updated to 2.0.2 - http issues

[googol]

http server keeps going down on one instance (I monitor it from my network/server monitoring tool)

I try to access the web interface when this alarm goes off, and yep, cant get to the web page, therefore no login.

I reboot the VM to restore http for now. I did the memory changes for 1.0.4, do I have to redo those changes I found in your knowledgebase for 2.0.2 to increase the memory used for elastica or something? I am giving the VM 12GB of RAM.

[npolovenko]

Hi, @googol. How many logs are you monitoring with the log server? Can you upload the acess_log, error_log, and the ssl_error_log from /var/log/ directory? What happens if you run service httpd status? Also please run a df command to check whether a hard drive is running out of space.

[googol]

attached logs for httpd, guessing thats what you wanted

disk space is not full

I'll try the service httpd status next time it dies

[npolovenko]

@googol, Yes, that's what I meant, sorry. Please open /etc/php.ini and change the:

Code: Select all

memory_limit = 128M
to
memory_limit = 512M

After that please run:

Code: Select all

service httpd restart

Let me know if that fixed your problem.

[googol]

not a problem, so basically after every time I update the software, I need to do set these settings?

looks like I set one for logstash, not for php.ini

[googol]

actually, memory_limit is already set to 512M in /etc/php.ini

[googol]

just happened again, confirmed that php.ini is 512M.

I did service httpd status and it says httpd pid 1656 is running...

but not responding to http requests.

[tmcdonald]

I'd like to get some more information about your environment. Please provide the following:

• OS and version?
• If multiple hosts in cluster, is it just this one that is having issues?
• How long does it take on average for the http daemon to die?
• Is there anything else running on this server aside from Log Server?

[googol]

This is your OVA (1.0.4 version), upgraded to 2.0.2 using your upgrade script following your steps, I've done yum update to see if updates to packages help.

nothing else installed on this system, dedicated to nagios and only nagios.

I have two instances, one in each datacenter. only my secondary datacenter's nagios instance is having this issue at this time. They both should be the same setup.

uname -r shows 2.6.32-279.11.1.el6.x86_64

nagios login prompt shows CentOS release 6.8 (Final)

havent seen anything to indicate a particular time or time period, or frame why it goes down when it does, but I will try to track that for you.

didnt have this problem on 1.0.4, dunno whats going on

[cdienger]

Attached is also a script to gather a profile from the command line. Copy this to the machine and from the command line run:

Code: Select all

chmod 755 profile.sh
./profile.sh

This will generate a file called system-profile.tar.gz in /tmp. Please PM me this as well as recent log files from /var/log/logstash and /var/log/elasticsearch.