- observium - lists.observium.org

Renamed interface - how to keep unified history?
by Adam Thompson 21 Feb '19

21 Feb '19

I made a change on one of our routers that caused a new ifIndex & name to be assigned to a customer circuit (interface). Observium is correctly reporting the data on the old virtual i/f (which still technically exists) and – naturally – separately on the new virtual i/f. Is there any way to (retroactively???) tell Observium that these are really the same interface, and to display only one graph for the two ifIndexes? Or, if I correct the situation so the old ifIndex & name is being showing traffic again, is there any way to move, or inject, the collected data on the ‘new’ port into the ‘old’ port’s RRDs? (i.e. avoid a multi-day gap in a multi-year dataset) We are reasonably up-to-date, running 19.2.9712 – not sure what other info will be helpful here. Thanks, -Adam Adam Thompson Consultant, Infrastructure Services [merlin-email-logo] 100 - 135 Innovation Drive Winnipeg, MB, R3T 6A8 (204) 977-6824 or 1-800-430-6404 (MB only) athompson(a)merlin.mb.ca<mailto:athompson@merlin.mb.ca> www.merlin.mb.ca<http://www.merlin.mb.ca/>

3 5

Alert Checker Question
by Ryan Huff 20 Feb '19

20 Feb '19

I have an alert checker called, "device down" and it tests for "device_status = 0". Occasionally, this alert will fire, but the underlying OS still has network connectivity to these device (ICMP). What this tells me, is that the underlying SNMP daemon on the server may be malfunctioning, because if OBS can't make and SNMP poll to a device, it makes sense to me that it would think the device_status = 0 for that device. So given that this doesn't seem to be an OBS issue, is it something others have encountered and found away around? I'm Ubuntu 18.04 here... just wondering if I need to throttle the poller or if there are other known way the community has addressed this? Thanks, Ryan

2 2

L2/L3VPN issues with Juniper MIBs
by Luis Balbinot 19 Feb '19

19 Feb '19

Hi. This has been bothering me for some time but I've only got the time to look at it now. When we list Ports for any given Juniper device we get a nice green label right below the speed/MTU/lag column with the name of the VRF when it's an L3VPN. Why not the same for L2VPNs? That would be nice, specially if the RT is encoded in the name (makes tracking issues easier). Second, on the righthand side of the Ports table there's an unnamed column right next to the MAC Address with some information regarding the interfaces, LAG membership, etc. There's a xconnect icon that is supposed to show the remote interface and device for L2VPNs, but that's completely messed up (see image attached). Are you aware of that? Can that be fixed? The device is correct, but the interface isn't. The same goes for L3VPN. The interfaces shown on that right column are not endpoints of the same VPN and even the devices are wrong. I'd love to help on that. Just checking ahead to see if there's anything stopping you from fixing it at MIB level or something. Thanks! Luis

1 0

Session lifetime
by Ricardo Franco Cantero 18 Feb '19

18 Feb '19

Hi, We just upgraded our Observium and (perhaps it's already mentioned but the archive doesn't work) since the upgrade we keep being logged out every 30 minutes or so. We have it running on a large screen but every 30 minutes (it looks like 30 minutes, haven't timed it) it need to enter my credentials again to log in. I know I can check the box Remember my Credentials but that's not what I'm asking. I want it to keep the session alive longer like it did. Kind regards, Ricardo Franco Cantero

6 10

IPv6 privacy addresses vs IP Session binding
by Darrell Budic 18 Feb '19

18 Feb '19

Pretty obvious when I figured out what was happening, IPv6 privacy addresses aren’t really compatible with IP Session binding. As I discovered and was driven nuts by for a day or two after my ISP added v6 :) If you’re using v6 and seeing rapid session expiry, try disabling IP Session binding. Consider this a RFE for IP Session binding on IPv4 addresses only. Maybe a second request for a loose form of IPv6 session binding to within the same network range (either /64 or user configurable?) to allow some restrictions on it? Not sure how feasible that would be, but it sounds like it could give us some reasonable v6 protections. Thanks for reading, -Darrell

