- observium - lists.observium.org

Frontpage Map Zoom
by Johnson, Edward A. 22 Apr '14

22 Apr '14

All of our devices are located within the state of Indiana. On the frontpage the map zooms from Utah to Nova Scotia with Indiana in the middle. Of course I can manually zoom in as far as I need but would like the default to zoom 1x more so I can see the entire state of indiana from north to south. Is there a setting in config.php that will let me do that?

2 2

Re: [Observium] Port 'Last Changed' visibility
by Jake Turner 22 Apr '14

22 Apr '14

After much debugging, tweaking the code and playing around with timestamp to epoch functions I’ve finally managed to get this working and it seems to be pretty accurate (for the Cisco devices at least). Thanks for your help Jake Turner [cid:94259173-8184-4F71-8163-56004829AE3C@turnerhouse.local] I am running 0.13.10 (rev. 4585) so yes, a pretty old version. I will update asap and that should shed a little more light (Y). Cheers Jake Turner On 19 Apr 2014, at 05:35, Adam Armstrong <adama(a)memetic.org<mailto:adama@memetic.org>> wrote: Are you running some horrifically old version? root@alpha:/home/observium/dev# wc -l includes/polling/ports.inc.php 719 includes/polling/ports.inc.php adam. On 2014-04-18 23:32, Jake Turner wrote: Weird, my /includes/polling/ports.inc.php file only has 617 lines. Cisco devices at least report the time that the system was up when the interface changed and I don’t mind having to calculate him many days/hours/minutes etc that was and then displaying that on the appropriate page. I see what you’re saying when it comes to providing the most compatibility with all the different vendors as they probably report it slightly different (or not at all) and so updating it with 'on update CURRENT_TIMESTAMP’ ensures that you’ll always have something there. I only have Cisco devices though so it’s not too much of a problem. Thanks, I’ll do some more digging tomorrow, need to sleep now! Jake Turner On 19 Apr 2014, at 05:10, Adam Armstrong <adama(a)memetic.org<mailto:adama@memetic.org>> wrote: Line 623 and below of includes/polling/ports.inc.php. This should use the port's ifLastChanged if it exists, else it'll use the current polling time. At some point we decided it was totally unrealiable and decided to track it ourselves by having the database update the value when the port configuration in the database changed. It sounds more like you'd want the "down since" number calculation changed to include the total time that the alert wasn't up (or unknown). adam. On 2014-04-18 22:58, Jake Turner wrote: Thank you for your reply. I take it from your response that there is some code that will update the ifLastChange field. Unfortunately I can not find it anywhere and as a desperate attempt to see if it were specified anywhere I performed a very dirty search: 'grep -r "ifLastChange" /etc/observium/‘ which yielded nothing of any significance. Sorry if you feel I’m being retarded but I simply can’t find it :/ Jake Turner On 19 Apr 2014, at 03:58, Adam Armstrong <adama(a)memetic.org<mailto:adama@memetic.org>> wrote: On 2014-04-18 21:34, Jake Turner wrote: Hmm, not sure what documentation is in the code, perhaps I was looking at the wrong files. It would appear however that the ifLastChange field is only changed whenever that particular record gets ‘updated’. The MySQL column has 'on update CURRENT_TIMESTAMP’ as it’s attribute. I’m thinking of removing that and creating a separate script to update this column with the interface change time as reported by the device. If you're not capable of finding where it's updated in the code, you possibly don't want to be modifying the code you can't be bothered to read. adam. _______________________________________________ observium mailing list observium(a)observium.org<mailto:observium@observium.org> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium _______________________________________________ observium mailing list observium(a)observium.org<mailto:observium@observium.org> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium _______________________________________________ observium mailing list observium(a)observium.org<mailto:observium@observium.org> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium _______________________________________________ observium mailing list observium(a)observium.org<mailto:observium@observium.org> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium _______________________________________________ observium mailing list observium(a)observium.org<mailto:observium@observium.org> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium _______________________________________________ observium mailing list observium(a)observium.org<mailto:observium@observium.org> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

