- observium - lists.observium.org

Juniper inventory
by Zalewski, Dominik 15 Apr '14

15 Apr '14

Hi, Are there any plans or development in progress of Juniper inventory module? Thanks Dominik

4 8

Alerting on high 'used' RAM, not counting 'cached'
by Robert Williams 15 Apr '14

15 Apr '14

Hi, Having my first experiment with the new alerting features - very nice! There is one thing I can’t seem to get quite right which is how to alert for Linux system when the ‘used’ RAM is greater than, say, 90% - which is best explained with an example of a device which looks like this: [cid:image001.png@01CF5826.67DDE2B0] Which shows that ‘used’ is 15.1GB (48%). However, for the purposes of alerting, if you use the mempool_perc gt 80 it alarms as being 98%. So I can’t specify ‘used’ as a pool because the only pools are “Swap Space”, “Virtual Memory”, “Physical Memory” etc. There isn’t one which is specifically the ‘used’ RAM element. So, is there a way around this or something I can tweak to get the desired result? Or what is everyone else doing to get alerting for linux boxes like this working for high RAM? Cheers guys! Robert Williams Custodian Data Centre Email: Robert(a)CustodianDC.com http://www.CustodianDC.com

3 2

Support for Huawei MA5600T GPON OLT
by Adam Wołk-Jankowski 14 Apr '14

14 Apr '14

Hello, I’m looking for NMS with support for full query and monitoring of MA5600T series Huawei GPON OLT, but with no luck. Huawei NMS – iManager U2000 costs huge amount of money and it not completely suit to my needs. We use Observium for few months with Cisco, Extreme Networks and D-Link devices and it seems to be ideal solution, but with MA5600T there works only basic things like traffic on Ethernet ports. I have full MIB library for this device but I’m not a programmer so cannot make support by myself. I have some devices in lab, so can share access and also would like to help if anyone with skills would like to work on this. -- Adam Jankowski a.jankowski(a)systel.pl

2 2

Device names on front page Google map?
by Ulrik Ivers 14 Apr '14

14 Apr '14

Hi, I'm new to Observium, so please be tolerant if I ask a stupid question. On the Google map on the front page - is it possible to have the device names displayed? Either "always" or when hovering on the green place marker. Now when I hover I get a label stating "1 hosts", "2 hosts" etc depending on the number of devices that have identical location. I've looked through the documentation and searched old discussion archives but haven't been able to find anything regarding this. Regards, /Ulrik Ivers

3 2

Re: [Observium] 6k Clock 2 status alarms
by Moerman, Maarten 14 Apr '14

14 Apr '14

I’m suffering from the same issue since last week or so. I’ve filed http://jira.observium.org/browse/OBSERVIUM-795 for this to get this solved/fixed. I didn’t exactly know what it was related to, but thanks for figuring that out! Thx, Maarten -- Maarten Moerman | Mgr, Network Engineering | eBay Classifieds | +31-655122247 | mmoerman(a)ebay.com From: Robert Williams <Robert(a)CustodianDC.com<mailto:Robert@CustodianDC.com>> Reply-To: Observium Network Observation System <observium(a)observium.org<mailto:observium@observium.org>> Date: Saturday, April 12, 2014 at 1:21 PM To: "Observium Network Observation System (observium(a)observium.org<mailto:observium@observium.org>)" <observium(a)observium.org<mailto:observium@observium.org>> Subject: [Observium] 6k Clock 2 status alarms Hi, Just a minor bug (or feature tweak, depending on your perspective!) when someone is next in the relevant area: [cid:image001.png@01CF5649.04D8EB80] The clock sensor number 2 on all 6k Catalyst chassis should probably not ‘alarm’ by default when it’s not in use. There will always be only one of the two clocks in use, so some kind of logic tweak is needed to treat this differently otherwise you always have a load of status warnings for the category of “State” under “Health”. Unless you go through them all and disable them, but then you won’t get an event when they do become active (which is usually a bad thing as something else has failed). I’ve seen it show up as these two names depending what IOS it’s running: OSR-7600 Clock FRU 2 - clock 2 clock-inuse Sensor<http://observium.cdc.local/graphs/id=579946/type=sensor_state> OSR-7600 Clock FRU 2 In Using Sensor - Clock 2 In Using Sensor<http://observium.cdc.local/graphs/id=580144/type=sensor_state> Actually, since there is also a value for the ‘ok-ness’ of the sensors it is probably worth totally skipping the ‘in-use’ element for the default alarming purposes, imho. If you need more info from me then let me know, cheers! Robert Williams Custodian Data Centre Email: Robert(a)CustodianDC.com<mailto:Robert@CustodianDC.com> http://www.CustodianDC.com

1 0

device status boxes on the front page show disabled devices
by Henrik Lund Kramshøj 14 Apr '14