1 0

Custom Widgets
by Mirditor Kuçaj 18 Feb '19

18 Feb '19

Hello I am searching for a way to customize the widgets. With the old dashboard it was easy with the frontpage filters. With the new one I cant EDIT the SYSLOG Widget for displaying only desired severity and any Filter (Only in widget, not in global config) And the device status boxes doesnt display more than 10 boxes even if I has a lot of dashboard space. Displays a unclickable XX more Is there any documentation where can I edit them. I am running Observium 19.1.9661 Regards Mirditor Kuçaj Backbone Network Director <mailto:miri@abissnet.al> miri(a)abissnet.al Abissnet Sh.a. +355 69 60 91 364 +355 (0)44 300 003 Ext 139 Rr. Ismail Qemali, Pallati Abissnet, Tiranë <http://www.abissnet.al/> www.abissnet.al <https://www.facebook.com/Abissnet> <https://plus.google.com/u/0/+AbissnetalISP> <https://www.linkedin.com/profile/view?id=AAIAABqLWRAB88ljU6t8b67oVm8PAlvPrh mPybE&trk=nav_responsive_tab_profile%20>

1 0

Question on alerts migration
by xomka686 16 Feb '19

16 Feb '19

Hi Adam, Recently you've mentioned "/migrate_alerts/" link to be used to migrate old alert format to the new one. Is it's usage documented anywhere? Or it's just 1-click operation? I was looking for such functionality and have a few customers with old alerting seaking for easy migration to new style. --- Best regards, Sergei

2 1

Questions regarding parsing
by Ryan Huff 15 Feb '19

15 Feb '19

Are there currently any built-in ways (that I clearly would have missed in the documentation) to create/add custom parsing/metrics? For example; I am a telephony engineer and my routers do the normal route-y things, but they also do specialized things with specialized configurations that are not present on all or most routers (e.g Dial Peers). Is there a way to define/capture configuration sections to parse (clearly, outside the bounds of SNMP, I get it..) and then either populate as a tile on the device page or some type of menu item added to the navigation bar; thinking perhaps the regex engine that is used to create the Syslog alerts or something.. I hate one-offing and "plugin'ing" code, cause then that's something I have to support; so I'm committed to using OB the way its offered, just trying to see if there was a creative way to get this done.. Thanks, Ryan Huff, CCDP, CCNP Cisco Certified Network and Design Professional

1 0

Edit Alert Association Ruleset
by Steve 15 Feb '19

15 Feb '19

Hi, is it possible to edit "Alert Association Ruleset" on an alert checker ? How? Many Thanks Steve

2 4

reverse geocoding not working
by Joey Stanford 14 Feb '19

14 Feb '19

Hi, We have our SNMP location strings set in all but a few cases to “NAME [xx.xxxxxx,-yy.yyyyyyy]” and as you can see from the screenshot that works for over a hundred hosts. For some reason though, not all of them are being interpreted properly. Any idea how to troubleshoot (or fix) this? We’re not overriding anything in config.php.

2 2

Add Lxd edfa/emtx devices patch
by Reto Zingg 14 Feb '19

14 Feb '19

Hi, In a network we monitor there are Langxunda (Lxd) Externally Modulated Optic Transmitters and Erbium Dotted Fibre Amplifiers. E.g: http://www.langxunda.com/en/goods/1167.html The current OS detection for the NSCRTV-ROOT mib does not work on them. So I added rules for them in os.inc.php. The limits which are set for the sensors in nscrtv-root.inc.php do not work for us. So I added some checks to not set a limit. Additionally I added some more sensors. We do not own the copyright of the Lxd logo, I don't know what your policy is, wrt the use of logos. I got it from the web page and cropped it to look nicer. It would be nice if the attached patch and logo could be added. If there are changes needed, please let me know (style, naming convention, etc). Thanks and best regards Reto

2 3

is there any solution to linux not having persistent port ifIndexes ?
by Edvinas K 13 Feb '19