1 0

graphs for power usage on Tripplite PDUMH30ATNET
by Jeff Courtade 22 Apr '14

22 Apr '14

Hello, I have successfully added in a Tripplite PDUMH30ATNET. It successfully detects and uses the OPS mib when it sets up and has a nice little tripplite icon. The only thing it seems to be graphing is the Ethernet traffic on the one ethernet interface on the device. What I want to see graphs of are the power usage on each of the power outlets and inputs. Any help would be appreciated. -- Thanks, Jeff Courtade Director of Systems Engineering Adjuggler Inc http://www.adjuggler.com M: 240.507.6116

4 5

Re: [Observium] Ubiquiti AirOS MIB (Robbie Wright)
by Josh Reynolds 22 Apr '14

22 Apr '14

Robbie, Thanks for staying on top of this. We've been trying to scare up support on the UBNT forum for about 6 months now to get a large group of people to donate to the Observium project, but until it works... it's hard to get people to pay for it. *Josh Reynolds* Chief Information Officer SPITwSPOTS josh(a)spitwspots.com | www.spitwspots.com

5 6

Re: [Observium] Ubiquiti AirOS MIB (Robbie Wright) (Moore, Cameron)
by Josh Reynolds 22 Apr '14

22 Apr '14

The new MIB will only work on 5.6-beta and above. Or AirFiber v2.0+ -- Josh Reynolds : Chief Information Officer : SPITwSPOTS : josh(a)spitwspots.com On Mon, 2014-04-21 at 23:58 +0100, observium-request(a)observium.org wrote: > Re: Ubiquiti AirOS MIB (Robbie Wright) (Moore, Cameron)

1 0

Trying to update Commnuity edition...
by Boardman, Nick 21 Apr '14

21 Apr '14

Good afternoon, I am trying to update my Community version .4585 to the latest version listed on the Observium page of .5229 but it fails. I have installed the Subversion client using apt-get. The specific error message when attempting to upgrade is: Skipped '.' Any help would be appreciated. Nick Boardman MCSE, MCSA: Security, CCA, Security+ HRSD Systems Engineer

3 4

Map issue
by Johnathon Voegeli 19 Apr '14

19 Apr '14

Does anyone know how to fix this? All the SysLocation fields are identical, but when they go on the map it places 2 blips a few miles a part. This is on all of my locations. I suspect its an easy fix to blow that part of the database away and let it reload them. I believe it started happening when the map updated the last time. Using the subscription. ~JV [cid:36A15709-7D97-40E0-AEAC-EFDF91B3CCBC] CONFIDENTIALITY NOTICE: This e-mail message, and any documents, files, or previous e-mail messages attached to it, may contain information that is confidential, proprietary, and/or legally privileged. If the reader of this e-mail message is neither the intended recipient nor an employee or an agent responsible for delivering this e-mail message to the intended recipient, you are hereby notified that any unauthorized review, use, disclosure, reproduction, distribution or the taking of action in reliance on the contents of the information of this communication is prohibited. If you have received this communication in error, please notify the sender by replying to the e-mail message and delete it from your computer, without reading it or saving it in any manner.

1 0

Port 'Last Changed' visibility
by Jake Turner 19 Apr '14

19 Apr '14

Hi all, I can't seem to see this feature on the port details page or the 'all ports' list but forgive me if it is visible somewhere. It would be useful to have a 'last changed' time (30 mins ago) etc... on the port details page. This info can be gained from SNMP and from what I can see is already in the DB so it would just be a case of adding it to the page code. Thanks Jake

3 12

Re: [Observium] Observium] Brocade MLXe and ICX support / question
by Adam Armstrong 18 Apr '14

