- observium - lists.observium.org

LDAP Configuration
by Michael Sweikata 02 Apr '13

02 Apr '13

Hey everyone; Answers so far have been wonderful. I've been pouring through the archives trying to find a good example for LDAP Authentication, and I believe I have the authentication portion working correctly, however security permissions in Observium aren't working correctly. If I provide the correct credentials, it lets me through but the main page has no content. If I provide incorrect credentials, it spits me back out to the login page. The structure of the containers is correct, the AD Security Group 'Network Ops' is the group our Network Management team uses, so I want us to have full access (level 10), and is located in (Canonical name): hh.nku.edu/NKU/HH/Groups/Security/Network Ops I've tried it with 'Domain Users' instead of 'Network Ops' but no dice there either. And yes, I have double-checked that my account is a member of the 'Network Ops' group, I didn't get fired today. My level 7 and level 1 groups have been commented out just to try and simplify any issues. My config.php (relevant portions) look like: // Authentication Model #Required LDAP Settings $config['auth_mechanism'] = "ldap"; $config['auth_ldap_version'] = 3; $config['auth_ldap_server'] = "nkuserv1.hh.nku.edu"; $config['auth_ldap_port'] = 389; $config['auth_ldap_starttls'] = "false"; #User Account $config['auth_ldap_prefix'] = "cn="; $config['auth_ldap_suffix'] = ",OU=Faculty_Staff,OU=Users,OU=HH,OU=NKU,DC=hh,DC=nku,DC=edu"; #Required Group Membership $config['auth_ldap_group'] = "cn=Network Ops,OU=Security,OU=Groups,OU=HH,OU=NKU,dc=hh,dc=nku,dc=edu"; $config['auth_ldap_groupbase'] = "ou=Security,ou=Groups,ou=hh,ou=nku,dc=hh,dc=nku,dc=edu"; #Optional LDAP Settings $config['auth_ldap_groups']['Network Ops']['level'] = 10; #$config['auth_ldap_groups']['Domain Admins']['level'] = 7; #$config['auth_ldap_groups']['Lab Admins']['level'] = 1; So, if anyone has any ideas or can help with this, it'd be greatly appreciated.

1 0

Re: [Observium] False "device rebooted" message
by Adam Armstrong 02 Apr '13

02 Apr '13

All counters. But we can't guarantee any of them exist at that point in the poller, or that they were ever greater than zero. Adam. Morgan McLean <wrx230(a)gmail.com> wrote: >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

1 0

Re: [Observium] False "device rebooted" message
by Adam Armstrong 02 Apr '13

02 Apr '13

It comes from SNMP. What does the uptime graph show? Stefano Lobrano <stefano.lobrano(a)staff.aruba.it> wrote: >Hello List, >I'm testing observium to monitor network devices in our datacenter, I've >added a range of different devices and it seems all good, but today it >has given me a false notice about a couple of nexus 5000 that are >reported as "rebooted" while the switches are correctly reporting uptime >of more than 600 days... >All the other hosts are reported without errors, apparently. > >I've looked in the mailing list archive and I found similar issues, with >server, but I understand that was something resolved. > >Where does observium takes the info about reboot? > >Observium is installed in a VM: >Observium 0.13.3.3824 >Apache 2.2.22 (Ubuntu) >PHP 5.3.10-1ubuntu3.6 >MySQL 5.5.29-0ubuntu0.12.04.2 >RRDtool 1.4.7 > >Thanks, >Stefano >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

4 8

observium alert not working
by Abul Bashar Azad 02 Apr '13

02 Apr '13

Hi Observium Team, BDCOM is one of the big ISP and IP Telephony Service provider in Bangladesh I am deploy it in my IP Telephony network. I am follow your documentation step by step. I am add host and view all the graph its working fine. But I can not see any alert message. I am configure it but its not working. Could you please help me how can i get specific alert like my ram memory 75% use, Processor 60% use. I need customised alert to monitor my netwrok device properly . I am wating for your valuable feedback. -- With regards --- Md. Abul Bashar Azad Assistant Manager, IPTSP *BDCOM Online Limited* IP Phone : 09666333666 Cell: +8801818301627

