is the relevan part of the process list... Output from ps: USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND root 11754 0.2 0.0 280872 49724 ? Ss 00:00 2:04 lfd - sleeping csf 11772 0.0 3...
- 13 Jul 2015, 22:15
- Forum: General Discussion (csf)
- Topic: LFD stopped working after 8.0 upgrade
- Replies: 3
- Views: 3002
Re: LFD stopped working after 8.0 upgrade
Yes, regex.custom.pm was the problem.
Thank you very much.
Thank you very much.
- 13 Jul 2015, 21:48
- Forum: General Discussion (csf)
- Topic: LFD stopped working after 8.0 upgrade
- Replies: 3
- Views: 3002
LFD stopped working after 8.0 upgrade
After upgrade to latest version LFD stopped working on one server only... This is the error reported in the lfd.log... : Can't use string ("31.8.173.147") as a SCALAR ref while "strict refs" in use at /usr/local/csf/lib/ConfigServer/CheckIP.pm line 33. : daemon stopped I tried to...
- 25 Jun 2014, 17:55
- Forum: General Discussion (csf)
- Topic: Log System?
- Replies: 2
- Views: 2949
Re: Log System?
You can ignore that warning when rsyslog is used.
- 23 Jun 2014, 20:14
- Forum: General Discussion (csf)
- Topic: IPs not being blocked for ModSec
- Replies: 5
- Views: 8447
Re: IPs not being blocked for ModSec
We have thousands of those login warnings per hour, and I always thought that CSF/LFD should block those repeated warnings. I just find out that all those login attempts are ignored by LFD, but they are actually very important!
Does anyone have a custom regex for this?
Does anyone have a custom regex for this?
- 26 Feb 2014, 17:18
- Forum: Report Bugs (csf)
- Topic: DROP_OUT_LOGGING option can't be disabled
- Replies: 2
- Views: 3539
DROP_OUT_LOGGING option can't be disabled
DROP_OUT_LOGGING option is disabled, but we still receive a lot of "Tracking Hit" notifications with "Firewall: *TCP_OUT Blocked*" on all servers. I also see that LOGDROPOUT chain is really still configured in iptables after CSF restart... Is this a bug, or I am missing some addi...
- 03 Jun 2013, 11:02
- Forum: Suggestions (csf)
- Topic: CSF&LFD standalone remote console (new product)
- Replies: 1
- Views: 3225
CSF&LFD standalone remote console (new product)
It would be great if we could have a central CSF&LFD standalone console. That should be a completely separate product that will run alone on separate server/vps, and allow complete control of all CSF&LFD servers in a cluster. It should also have multiple admin user authentication with differ...
- 18 Feb 2013, 18:21
- Forum: General Discussion (csf)
- Topic: OpenVZ VPS's inaccesible after update to v5.76
- Replies: 3
- Views: 4393
Re: OpenVZ VPS's inaccesible after update to v5.76
I just realised what the problem is. We didn't have ipt_conntrack module loaded on any VZ node. # modprobe ipt_conntrack All VE's that use CSF/LFD should also be restarted after that. This should also be added to /etc/vz/vz.conf module list. Since all VE's should be restarted for new modules to be l...