- observium - lists.observium.org

Re: [Observium] SNMP support for blades' internal components (CPU,
by Youssef BENGELLOUN - ZAHR 27 Feb '17

27 Feb '17

Dear Michael, SNMP is perfectly responsive on those hosts : root@prod-monit-002:/opt/observium# snmpwalk -v 2c -c asp par01-esx2-bl2 sysDescr SNMPv2-MIB::sysDescr.0 = STRING: VMware ESXi 5.5.0 build-3568722 VMware, Inc. x86_64 I have attached a full snmwalk output from one of the host as an example. Best regards. Le 27/02/2017 16:19, « observium au nom de Michael » <observium-bounces(a)observium.org au nom de obslist(a)smarsz.com> a écrit : Looks to me like there's no data at all coming back from either of those esxi hosts. "WARNING: 0 devices discovered" Could you try the following query to see what comes back snmpwalk -v3 <your-snmpv3-details-here> par01-esx02-bl2 sysDescr I think you need to validate that your server is actually answering SNMP queries for you. VMware have long been reluctant to do much with SNMP (prefering to push people towards vCenter/vCOPS). As a result, their support for revealing lots of physical host information leaves a bit to be desired. Cheers, Michael > On 27 Feb 2017, at 9:35 pm, Youssef BENGELLOUN - ZAHR <ybzahr(a)prodware.fr> wrote: > > > Dear Observium, > > This is a request for information / enhancement, I hope you will find it relevant / useful. Actually running latest stable version 17.2.8364. > > We run quite a lot of blade chassis (DELL M1000e, IBM bladecenterH, IBM pureflex) and monitor the blades in the enclosures, more specifically the ESXi hosts servers. > > I recently noticed that we were not able to recover SNMP information about physical components’ status (CPU, memory DIMMs, etc.). We are more interested in memory DIMM status. > > I have run a discovery against a couple of hosts in debug mode and can see that recovered SNMP information is pretty basic (see attached). > > Do you think it would be possible / interesting to recover such information ? > > Thank you for your feedback. > > Best regards. > > > Youssef BENGELLOUN - ZAHR - Consultant Expert > Prodware France > T : +33 979 999 000 - F : +33 988 814 001 - ybzahr(a)prodware.fr > <image23bec5.PNG>Web : prodware.fr > > <image9b26f1.JPG> <image5c3e4c.JPG> <image0119e3.JPG> <imagee569c0.JPG> <image93eaa0.JPG> <image039eb0.JPG> > > > > <par01-esx02-bl2.txt><par01-esx03-bl2.txt> BENGELLOUN - ZAHR Youssef - Consultant Expert Prodware France T : +33 979 999 000 F : +33 988 814 001 - ybzahr(a)prodware.fr Web : prodware.fr _______________________________________________ > 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 3

No graph policy description on Cisco CBQoS graph
by Sophanith Chhom 27 Feb '17

27 Feb '17

Dear All, I wonder why Cisco CBQoS graph on Observium Pro have no graph description(policy name) show up? Since I observe there is policy description name showing up on Observium Community Edition. It should display something like: CUST-Name / class-default (input) CUST-Name / class-default (output) Can you guide if we can enable it on the script somewhere? Thanks so much.

