- observium - lists.observium.org

Overview
by Adam Summerfield 15 Jan '13

15 Jan '13

Attempting to put this into production with our NOC. When they logon with the user class of "Global Read" they can see the devices but nothing appears on the overview page. If I create the user with class of "Normal" they cannot see anything. Admin class of course works as expected. Regards, Adam

1 0

Postgres monitor script
by Lance Raymond 15 Jan '13

15 Jan '13

Afternoon all, have my apache working nicely, mysql and moving through the list of to-do's. Next is our primary DB which is running postgres. I have that server added, SNMP working, the agent is loaded and I can telnet from the observium server on 6556 port with no issues. Looking at the install http://www.observium.org/wiki/Application/Postgresqlthe only thing left from that section and my notes from mysql is to copy the script from the local observium server (under /opt/observium/scripts/agent-local) to the postgres server and update the config with the user/pass but those files are not there. I also don't see postgres listed in the applications list for the device. Is postgres not supported anymore? -- *e*. lraymond(a)weatherflow.com *c.* 386.561.1509 *f.* 386.409.5178

1 0

Function scope issues?
by Paul Gear 11 Jan '13

11 Jan '13

Hi all, I'm just looking into a couple of nagging problems on my Observium install and i found this: 1. includes/discovery/discovery-protocols.inc.php uses ifNameDescr(), which is defined in includes/rewrites.php (called from includes/functions.php). 2. ifNameDescr() is a wrapper around ifLabel(), which itself is a wrapper around humanize_port() 3. humanize_port() calls humanspeed(), which is defined in html/includes/functions.inc.php This seems to indicate that ifNameDescr/ifLabel/humanize port, when called from discovery functions, will never work. Is this intentional? Or should humanspeed() (and possibly others of its peers within html/includes/functions.inc.php) be moved to includes/rewrites.php? Regards, Paul

3 3

Disable Monitoring of Linux Interfaces
by admin 11 Jan '13

11 Jan '13

Good Evening, I was able to get Observium up and running with no issues, but I do have a bit of an odd situation. I happen to have a server that I wish to monitor that has over 300 PPP interfaces. Observium is attempting to poll them, and its hammering the server hard, and Observium. Is there a way to remove polling of interfaces (or maybe just ppp interfaces) on that server? Thanks!

2 6

Drilling down to specific times in graphs
by Jason Lixfeld 11 Jan '13

11 Jan '13

I'm having a little trouble understanding how the graph drill downs are supposed to work. I select a 6 hour graph of something, From: says 17:44 To: says 23:44. Case 1: I change 17:44 to 16:44 and click Submit, From: says 11:44 To: says 18:44. Case 2: I leave From: at 17:44 but change To: to 18:44 and click Submit, From: says 12:44 To: says 13:44. Am I missing something on how these timeframes are supposed to work?

3 2

Re: [Observium] Apache application monitor problem
by Adam Armstrong 09 Jan '13

09 Jan '13

Observium doesn't hit apache directly. That is done via the agent script. Is it working? Lance Raymond <lraymond(a)weatherflow.com> wrote: >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

4 6

Re: [Observium] Function scope issues?
by Adam Armstrong 09 Jan '13

09 Jan '13

yes :-) Though really the processing should be done at polling time. Its all a mess caused by gradual evolution! Paul Gear <observium(a)gear.dyndns.org> wrote: >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

