- observium - lists.observium.org

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