2 3

Disable IF-MIB cisco nexus
by Svensson Fredrik A 24 Feb '17

24 Feb '17

Hi there! Due to a bug I want to disable the IF-MIB, on my nexus devices, is that possible? There is no specific IF-mib in the Device-MIB, is it included in some of the other MIBs? ADSL-LINE-MIB enabled  Disable BGP4-MIB enabled  Disable CISCO-AAA-SESSION-MIB enabled  Disable CISCO-BGP4-MIB enabled  Disable CISCO-CDP-MIB enabled  Disable CISCO-CEF-MIB enabled  Disable CISCO-EIGRP-MIB enabled  Disable CISCO-ENHANCED-MEMPOOL-MIB enabled  Disable CISCO-ENTITY-FRU-CONTROL-MIB enabled  Disable CISCO-ENTITY-PFE-MIB enabled  Disable CISCO-ENTITY-QFP-MIB enabled  Disable CISCO-ENTITY-SENSOR-MIB enabled  Disable CISCO-ENTITY-VENDORTYPE-OID-MIB enabled  Disable CISCO-ENVMON-MIB enabled  Disable CISCO-FLASH-MIB enabled  Disable CISCO-IETF-IP-MIB enabled  Disable CISCO-IETF-PW-MIB enabled  Disable CISCO-IP-STAT-MIB enabled  Disable CISCO-IPSEC-FLOW-MONITOR-MIB enabled  Disable CISCO-MEMORY-POOL-MIB enabled  Disable CISCO-POWER-ETHERNET-EXT-MIB enabled  Disable CISCO-PROCESS-MIB enabled  Disable CISCO-REMOTE-ACCESS-MONITOR-MIB enabled  Disable CISCO-RTTMON-MIB enabled  Disable CISCO-TRUSTSEC-INTERFACE-MIB enabled  Disable CISCO-VPDN-MGMT-MIB enabled  Disable CISCO-VTP-MIB enabled  Disable ENTITY-MIB enabled  Disable ENTITY-SENSOR-MIB enabled  Disable EtherLike-MIB enabled  Disable HOST-RESOURCES-MIB enabled  Disable IP-MIB enabled  Disable LLDP-MIB enabled  Disable MPLS-L3VPN-STD-MIB enabled  Disable MPLS-VPN-MIB enabled  Disable OSPF-MIB enabled  Disable POWER-ETHERNET-MIB enabled  Disable Q-BRIDGE-MIB enabled  Disable SMON-MIB enabled  Disable SNMP-FRAMEWORK-MIB enabled  Disable SNMPv2-MIB enabled  Disable UCD-SNMP-MIB enabled  Disable

2 1

Can't login to Subscription Management
by Karolis Birgėla 24 Feb '17

24 Feb '17

Hello, as our staff is changing over time, we lost the credentials to the client area. Is there a possible way to get the info back? We only know the email.. Thank you. UAB "Kauno Interneto Sistemos", Lithuania

1 0

Observium 1min poller interval?
by Sophanith Chhom 23 Feb '17

23 Feb '17

Dear All, Is it possible to schedule 1min for poller interval? By default it is 5min interval. Thanks

5 5

Graph direct link changes when group membership changes
by Jesper Frank Nemholt 23 Feb '17

23 Feb '17

Hi! I'm using the direct link to graphs for a dashboard. Not sure this was the intended usage, but I found an issue that I think could be avoided related to this. It looks like graphs in this view are getting encoded IDs that somehow relate to the devices in the group, so the URL looks like this below. The issue is that when having lots of devices in a view, the id field gets really long and eventually will cause the graph creation to fail....however another and bigger issue is that it seems the id actually change when new devices are added (which happens during auto-discovery). Is there any way to fix the id to a numeric number like used with other graphs ? http://observium/graph.php?to=1487801959&id=WyIzMjE5MSIsIjMyNTE1IiwiMzMxNjM… /Jesper

3 2

Re: [Observium] Can't write new info to Observium
by Karolis Birgėla 22 Feb '17

22 Feb '17

On 2017.02.22 14:00, observium-request(a)observium.org wrote: > Can't write new info to Observium Also, another problem we have noticed, I wonder if it's related to the caching update. When using device search (Devices->All Devices), then selecting hardware type (for example DES-1210-26/ME), we can find 8 devices from our network, that's ok. But we have updated the firmware version of this specific device, and when trying to do the same search in Observium, It can't find the device with updated firmware. And this happened with every device we upgraded. Quick search bar finds the device without any problems. Could this be related to the caching update? Thank you