14 Apr '14

Hi We have a problem with the front page, the red alert boxes are shown for devices that are disabled. We use the config.php $config['frontpage']['order'] = array('status_summary', 'device_status_boxes', 'device_status', 'overall_traffic' ); and I guess its the device_status_boxes is the one. Did something change, or do we need to change something? Devices shown (some of them) has check mark in: "Disable Disables polling and discovery. Device ignore Device ignore. " So they shouldnt show up, didnt sometime before - but unknown when they stopped working correctly. Version, subscription, did a svn up today. man apr 14 09:14:08 CEST 2014 Best regards Henrik -- Henrik Lund Kramshøj, Follower of the Great Way of Unix internet samurai cand.scient CISSP hlk(a)kramse.org hlk(a)solidonetworks.com +45 2026 6000 http://solidonetworks.com/ Network Security is a business enabler

3 3

6k Clock 2 status alarms
by Robert Williams 12 Apr '14

12 Apr '14

Hi, Just a minor bug (or feature tweak, depending on your perspective!) when someone is next in the relevant area: [cid:image001.png@01CF5649.04D8EB80] The clock sensor number 2 on all 6k Catalyst chassis should probably not ‘alarm’ by default when it’s not in use. There will always be only one of the two clocks in use, so some kind of logic tweak is needed to treat this differently otherwise you always have a load of status warnings for the category of “State” under “Health”. Unless you go through them all and disable them, but then you won’t get an event when they do become active (which is usually a bad thing as something else has failed). I’ve seen it show up as these two names depending what IOS it’s running: OSR-7600 Clock FRU 2 - clock 2 clock-inuse Sensor<http://observium.cdc.local/graphs/id=579946/type=sensor_state> OSR-7600 Clock FRU 2 In Using Sensor - Clock 2 In Using Sensor<http://observium.cdc.local/graphs/id=580144/type=sensor_state> Actually, since there is also a value for the ‘ok-ness’ of the sensors it is probably worth totally skipping the ‘in-use’ element for the default alarming purposes, imho. If you need more info from me then let me know, cheers! Robert Williams Custodian Data Centre Email: Robert(a)CustodianDC.com http://www.CustodianDC.com

1 0

How best to tell Observium that VMware is no longer unix [SEC=UNCLASSIFIED]
by Peter.Hine＠familycourt.gov.au 11 Apr '14

11 Apr '14

Back in the bad old days, esx was run on a version of RedHat Linux (RHEL). Now they run their own kernel. During discovery, i noticed that the unix agent was trying to be contacted and also has (unix). Reasonably sure VMware won't support it. Your native VM detection and polling is sufficient anyway. adlvm03 6 vmware (unix) Module [ junose-atm-vp ] excluded for device. Module [ cisco-ipsec-flow-monitor ] excluded for device. Module [ cisco-remote-access-monitor ] excluded for device. Module [ cisco-cef ] excluded for device. Module [ cisco-sla ] excluded for device. Module [ arista-software-ip-forwarding ] excluded for device. Module [ cipsec-tunnels ] excluded for device. Module [ cisco-ace-loadbalancer ] excluded for device. Module [ cisco-ace-serverfarms ] excluded for device. Module [ cisco-cbqos ] excluded for device. Module [ cisco-eigrp ] excluded for device. Module [ netscaler-vsvr ] excluded for device. Module [ aruba-controller ] excluded for device. Module [ cisco-vpdn ] disabled globally. Module [ wmi ] disabled globally. Observium UNIX Agent: Connection to UNIX agent on tcp://adlvm03:36602 failed. ERROR: 110 Connection timed out Connection to UNIX agent on tcp://adlvm03:6556 failed. ERROR: 110 Connection timed out Sensors: Module time: 20.0376s Uptime: 129 days, 2h 18m 13s Module time: 0.308s VMware VM: thanks Peter Hine Senior Technical Support Engineer (Servers) FCoA ITS peter.hine(a)familycourt.gov.au ********************************************************************** The information contained in this e-mail (including any attachments) is for the exclusive use of the addressee. If you are not the intended recipient please notify the sender immediately and delete this e-mail. It is noted that legal privilege is not waived because you have read this e-mail. **********************************************************************

2 1

Frontpage alert hostnames
by Spencer Gaw 11 Apr '14

11 Apr '14

Hey all, Has anyone noticed that some devices display the subdomain in their hostname on frontpage alerts and others do not? Example: Here are the hostnames configured on each device: /ring-rmcit.ftc> system show host-name// // //+------------------------------ HOST NAME -------------------------------+// //| Oper | ring-rmcit.ftc |// //| User | ring-rmcit.ftc |// //| DHCP | |// //+------+-----------------------------------------------------------------+// // //sw01.cos> system show host-name// // //+------------------------------ HOST NAME -------------------------------+// //| Oper | sw01.cos |// //| User | sw01.cos |// //| DHCP | |// //+------+-----------------------------------------------------------------+/ /redacted(a)access03.den> show version | match hostname// //Hostname: access03.den /Any ideas here? / /Regards, SG

4 4

Enhancement Request: Latest Reported Operating System
by techuser6＠gmail.com 11 Apr '14

11 Apr '14

It would be nice if there was another column next to "Operating System" for Network devices that had the latest reported IOS from Cisco/Dell. That way one could easily see which devices need to be upgraded to the latest IOS to ensure that security and bug fixes are addressed.

4 4