18 Apr '14

No such object is irrelevant, it is from another polled module. I'm not sure why it's discovering incorrectly, I will try to look tomorrow. Adam. "Swen Veckes (KDDI)" <s.veckes(a)de.kddi.com> wrote: OK, I’m not a programmer, so I can’t proof. But I have deleted the device „br00“ from observium and added it again. Now the result is the same as for the other device „mlx01“. So observium changed the way it calculates processor load. But why should this been changed? Any of the developers who can help me to understand, especially on the „No Such Object …“ messages which makes me think that Broke support does not work correctly for MLXe devices? Regards, Swen Am 17.04.2014 um 09:14 schrieb Swen Veckes (KDDI) <s.veckes(a)de.kddi.com>: > OK then. I have tried to further investigate my problem and run poller.php and discover.php in debug mode and got the following result for the processors: > > SQL[SELECT `processors`.*, `processors-state`.processor_usage, `processors-state`.processor_polled FROM `processors` LEFT JOIN `processors-state` ON `processors`.processor_id = `processors-state`.processor_id WHERE `device_id` = '88'] > Processor Slot 1 BR-MLX-10Gx8-X [1] DEBUG: SNMP Auth options = -v2c -c '****' > /usr/bin/snmpget -v2c -c '****' -O Uqnv -m "" -M /opt/observium/mibs 'udp':'ffm-th-mlx01':'161' .1.3.6.1.4.1.1991.1.1.2.11.1.1.6.1.1.300 > 100 > > 100% > RRD[cmd[update /opt/observium/rrd/ffm-th-mlx01/processor-ironware-1.1.300.rrd N:100] > stdout[OK u:0.00 s:0.00 r:1.27] > stderr[]] > > SQL[UPDATE `processors-state` set `processor_usage` ='100',`processor_polled` ='1397716569' WHERE `processor_id` = '480'] > > > > SQL[SELECT `processors`.*, `processors-state`.processor_usage, `processors-state`.processor_polled FROM `processors` LEFT JOIN `processors-state` ON `processors`.processor_id = `processors-state`.processor_id WHERE `device_id` = '49'] > Processor Slot 1 BR-MLX-10Gx8-X [1] DEBUG: SNMP Auth options = -v2c -c '****' > /usr/bin/snmpget -v2c -c '****' -O Uqnv -m "" -M /opt/observium/mibs 'udp':'br00':'161' .1.3.6.1.4.1.1991.1.1.2.11.1.1.6.1.1.300 > 100 > > 1% > RRD[cmd[update /opt/observium/rrd/br00/processor-ironware-1.1.300.rrd N:1] > stdout[OK u:0.00 s:0.00 r:0.90] > stderr[]] > > SQL[UPDATE `processors-state` set `processor_usage` ='1',`processor_polled` ='1397716171' WHERE `processor_id` = '160'] > > > > > DEBUG: SNMP Auth options = -v2c -c '****' > /usr/bin/snmpget -v2c -c '****' -Oqv -m FOUNDRY-SN-AGENT-MIB -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp:/opt/observium/mibs/foundry 'udp':'ffm-th-mlx01':'161' snAgentConfigModuleDescription.1 > "BR-MLX-10Gx8-X 8-port 10GbE (X) Module" > > 88 -> .1.3.6.1.4.1.1991.1.1.2.11.1.1.6.1.1.300, 1.1.300, ironware, Slot 1 BR-MLX-10Gx8-X [1], 1, 100, , > > SQL[SELECT * FROM `processors` WHERE `device_id` = '88' AND `processor_index` = '1.1.300' AND `processor_type` = 'ironware'] > . hrDevice: DEBUG: SNMP Auth options = -v2c -c '****' > /usr/bin/snmpbulkwalk -v2c -c '****' -OQUs -m HOST-RESOURCES-MIB:HOST-RESOURCES-TYPES -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp 'udp':'ffm-th-mlx01':'161' hrDevice > hrDevice = No Such Object available on this agent at this OID > > DEBUG: SNMP Auth options = -v2c -c '****' > /usr/bin/snmpbulkwalk -v2c -c '****' -OQUs -m HOST-RESOURCES-MIB:HOST-RESOURCES-TYPES -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp 'udp':'ffm-th-mlx01':'161' hrProcessorLoad > hrProcessorLoad = No Such Object available on this agent at this OID > > > > DEBUG: SNMP Auth options = -v2c -c '****' > /usr/bin/snmpget -v2c -c '****' -Oqv -m FOUNDRY-SN-AGENT-MIB -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp:/opt/observium/mibs/foundry 'udp':'br00':'161' snAgentConfigModuleDescription.1 > "BR-MLX-10Gx8-X 8-port 10GbE (X) Module" > > 49 -> .1.3.6.1.4.1.1991.1.1.2.11.1.1.6.1.1.300, 1.1.300, ironware, Slot 1 BR-MLX-10Gx8-X [1], 1, 100, , > > SQL[SELECT * FROM `processors` WHERE `device_id` = '49' AND `processor_index` = '1.1.300' AND `processor_type` = 'ironware'] > . hrDevice: DEBUG: SNMP Auth options = -v2c -c '****' > /usr/bin/snmpbulkwalk -v2c -c '****' -OQUs -m HOST-RESOURCES-MIB:HOST-RESOURCES-TYPES -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp 'udp':'br00':'161' hrDevice > hrDevice = No Such Object available on this agent at this OID > > DEBUG: SNMP Auth options = -v2c -c '****' > /usr/bin/snmpbulkwalk -v2c -c '****' -OQUs -m HOST-RESOURCES-MIB:HOST-RESOURCES-TYPES -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp 'udp':'br00':'161' hrProcessorLoad > hrProcessorLoad = No Such Object available on this agent at this OID > > > > Even the result of CPU load is the same for both devices, one is inserted to the database as 1% the other as 100%. Both queried OID’s are the same. Not sure about the „No such Object ….