13 Feb '19

Hello, Since we'all know that Linux is not having persistent port if Indexes, is there any quick-fix workaround to make the ifindexes stay after server reboot ? Observium wise or Linux wise ? Thanks

3 6

Assign specific dashboard to user
by Fiberwide Sysadmins 12 Feb '19

12 Feb '19

Hi All, i'm trying to setup a configuration to give access to traffics data to our customers, but I've encounter some problems on the dashboard. Now You can define more dashboards but I can find a way to define a dashboard for our NOC and a dashboard for the customers... The default is applied to all users, but i set this with backbone graphs, so in the case of customer with only USER privilege log in all the graph are showed as "Permission error", that's because are related to device or ports not assigned to that specific customer. So my idea was to create a dashboard only with "Status summary" and "Event Log", those to widget show only the info related to the customer. Other minor question, can You tell Me a way to change the name of the dashboard? Distinti saluti --- Davide Lima Daum - System & Network Engineer Fibering S.p.a. - www.fiberwide.com T: +39.02.00632100 F: +39.02.00632199 e: sysadmin(a)fiberwide.com

2 1

Changelog
by Benjamin Moore 12 Feb '19

12 Feb '19

Is there somewhere I can see the changelog for the rolling version of Observium?

1 0

Having issues with the email alerting..
by Ryan Huff 11 Feb '19

11 Feb '19

Having issues getting the alert to mail. I am not using SMTP, I have sendmail installed locally to the OS. If I use the PHP mail(); function in a test script and execute it in the broswer, it sends mail just fine. Here is the email params in my 'config.php' file: $config['email']['enable'] = TRUE; $config['email']['backend'] = 'mail'; $config['email']['sendmail_path'] = '/usr/sbin/sendmail'; $config['alerts']['interval'] = 40000; $config['email']['default'] = "me(a)company.com"; $config['email']['from'] = "Observium <observium(a)company.com>"; $config['email']['default_only'] = TRUE; Any thoughts mates? Thanks, Ryan Huff, CCDP, CCNP Cisco Certified Network and Design Professional

2 2

mcrypt deprecated
by Lane Russell 11 Feb '19

11 Feb '19

It appears that the install instructions for Ubuntu 18.04 still require that mcrypt be enabled. However, mcrypt was deprecated in PHP 7.1 and fully removed in PHP 7.2, which is used in Ubuntu 18.04. In addition to updating the install documentation, what is the timeline for replacing mcrypt in Observium? What are the correct steps for installing Observium on Ubuntu 18.04? Regards, Lane Russell Sent with [ProtonMail](https://protonmail.com) Secure Email.

2 7

Dealing with Alert Storms
by Joey Stanford 11 Feb '19

11 Feb '19

Hi All, I’m looking for advice on how to deal with alert storms. For example, we lost a UPS last night at a critical hub which cause Observium not to be able to reach most of our networks (and of course, all the alternative OSPF routes were powered via that UPS). Consequently we had several hundred alerts sent out via email and telegram. Is there any interesting way to deal with this that anyone can recommend? Thanks, Joey

4 8

Printing reports
by Ryan Huff 09 Feb '19

09 Feb '19

Using latest CE.. Just making sure I haven’t missed something in the docs and config guides.. I do not see a way to print any type of reports/PDFs.. but I do see a ../html/pdf.php which looks like it would do that. Did I miss a page in the docs? -Ryan

3 7

Map Location
by Ryan Huff 08 Feb '19

08 Feb '19

It seems that no matter what addressing schema I've tried to set with config['location_map']['nice_name'] ['real_address'], it continues to populate the front page map with the proper location names, but all located in Ponta Delgada on the island of São Miguel. The USA addressing formats I've tried are: street address, city, state, country and street address, city, state (Ex: 180 Peachtree Street, Atlanta, Georgia, USA) Is there an additional override I am missing, or a way to verify the location value I'm typing is understandable to the correct geopositioning on the map? Thanks, Ryan Huff, CCDP, CCNP Cisco Certified Network and Design Professional

2 2

Alert Priority
by Mike Hyslop 08 Feb '19

08 Feb '19

Hi All Is there a way to add priorities to alerts at the moment all I have to choose from is critical Cheers Mike.