1 0

all VMs Removed - why ?
by Thoralf Freitag 22 Feb '17

22 Feb '17

Hi, is there any idea, why observium has removed all VMs from the ESXi Server, like this: 1d 2h 3m Vm Virtual Machine deleted: vmware 564da96d-e3b0-b511-66cc-2a184a576652 1d 2h 3m Vm Virtual Machine deleted: vmware 564d8984-0864-6b78-7d91-36f2c13e53d6 1d 2h 3m Vm Virtual Machine deleted: vmware 564d4905-87d7-6fdf-bc72-753c916ef3c1 1d 2h 3m Vm Virtual Machine deleted: vmware 564d2060-e2e4-9949-97a6-28cb5f02e801 1d 2h 3m Vm Virtual Machine deleted: vmware 564d3421-27c8-04ea-7ef8-139af5ebcb5c 1d 2h 3m Vm Virtual Machine deleted: vmware 564d18eb-bad5-0691-7acf-3cb24787c0c1 1d 2h 3m Vm Virtual Machine deleted: vmware 564d523c-b392-e037-12e6-4c93b9dc82b1 1d 2h 3m Vm Virtual Machine deleted: vmware 564dc378-b321-2b72-176f-09c8fc5a1116 1d 2h 3m Vm Virtual Machine deleted: vmware 564db385-e361-5d0a-3167-d9a416357692 ... The tab for virtual machines no longer appears ... . There was no change at the ESXi. VMWARE-SYSTEM-MIB and VMWARE-VMINFO-MIB are enabled. I restarted the snmp agant at the ESXi host, run poller.php again , but it changes nothing. Thanks for any suggestions. lg. TF

2 4

Can't write new info to Observium
by Karolis Birgėla 22 Feb '17

22 Feb '17

Hello all, never happened to me before, but Obsevium acts strangely lately for me. When trying to add a new device via web, it sucessfully adds it, but it's page appears to be missing. It says something like 'You entered a device which does not exist' or something similar. What is more, when I want to manually change device's description on the web, it says 'Device record updated'. But the info that I've written has not appeared. What could have caused this? I don't remember doing any changes to my Observium server, except doing an update. Version 17.2.8364 (stable) Everything else works as it should. Thank you

2 1

ASR9k RAM Changes
by Robert Williams 22 Feb '17

22 Feb '17

Hi, Have there just been some changes to the ASR9k RAM detection? Suddenly after the overnight svn up we are seeing this for all 9k chassis: [cid:image001.png@01D28C23.56E5F220] These 3 new items are in alarm. The “reserved” and “image” are filtered from all other devices by existing config.php – but obviously no brackets on these ones. $config['ignore_mempool_string'][] = '(reserved)'; $config['ignore_mempool_string'][] = '(image)'; However, the ‘processor’ one is actually wrong as it seems to be using the CPU0 usage but a strange maximum of 3.81GB instead of 7.81GB. I can’t filter this out because it will drop all ‘processor’ types. I remember something historic relating to this where the maximums were showing as around 4GB even though there is 8GB installed and someone (Adam?) found and sorted it. This may or may not be related though. Cheers! Robert Williams Custodian Data Centre Email: Robert(a)CustodianDC.com http://www.CustodianDC.com

4 5

Alert regeneration via cli?
by Michael 22 Feb '17

22 Feb '17

Is there a way to regenerate the alerts from the cli? We have some checks that alert based on port state, dependent on conditions within a port description. But that matching is only performed during the alert regeneration, not during polling. So if the port description gets updated, the alert still triggers based on the old condition match... I'm thinking of running a regen via cron on a regular basis to try and minimise the false positives. Unless somebody has a better way to achieve a similar outcome? Thanks, Michael

2 2

Poller wrapper history graphs not updating
by Jesper Frank Nemholt 21 Feb '17

21 Feb '17

Hi! I have moved an instance from a physical server to a VM by moving over the database & RRD files + Observium $HOME. All works except one minor thing : The poller wrapper history (/pollerlog/ page) graphs do not update anymore. Also the RRD files do not update their timestamp since I moved the files to the new VM. The poller discovery/timing part of the same page does work properly, so it's only the top two graphs missing (or rather the graphs are there, but no new data is being added). Any idea why ? /Jesper