“ but that may be the problem!? > > > Regards > Swen > > > > > > > Am 16.04.2014 um 17:34 schrieb Raphael Maunier <raphael.maunier(a)gmail.com>: > >> Not sure, >> >> I will upgrade this observium later this month, but for now, I can’t :) >> >> I’ll let you know when this will be done, and I’ll update here >> >> Raphael >> >> On 16 Apr 2014, at 17:13, Swen Veckes <s.veckes(a)de.kddi.com> wrote: >> >>> Hi Raphael, >>> >>> thanks a lot for your answer. I tried downgrading to 5.4e and memory seems ok now, however, CPU is still 100%. Observium version is at v0.14.4.5323, as you are running a much older version that might make a difference!? >>> >>> Regards, >>> Swen >>> >>> >>> ----- Ursprüngliche Mail ----- >>> Von: "Raphael Maunier" <raphael.maunier(a)gmail.com> >>> An: "Observium Network Observation System" <observium(a)observium.org> >>> Gesendet: Dienstag, 15. April 2014 22:43:12 >>> Betreff: Re: [Observium] Brocade MLXe and ICX support / question >>> >>> >>> Hello Swen, >>> >>> >>> I have a couple MLX-8 MLX-16 running 5.4E and I don’t have this issue for Memory / cpu. >>> For your vlan issue, I can’t check it as our mlx are a full vpls fabric, so all the ports are seen in vlan1 ( even if the vpls vlan is 4 ) >>> >>> >>> There is plenty of bugs on Broke -aid , the most stable version we have in production is this one. >>> >>> >>> 5.4B is bugued on the lag( ask brocade about this internal ID : DEFECT000440912 ) , so the minimum version you have to use is 5.4C >>> >>> >>> Regarding the ICX, I tried them and … I removed them and I will forget them :) >>> >>> >>> Attached a capture of one MLX16. >>> >>> >>> This customer is running CE 0.13.10.4585 >>> >>> >>> Raphael >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> On 15 Apr 2014, at 14:28, Swen Veckes (KDDI) < s.veckes(a)de.kddi.com > wrote: >>> >>> >>> >>> Does no one have a glue on this? >>> >>> >>> cheers >>> Swen >>> >>> >>> >>> Am 11.04.2014 um 11:14 schrieb Swen Veckes (KDDI) < s.veckes(a)de.kddi.com >: >>> >>> >>> >>> Hi, >>> >>> we use a couple of MLXe’s and ICX’s which seem fine at the first place running observium’s subscriber edition. However, a few part seems not working correctly: >>> >>> + MLXE’s >>> - Memory shown as -1 on the device main page and has no graphs „nan“ at all >>> - Memorypool usage graphs show -nan >>> - Observium recognizes the installed Line card CPU but report 100% CPU on another machine adder 3 month earlier it reports OK >>> - VLAN’s are reported fine, however, on the VLAN tab only untagged vlan1 (default vlan) has port associated with it, others are empt >>> >>> + ICX’s >>> - CPU reported 100% all time, but same switch installed earlier shows fine >>> - Compared to the MLXe it does not show LAG interfaces >>> >>> >>> Any hints on the the graphing which works for one device but another added later does not as well for the memory? I Don’t hink that observium suddenly broke the support for this, as it is still working for the other devices (exactly the same device). >>> >>> What do I need to provide to help fixing the other things such as the LAG display for the ICX or others and to get more current Brocade device support? >>> >>> >>> Cheers, >>> Swen >>> _______________________________________________ >>> observium mailing list >>> observium(a)observium.org >>> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium >>> >>> >>> _______________________________________________ >>> observium mailing list >>> observium(a)observium.org >>> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium >>> <observium_mlx8.PNG>_______________________________________________ >>> observium mailing list >>> observium(a)observium.org >>> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium >> >> _______________________________________________ >> observium mailing list >> observium(a)observium.org >> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium > _______________________________________________ observium mailing list observium(a)observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