1 0

Graph scale question
by Ryan Huff 07 Feb '19

07 Feb '19

Using the latest CE.. All the graphs’ time scale seems to be off by a few hours. For example, a device’s port graph shows traffic for today at 2100 hours, and it’s only 1723 currently. I have verified the PHP instance Observium is using reports the correct TZ (America/New_York) and also the monitored device is in the correct TZ. Is there a UTC offset I need to adjust somewhere? Thanks, -Ryan

3 3

APC PDU's, some change have made outlet status to show twice.
by Henrik Cednert (Filmlance) 07 Feb '19

07 Feb '19

Hello Some fairly recent update and made my PDU’s a bit messy in obervium. I know have two statuses per outlet. See pictures of two different APC models. Cheers [cid:BE5E0B1D-C29C-4C36-B635-931B3D24A996@metronome.local] [cid:11A976F8-A9AD-4931-A332-6F143F3B7EA3@metronome.local] -- Henrik Cednert / + 46 704 71 89 54 / CTO / Filmlance Disclaimer, the hideous bs disclaimer at the bottom is forced, sorry. ¯\_(ツ)_/¯ Disclaimer The information contained in this communication from the sender is confidential. It is intended solely for use by the recipient and others authorized to receive it. If you are not the recipient, you are hereby notified that any disclosure, copying, distribution or taking action in relation of the contents of this information is strictly prohibited and may be unlawful.

2 3

Minigraphs, names and labels. Some issues.
by Henrik Cednert (Filmlance) 07 Feb '19

07 Feb '19

Hello I mentioned this in a thread about something else but think it drowned and got lost. lifting it out to a separate thread since I find these issues very frustrating. Not sure if I’m alone or not. 1. Can’t I rename graphs like we could in config.php? 2. Is there any way to move the label so that it doesn’t hide parts of the graph? 3. They mini graphs are a lot slower to refresh now, can this be tweaked? 4. I get the feeling that “show legend” on the larger graphs is semi-broken and is more or less always off here. Or is there some global setting that can have been changed when upgrading? Cheers and thanks [cid:782CC26C-348D-4DBE-95F4-097AC52CBFBA@metronome.local] -- Henrik Cednert / + 46 704 71 89 54 / CTO / Filmlance Disclaimer, the hideous bs disclaimer at the bottom is forced, sorry. ¯\_(ツ)_/¯ Disclaimer The information contained in this communication from the sender is confidential. It is intended solely for use by the recipient and others authorized to receive it. If you are not the recipient, you are hereby notified that any disclosure, copying, distribution or taking action in relation of the contents of this information is strictly prohibited and may be unlawful.

2 1

Cisco RV340/345
by Benjamin Moore 06 Feb '19

06 Feb '19

It appears that the interfaces and everything else isn't working in Observium. Does anyone have any experience with this device? I'm happy to post an snmpwalk if it might help it get better support.

1 0

CDR Overage Alerts
by Andrew Lemin 04 Feb '19

04 Feb '19

Hi List, I know there has been a long-time request to enable the ability to generate an Alert when traffic (peak or 95%) goes over the CDR (as configured within [CDR] on the interface), and we would like to know if this has been realised yet? We just need a way of dynamically detecting and reporting when traffic goes over the configured CDR, using the port description’s [CDR] value. We cannot create Observium billing reports for every single customer facing port as we have thousands, and we are trying to keep manual processes to a minimum as this could easily be a generic alert of the form; ifInUcastPkts_rate > CDR OR ifOutUcastPkts_rate > CDR I believe historically there was some outstanding work to convert the discovered [CDR] field data into a normalised bps value in the database (as CDR can be given in units of Bits/Megabits/Gigabits etc). Thanks, Andy.

1 0