3 6

Memory graph missing
by Simon Mousey Smith 21 Feb '17

21 Feb '17

Hi All, I’ve got up this morning to find our nightly SVN UP has messed up the memory graphs for all our switches/servers/etc…? What appears to be be happening is its looking for the file 'mempool-HOST-RESOURCES-MIB-65536.rrd' which doesn’t exist BUT 'mempool-host-resources-mib-65536.rrd’ does exist Case-senstive I believe Anybody else had this issue? Regards Simon

2 2

Re: [Observium] ASR9k RAM Changes
by Moerman, Maarten 21 Feb '17

21 Feb '17

There’s currently work being done on redoing memory pools to ‘definitions’ – probably related. Maarten From: observium <observium-bounces(a)observium.org> on behalf of Robert Williams <Robert(a)CustodianDC.com> Reply-To: Observium Public Support <observium(a)observium.org> Date: Tuesday, February 21, 2017 at 10:20 AM To: "Observium Network Observation System (observium(a)observium.org)" <observium(a)observium.org> Subject: [Observium] ASR9k RAM Changes Hi, Have there just been some changes to the ASR9k RAM detection? Suddenly after the overnight svn up we are seeing this for all 9k chassis: [cid:image001.png@01D28C2D.94872EF0] These 3 new items are in alarm. The “reserved” and “image” are filtered from all other devices by existing config.php – but obviously no brackets on these ones. $config['ignore_mempool_string'][] = '(reserved)'; $config['ignore_mempool_string'][] = '(image)'; However, the ‘processor’ one is actually wrong as it seems to be using the CPU0 usage but a strange maximum of 3.81GB instead of 7.81GB. I can’t filter this out because it will drop all ‘processor’ types. I remember something historic relating to this where the maximums were showing as around 4GB even though there is 8GB installed and someone (Adam?) found and sorted it. This may or may not be related though. Cheers! Robert Williams Custodian Data Centre Email: Robert(a)CustodianDC.com http://www.CustodianDC.com

2 1

Specify snmp timeout for single device.
by Gabriel Zamorski 20 Feb '17

20 Feb '17

Hey, I would like to ask if it is possible to specify snmp timeout for a single device in Observium? I tried to do that in Device > Properties > SNMP > Timeout, but each time I try to change this empty field to 100 or 200… I get info: “Device SNMP configuration update is not required” Regards, Gabriel

2 1

Observium tunning
by chott＠praha1.net 19 Feb '17

19 Feb '17

Hello team what is best solution for tunning performance like different Hardrive for MySQL/RRD and System? What about performance SSD Raid or RAM Disk? We are using VMware, with iSCSI SAS and SSD , for System SAS drive for RRD/MySQL two drive from SSD Raid any idea? like HW Raid without iSCSI? Now we have more than 500 devices, 4000 ports and 3000 sensors Tom [email-header-praha1] Tomas Chott technický ředitel chott(a)praha1.net<mailto:chott@praha1.net> Phone: +420 245 004 005 Mobile: +420 733 123 135 Metropolitní síť Praha 1 z.s.p.o. Jindrisska 18 110 00 Praha 1 Czech Republic www.praha1.net<http://www.praha1.net> [Popis: Facebook]<http://www.facebook.com/praha1net> <http://www.linkedin.com/pub/tomas-chott/49/97a/1aa>

4 4

using hardware sensors
by Thoralf Freitag 19 Feb '17

19 Feb '17

Dear Observium community, is there any description of using hardware sensors? I've following issue. I'm operating a HP server with an Ilom under Ubuntu Linux. I configured in the device properties for ipmi: IPMI Hostname (IP address of the Ilom), IPMI Username (as configured at Ilom) IPMI Password, IPMI Userlevel (User) IPMI Interface (IPMI v2.0 RMCP+LAN Interface). Observium rekognized the sensors for temperature und fan speed. I did the same at an other HP server (same type) which is running ESXi 6.5 (HP customized Image). The Sensors are shown in the vSphre-Client and ESXi empeded host client UI. But for this host, Observium does not rekognize the sensors. How can this be explained ? Thanks. Lg. TF

2 2

Group with 500+ devices
by Karolis Birgėla 19 Feb '17

19 Feb '17

Hi, I have a problem with device classification to specific groups. We currently have 750 devices added in our network. Some users can only have access to specific devices, which led us to creating a group associated with them. The problem is, that more than 500 devices belong to the group, and when we attach this group to a user, it lags like hell. The main window of Observium loads slowly, and the search works very slow. I dont think that the server is incapable of proccesing this quickly It's a server with 80gb of ram, 24core cpu. By the way, admin accounts work perfectly . Is this a bug? Or something needs to be done in order? Let me know, thank you.