2 1

Interface pop up bubble "up"
by David Peall 18 Apr '14

18 Apr '14

Hi I’ve got a problem on the interface where the pop up blocks the mouse pointer from clicking on the interface link, i.e. the mouse pointer is inside the bubble. For interfaces at the bottom of the page when the bubble goes up, down ones are fine. Regards — David Peall

4 3

graph NVIDIA GPU parameters
by Mihai - Cristian Satmarean 17 Apr '14

17 Apr '14

Dear all, sorry if this was already addressed, is it somehow possible to graph the Nvidia GPU parameters in observium? Load, memory, temp etc. Thank you! Mihai

3 6

1 rogue host on map in the mid Atlantic
by Ryan Prihoda 17 Apr '14

17 Apr '14

Hey all, I recently updated to the new community edition and once everything was up again, I have 1 rogue host showing in the mid Atlantic on the map. None of my hosts use the override "syslocation" function. The number of devices in "All Locations" is equal to devices my syslocation. Here's the kicker, when I click on the rogue host, it takes me to my devices page for my syslocation. Is there anything I can do to figure out what device is causing this issue ? Thanks, R. Prihoda

2 4

IP address weirdness
by Sander Steffann 16 Apr '14

16 Apr '14

Hi, This morning at 6:30 the IPv4 addresses of my servers suddenly started changing. I see a lot of messages like: IPv4 changed: 83.247.10.20/27 -> 83.247.10.20/3 IPv4 changed: 83.247.10.21/27 -> 83.247.10.21/3 IPv4 changed: 83.247.10.22/27 -> 83.247.10.22/3 IPv4 changed: 94.142.242.194/29 -> 94.142.242.194/5 IPv4 changed: 94.142.242.195/29 -> 94.142.242.195/5 IPv4 changed: 94.142.242.213/29 -> 94.142.242.213/5 IPv4 changed: 94.142.242.214/28 -> 94.142.242.214/4 IPv4 changed: 94.142.242.215/28 -> 94.142.242.215/4 IPv4 removed: 127.0.0.1/8 Besides showing me that I have some misconfigured subnet masks in the 94.142.242.208/28 subnet, it also seems to have some consistency: - A /27 turns into a /3 - A /28 turns into a /4 - A /29 turns into a /5 My VMware ESXi machines also suddenly show: IPv4 added: 83.247.10.11/32 IPv4 added: 83.247.10.12/32 IPv4 added: 83.247.10.13/32 I also noticed some broadcast addresses being added: IPv4 added: 83.247.10.31/27 (this is an Ubuntu 14.04 server with 83.247.10.29/27) IPv4 added: 83.247.10.31/27 (this is is a CentOS 5.10 server with 83.247.10.7/27) IPv4 added: 83.247.10.47/28 (this is a vcenter applicance) IPv4 added: 83.247.10.47/28 (this is a Dell DRAC card) IPv4 added: 127.255.255.255/8 (this is a vcenter applicance) So, subnet sizes went weird on some systems, 127.0.0.1 disappeared from some systems, some ethernet adapters now show both the real IP address and the broadcast address on some systems. But different mixes of results across different platforms. Might this be related to r5317? Cheers, Sander

