Extremely slow load times for graphs after PHP upgrade

[rkennedy]

Could you PM a profile over for us to look at? I'd like to take a deeper look at your environment. (Admin -> System Profile -> Download Profile)

EDIT: Profile received.

[rkennedy]

I noticed a few things when looking over your profile, which I'll point out here. Is there a reason you're running gnome on top of XI? This isn't reccomended at all.

Looks older, but regardless, something is spiking your load. -

Code: Select all

[04-03-2016 13:09:47] NPCD: WARN: MAX load reached: load 10.640000/10.000000 at i=1
[04-03-2016 13:12:04] NPCD: WARN: MAX load reached: load 11.410000/10.000000 at i=0
[04-03-2016 13:12:19] NPCD: WARN: MAX load reached: load 10.150000/10.000000 at i=1
[04-04-2016 09:26:18] NPCD: WARN: MAX load reached: load 13.770000/10.000000 at i=0
[04-04-2016 09:26:33] NPCD: WARN: MAX load reached: load 11.520000/10.000000 at i=1

SQL and Postgres both possibly not running - what happens if you start them? -

Code: Select all

[Fri Apr 15 09:50:25.349252 2016] [:error] [pid 6282] [client 10.0.219.126:44365] PHP Warning:  mysql_pconnect(): Link to server lost, unable to reconnect in /usr/local/nagiosxi/html/db/adodb/drivers/adodb-mysql.inc.php on line 383, referer: http://monitoring.atriacom.com/nagiosxi///includes/page-home-main.php?&=
[Fri Apr 15 09:50:25.365107 2016] [:error] [pid 6283] [client 10.0.219.126:44364] PHP Warning:  pg_pconnect(): PostgreSQL link lost, unable to reconnect in /usr/local/nagiosxi/html/db/adodb/drivers/adodb-postgres64.inc.php on line 682, referer: http://monitoring.atriacom.com/nagiosxi///includes/page-home-main.php?&=

I see there are multiple NICs on the system, can you please run nslookup monitoring.atriacom.com from the XI system? Also, attach your /etc/hosts file.

How much resources do you have allocated to this machine? (Ram / CPU's)

[Atria]

When I check my running services I see the following

Code: Select all

[frack@monitoring ~]$ systemctl list-units | grep sql
  postgresql.service                                                                        loaded active running   PostgreSQL database server

an nslookup from my server returns

Code: Select all

[frack@monitoring ~]$ nslookup monitoring.atriacom.com
Server:         10.0.4.20
Address:        10.0.4.20#53

Name:   monitoring.atriacom.com
Address: 172.30.0.14

Gnome is running for the Linux challenged among the team to be able to access an RDP. If having it running is a problem I can strip it off.

specs of the VM are as follows

RAM: 16 GB
CPUS: 2 Sockets, 4 Cores
Single NIC card assigned in VM

I will say that when I ran systemctl list-units I noticed a few failed services, a couple of which are likely due to a failed attempt to get LDAP authentication working.

Code: Select all

[frack@monitoring ~]$ systemctl list-units | grep failed
● fprintd.service                                                                           loaded failed failed    Fingerprint Authentication Daemon
● rngd.service                                                                              loaded failed failed    Hardware RNG Entropy Gatherer Daemon
● sssd.service                                                                              loaded failed failed    System Security Services Daemon
● winbind.service                                                                           loaded failed failed    Samba Winbind Daemon

Is it likely sssd or winbind are causing some sort of problem?

[rkennedy]

All looks good from the outputs.

The upgrade of PHP might have done something. Can you open up a ticket so that we can get this moved into a remote?