4 12

Poller coding scheme
by Youssef BENGELLOUN - ZAHR 19 Feb '17

19 Feb '17

3 4

Copy from Old to New server
by chott＠praha1.net 19 Feb '17

19 Feb '17

Hello everybody how we can copy all Observium data from OLD to NEW server? i means RRD probably [email-header-praha1] Tomas Chott technický ředitel chott(a)praha1.net<mailto:chott@praha1.net> Phone: +420 245 004 005 Mobile: +420 733 123 135 Metropolitní síť Praha 1 z.s.p.o. Jindrisska 18 110 00 Praha 1 Czech Republic www.praha1.net<http://www.praha1.net> [Popis: Facebook]<http://www.facebook.com/praha1net> <http://www.linkedin.com/pub/tomas-chott/49/97a/1aa>

2 1

SummitD support like Siklu
by chott＠praha1.net 18 Feb '17

18 Feb '17

Hello Mike there is support for Siklu device in r8371 what about support similar device Summit Development http://www.summitd.cz/cz/download/mib-file Tom [email-header-praha1] Tomas Chott technický ředitel chott(a)praha1.net<mailto:chott@praha1.net> Phone: +420 245 004 005 Mobile: +420 733 123 135 Metropolitní síť Praha 1 z.s.p.o. Jindrisska 18 110 00 Praha 1 Czech Republic www.praha1.net<http://www.praha1.net> [Popis: Facebook]<http://www.facebook.com/praha1net> <http://www.linkedin.com/pub/tomas-chott/49/97a/1aa>

1 0

Mail list Archives throw 404
by Ramiro Rojo 17 Feb '17

17 Feb '17

Greetings, Trying to acces to the Observium Archives (http://postman.memetic.org/pipermail/observium/) is not working. Are you looking for a solution for it, or is the page on the documentation outdated?

1 0

Ruckus Virtual SmartZone (vSZ) controller
by Philippe Laurent 17 Feb '17

17 Feb '17

It seems that either the Ruckus MIB changed, or my expectation of function is different than reality. We run the Ruckus Virtual SmartZone (vSZ) to control our AP's. At the moment, Observium provides just the local attributes (CPU, NIC, RAM, Disk) of the virtual machine, but it was my impression that I was supposed to get a much larger subset of data regarding AP's, configured WLANS, clients and traffic, etc. It appears in the MIB properties that the data should be available, but perhaps there's been a change of MIB from the Ruckus side (in which case I would also like to pay for getting that updated). Attached is a screenshot of the Device MIBs and SNMP errors as reported by Observium. If you require something else, please let me know. I am using the most recent software version for both vSZ and the APs (R600). Observium is updated from SVN daily(ish). If this requires more heavy lifting or outside of the scope, then please contact me offline for pricing and timeline, and I'll sponsor the additions. Philippe [image: Inline image 1]

1 0

Generate an BGP group with only like clue the interface description?
by Johann Mallet 17 Feb '17

17 Feb '17

Hello, We would like create alert on BGP sessions only on private peering. The only clue for detected these BGP sessions is the description on interface. But Observium seems don't can link session<->network ports. I don't find easy solution to match these sessions. Anybody have an idea? Observium seems to be able link one IP to network port. Is it possible for you add functionality to link BGP sessions (with their IP) to network port too? We could do a really great monitoring and see BGP sessions linked to the network port (on new tab?) :-) Johann

3 4

CSS Bug on 8360?
by Johann Mallet 17 Feb '17

17 Feb '17

Hello, I have update to SVN 8360. And now we don't have all field in search bar on device/port page : Do you have same issue? Thanks, Johann

3 7