2 2

dot1dbaseportifindex on Cisco not listing UP interfaces?
by Ulrik Ivers 16 Apr '14

16 Apr '14

Hi, I see a strange behavior on our Cisco-switches (different models) when listing the interfaces via SNMP using the mib BRIDGE-MIB. When walking dot1dbaseportifindex it excludes interfaces that are UP. Only interfaces that are DOWN or DISABLED are returned. Is this something that is well known? Seems to me that is a bug in ios... It breaks neighbor discovery via LLDP (lldp-mib.inc.php). Examples: !!!!!!!!!!!!!!!!!!!!!!!!!!! Cisco Switch - ports 1-7 are UP, the rest DOWN /usr/bin/snmpbulkwalk -v2c -c 'XXXX' -OQUs -m BRIDGE-MIB -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp 'udp':'test-host' dot1dBasePortIfIndex dot1dBasePortIfIndex.8 = 10108 dot1dBasePortIfIndex.9 = 10109 dot1dBasePortIfIndex.10 = 10110 dot1dBasePortIfIndex.11 = 10111 dot1dBasePortIfIndex.12 = 10112 dot1dBasePortIfIndex.13 = 10113 dot1dBasePortIfIndex.14 = 10114 dot1dBasePortIfIndex.15 = 10115 dot1dBasePortIfIndex.16 = 10116 dot1dBasePortIfIndex.17 = 10117 dot1dBasePortIfIndex.18 = 10118 dot1dBasePortIfIndex.19 = 10119 dot1dBasePortIfIndex.20 = 10120 dot1dBasePortIfIndex.21 = 10121 dot1dBasePortIfIndex.22 = 10122 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!! HP 10 port switch , all interfaces are always listed /usr/bin/snmpbulkwalk -v2c -c 'XXXX' -OQUs -m BRIDGE-MIB -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp 'udp':'test-host' dot1dBasePortIfIndex dot1dBasePortIfIndex.1 = 1 dot1dBasePortIfIndex.2 = 2 dot1dBasePortIfIndex.3 = 3 dot1dBasePortIfIndex.4 = 4 dot1dBasePortIfIndex.5 = 5 dot1dBasePortIfIndex.6 = 6 dot1dBasePortIfIndex.7 = 7 dot1dBasePortIfIndex.8 = 8 dot1dBasePortIfIndex.9 = 9 dot1dBasePortIfIndex.10 = 10 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Regards, /Ulrik Ivers Ulrik Ivers -- Excanto AB Sveavägen 63 | 113 59 Stockholm Mobil: 0709-90 95 06 Växel: 08-545 726 00 Web: www.excanto.se<http://www.excanto.se/> [cid:image001.jpg@01CF595E.6CC06620]

1 0

PPP links for Mikrotik
by David Peall 15 Apr '14

15 Apr '14

Hi The link names for ppp interfaces on the mikrotik routers are wrapped in <> i.e. <pptp-david> this seems to be causing some issues on the website. For example: <span class="entity-title"> <a href="device/device=28/tab=port/port=1462/" class="entity-popup " data-eid="1462" data-etype="port" data-hasqtip="1"><pptp-david></pptp-david></a> </span> Regards — David Peall

2 3

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