2 2

Re: [Observium] Observium Storage Requirements
by Tom Laermans 01 Apr '13

01 Apr '13

Next to the rrd requirements, you may want to put some pruning in place for the eventlog and syslog tables. Size of storage per device depends on what is monitored on it. But as said in other answers the data files itself do not expand. Sent from my mobile, please excuse brevity and spelling errors. ----- Reply message ----- From: "Michael Sweikata" <sweikatam1(a)nku.edu> To: "observium(a)observium.org" <observium(a)observium.org> Subject: [Observium] Observium Storage Requirements Date: Mon, Apr 1, 2013 16:51 Hey everyone; I mentioned in a previous e-mail that we’re rolling out a deployment of Observium, and we absolutely love the product. I know that it’s designed as a historical data gathering tool, but I am concerned about the storage requirements for the system. I have approximately 600 routers and switches on my campus, and eventually would like to get my servers involved (pushing the device count beyond two-thousand), and I’m concerned at the storage requirements for this. I couldn’t find this information anywhere online, so I figured I would check with you guys. Does Observium have any kind of roll-over archiving for device history beyond a certain point, or will it continually just grow the data in the files?

2 1

Re: [Observium] Observium Storage Requirements
by Adam Armstrong 01 Apr '13

01 Apr '13

Observium uses red, which is fixed size round robin database. You can find information about how it works online. Perhaps we should document this too. Data usage for a single host shouldn't grow over time, except event log entries and the like. For an installation of that size you might want to use ssd or solid state ram drives. Spindles just won't survive the I/o load. Adam. Michael Sweikata <sweikatam1(a)nku.edu> wrote: >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

1 0

Observium Storage Requirements
by Michael Sweikata 01 Apr '13

01 Apr '13

Hey everyone; I mentioned in a previous e-mail that we're rolling out a deployment of Observium, and we absolutely love the product. I know that it's designed as a historical data gathering tool, but I am concerned about the storage requirements for the system. I have approximately 600 routers and switches on my campus, and eventually would like to get my servers involved (pushing the device count beyond two-thousand), and I'm concerned at the storage requirements for this. I couldn't find this information anywhere online, so I figured I would check with you guys. Does Observium have any kind of roll-over archiving for device history beyond a certain point, or will it continually just grow the data in the files?

2 1

Re: [Observium] only interface counters wrapped
by Adam Armstrong 01 Apr '13

01 Apr '13

Interface counters update much quicker. Probably 500 times quicker than even packets in net stats. But it is perhaps related to old missed data problem. :( Adam Nikolay Shopik <shopik(a)inblock.ru> wrote: >Hey, > >So I just wondering is this normal when some of your counters wrapped >(huge peak) while others are perfectly normal? > >I have device running for 1 years, 103 days, and only interface graph >are wrapped while netstat are just perfectly fine. > >snmpwalk seems report that HC counters supported, device running 12.3 IOS. >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

2 2

only interface counters wrapped
by Nikolay Shopik 01 Apr '13

01 Apr '13

Hey, So I just wondering is this normal when some of your counters wrapped (huge peak) while others are perfectly normal? I have device running for 1 years, 103 days, and only interface graph are wrapped while netstat are just perfectly fine. snmpwalk seems report that HC counters supported, device running 12.3 IOS.

1 0

False "device rebooted" message
by Stefano Lobrano 31 Mar '13

31 Mar '13

Hello List, I'm testing observium to monitor network devices in our datacenter, I've added a range of different devices and it seems all good, but today it has given me a false notice about a couple of nexus 5000 that are reported as "rebooted" while the switches are correctly reporting uptime of more than 600 days... All the other hosts are reported without errors, apparently. I've looked in the mailing list archive and I found similar issues, with server, but I understand that was something resolved. Where does observium takes the info about reboot? Observium is installed in a VM: Observium 0.13.3.3824 Apache 2.2.22 (Ubuntu) PHP 5.3.10-1ubuntu3.6 MySQL 5.5.29-0ubuntu0.12.04.2 RRDtool 1.4.7 Thanks, Stefano

3 2