1 0

Apache application monitor problem
by Lance Raymond 09 Jan '13

09 Jan '13

After a successful implement of the mysql module, I now am looking to monitor apache. I will try to give as much on the 1st post as I do now understand how the agents work a bit more (Thanks Adam) so the info and questions are; Unix-agent is installed, enabled, port is open, and I can hit the server-status from the observium server via curl; curl stats.domain.com/server-status <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2 Final//EN"> <html><head> <title>Apache Status</title> </head><body> <h1>Apache Server Status for stats.domain.com</h1> <dl><dt>Server Version: Apache/2.2.22 (Ubuntu) PHP/5.3.10-1ubuntu3.4 with Suhosin-Patch</dt> <dt>Server Built: Nov 8 2012 21:37:30 </dt></dl><hr /><dl> <dt>Current Time: Wednesday, 09-Jan-2013 04:00:49 UTC</dt> <dt>Restart Time: Tuesday, 08-Jan-2013 20:58:01 UTC</dt> <dt>Parent Server Generation: 3</dt> <dt>Server uptime: 7 hours 2 minutes 47 seconds</dt> <dt>Total accesses: 651 - Total Traffic: 1.7 MB</dt> <dt>CPU Usage: u9.48 s1.89 cu0 cs0 - .0448% CPU load</dt> <dt>.0257 requests/sec - 71 B/second - 2773 B/request</dt> <dt>1 requests currently being processed, 9 idle workers</dt> Using the poller command to test server; ubuntu@websvr1:/opt/observium$ ./poller.php -v -h stats.domain.com Observium Poller v0.12.11.3407 Starting polling run: stats.domain.com 5 linux (unix) Observium UNIX Agent: execution time: 328ms Processes: Sensors: Using UNIX Agent Uptime (2877622) Uptime: 33 days, 7h 20m 22s Hardware: Generic x86 64-bit Version: 3.2.0-31-virtual Features: /usr/bin/distro: No such file or directory Serial: IPMI: Processor Intel Xeon E5507 @ 2.27GHz 1% Mempool Physical memory: 79.6% Mempool Virtual memory: 51.63% Mempool Swap space: 0.01% Storage /: 17% Storage /mnt: 0% Polling Netstats: IP TCP UDP ICMP SNMP HR Stats: Processes Users Polling IP-MIB ipSystemStats ipv4 ipv6 Caching Oids: ifEntry ifXEntry dot3StatsDuplexStatusvalidvalid Port lo(1) HC HC VLAN == bps(0bps/0bps)bytes(0B/0B)pkts(0pps/0pps) Port eth0(2) HC HC VLAN == bps(1.32kbps/4kbps)bytes(4.83kB/14.6kB)pkts(2.2pps/2.2pps) Checking UCD DiskIO MIB: xvda1 xvdb xvda3 OSPF: Processes: Areas: Ports: Neighbours: Entity Physical: Applications: agent uptime netstat_ip netstat_ip_frag netstat_tcp netstat_udp netstat_icmp netstat_icmp_info netstat_snmp netstat_snmp_pkt hr_processes hr_users ucd_cpu ucd_swap_io ucd_io ucd_contexts ucd_interrupts ucd_memory ucd_load ipsystemstats_ipv4 ipsystemstats_ipv4_frag ipsystemstats_ipv6 ipsystemstats_ipv6_frag Polled in 4.680 seconds UPDATED! MySQL: Cell[0/0s] Row[3/0s] Rows[25/0.01s] Column[0/0s] Update[9/0.01s] Insert[5/0.01s] Delete[0/0s] So, it seems the agent is running, apache is reachable by the server with the server-status, just not sure why the graph page is broken. The main graph's all say Error Drawing Graph, and the scoreboard at the bottom just has a few broken image links. -- *e*. lraymond(a)weatherflow.com *c.* 386.561.1509 *f.* 386.409.5178

1 0

Re: [Observium] Adding mysql monitoring
by Adam Armstrong 09 Jan '13

09 Jan '13

The agent is not enabled for that host :-) Lance Raymond <lraymond(a)weatherflow.com> wrote: >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

3 3

Re: [Observium] Adding mysql monitoring
by Adam Armstrong 08 Jan '13

08 Jan '13

I just added apache/mysql agents to a clean install and it worked OK. Run the poller with -d Adam. Lance Raymond <lraymond(a)weatherflow.com> wrote: >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

2 1