Skip to main content

lfd - Suspicious process running under user nobody

Comments

5 comments

  • rbairwell
    If you have (under WHM->Tweak Settings) " Thunderbird and Outlook autodiscover and autoconfig support" enabled, then this script will operate on autodiscover/autoconfig subdomains to help provide relevant information to Outlook/Thunderbird etc to configure it. To stop this messages, you can either disable that option or in CSF->LFD->Edit lfd ignore file->csf.pignore (Process tracking) and add at the end "exe:/usr/local/cpanel/cgi-sys/autodiscover.cgi" which should cause LFD to ignore that script from then on.
    0
  • Otávio Serra
    If you have (under WHM->Tweak Settings) " Thunderbird and Outlook autodiscover and autoconfig support" enabled, then this script will operate on autodiscover/autoconfig subdomains to help provide relevant information to Outlook/Thunderbird etc to configure it. To stop this messages, you can either disable that option or in CSF->LFD->Edit lfd ignore file->csf.pignore (Process tracking) and add at the end "exe:/usr/local/cpanel/cgi-sys/autodiscover.cgi" which should cause LFD to ignore that script from then on.

    Ok. But this overload is a problem or not? This behavior is occurring on my server more or less three times a day. Because after this message, I receive one other message like this: Time: Thu Oct 27 08:39:35 2022 -0300 1 Min Load Avg: 21.58 5 Min Load Avg: 7.69 15 Min Load Avg: 2.81 Running/Total Processes: 2/397
    0
  • martin MHC
    A load of 21 looks quite high to me, but I think the actual load rating (ie if it's a lot or not for the system to handle) depends on how many Cores you're operating with on your server
    0
  • rbairwell
    Yep, I agree the load is very high and the autoconfiguration script shouldn't be producing that much load (in fact, it should be negligible). You say you've had this problem for a while now: do you always get the Suspicious Process email just before the high load issue? (the two issues could be connected or they could be two separate issues - or there could actually be a third issue underlying them. I've not used the autodiscover tool myself (yet) so I can't quite comment - but I'm spinning up a test instance to investigate further (as I want to know myself ; ) ) If you are able at the time, run the following command as root: netstat -Wat -o state established | egrep '(:http|:https) ' | awk '{print $5 " connected to " $4}' | sort | uniq -c
    which will list all established connections to http/https ports (with a count of connections) of which hosts are connecting to your server - it might give some guidance if an attack of some sort is happening. If you've got something like 10+ connections from a single host, then that's the source of your problems. Looking at the "Apache Status" page in WHM might give some clues as well.
    0
  • Otávio Serra
    I found the problem: my server had few memory. Output from vmstat: procs -----------memory---------- ---swap-- -----io---- -system-- ------cpu----- r b swpd free buff cache si so bi bo in cs us sy id wa st 13 37 1298568 201152 108 241724 14 13 1857 435 5 10 5 1 91 1 2
    0

Please sign in to leave a comment.