- observium - lists.observium.org

CBQoS 'drop' vs 'policed-transmit'
by Robert Williams 14 Aug '14

14 Aug '14

Hi, Just noticed that the CBQoS stuff has gone active recently (which is very cool) - currently we get all this via (very) painful integration with Cacti. On looking closer I've found what I think is an issue, but stand ready to be corrected. What I think is happening is that Observium is considering packets the Cisco counted as "Exceeded" to have been "Dropped" where in reality they were just re-marked and transmitted normally. As an example, here is the graph for gi3/1 (class-default) on a 6500 chassis: Here is the output of the show policy-map int gi3/1: GigabitEthernet3/1 Service-policy input: <name> class-map: <Class A> (match-any) Match: protocol arp Match: access-group name <ACL A1> Match: access-group name <ACL A2> police : 32000 bps 102400 limit 102400 extended limit Earl in slot 3 : 1377162657 bytes 5 minute offered rate 2280 bps aggregate-forwarded 1377162657 bytes action: transmit exceeded 0 bytes action: drop <- Zero drops aggregate-forward 2520 bps exceed 0 bps class-map: class-default (match-any) Match: any police : 40000000 bps 3200000 limit 3200000 extended limit 1000000000 pir-bps Earl in slot 3 : 10562039041987 bytes 5 minute offered rate 11748848 bps aggregate-forwarded 10562039041987 bytes action: set-dscp-transmit exceeded 297846881624 bytes action: policed-dscp-transmit violated 0 bytes action: drop <- Zero drops aggregate-forward 13842064 bps exceed 0 bps violate 0 bps So in summary, there have never been any actual drops on the port, only re-marking of the DSCP (incrementing the exceeded counter) and then a transmit. Since the re-marking occurs in the "exceeded" counter, I think Observium is considering them as dropped, instead of just exceeded. This also messes up the pre-policy and post-policy counters, because it implies that traffic didn't make it through when in reality it all did. We use re-marking extensively and so all of the CBQoS graphs show incorrect drop and post-policy information. Either that or we've got a bug/misconfiguration (or I just don't get it), so maybe someone else can confirm? Let me know what debug info is needed if it helps :) Cheers! Robert Williams Custodian Data Centre Email: Robert(a)CustodianDC.com http://www.CustodianDC.com

3 9

Possible bug in BGP Prefixes graphs?
by Erwin Heringa 14 Aug '14

14 Aug '14

Hi Observium team, Earlier this week the global route table hit 512k prefixes, so I was looking in Observium how many prefixes each peer announces. But there is something wrong that I can't explain and it could be a bug. This morning I upgraded to the latest (paid) version r5747, that didn't change this. For peer 195.69.144.192 my Juniper router shows 89 received prefixes: rtr1> show bgp neighbor 195.69.144.192 | match prefixes Active prefixes: 47 Received prefixes: 89 Accepted prefixes: 89 Advertised prefixes: 7 Via snmp on the router: This peer has index 155, and Peer with index 155 has 89 prefixes: rtr1> show snmp mib wal decimal 1.3.6.1.4.1.2636.5.1.1.2.1.1 | match 144.192 | match PeerIndex jnxBgpM2PeerIndex.0.1.195.69.144.49.1.195.69.144.192 = 155 rtr1> show snmp mib wal decimal 1.3.6.1.4.1.2636.5.1.1.2.6.2 | match \.155\. | match Prefixes jnxBgpM2PrefixInPrefixes.155.1.1 = 89 jnxBgpM2PrefixInPrefixesAccepted.155.1.1 = 89 jnxBgpM2PrefixInPrefixesRejected.155.1.1 = 42 jnxBgpM2PrefixOutPrefixes.155.1.1 = 7 Via snmp on the Observium server: snmpwalk -v3 -l authPriv -u ***** -a SHA -A ****** -x AES -X ******* rtr1 1.3.6.1.4.1.2636.5.1.1.2.6.2 | grep \.155\. iso.3.6.1.4.1.2636.5.1.1.2.6.2.1.7.155.1.1 = Gauge32: 89 iso.3.6.1.4.1.2636.5.1.1.2.6.2.1.8.155.1.1 = Gauge32: 89 iso.3.6.1.4.1.2636.5.1.1.2.6.2.1.9.155.1.1 = Gauge32: 42 iso.3.6.1.4.1.2636.5.1.1.2.6.2.1.10.155.1.1 = Gauge32: 7 But the graphs looks like this, with only 8 prefixes: [cid:image001.png@01CFB7A9.99364360] And even worse for our transit provider: rtr1> show bgp neighbor 195.69.144.34 | match prefixes Active prefixes: 158327 Received prefixes: 505071 Accepted prefixes: 505070 Advertised prefixes: 7 snmpwalk -v3 -l authPriv -u ***** -a SHA -A ***** -x AES -X ****** rtr1 1.3.6.1.4.1.2636.5.1.1.2.6.2.1 | grep '\.4\.1\.1' iso.3.6.1.4.1.2636.5.1.1.2.6.2.1.7.4.1.1 = Gauge32: 505068 iso.3.6.1.4.1.2636.5.1.1.2.6.2.1.8.4.1.1 = Gauge32: 505066 iso.3.6.1.4.1.2636.5.1.1.2.6.2.1.9.4.1.1 = Gauge32: 346727 iso.3.6.1.4.1.2636.5.1.1.2.6.2.1.10.4.1.1 = Gauge32: 7 [cid:image002.png@01CFB7AC.69D3A650] This behavior seems to have changed around March 1st. Any thoughts? Met vriendelijke groet, kind regards, Erwin Heringa SIDN | Meander 501 | 6825 MD | Postbus 5022 | 6802 EA | ARNHEM T +31 (0)26 352 55 59 | F +31 (0)26 352 55 05 erwin.heringa(a)sidn.nl | www.sidn.nl

1 0

syslog no msg
by Vilberg Eiríksson 14 Aug '14

14 Aug '14

Hi All, I have just set up observium now and successfulli integrated it with syslog-ng but i am not seeing anything in the message column. I have tried the patch from her: http://jira.observium.org/browse/OBSERVIUM-850?jql=text%20~%20%22msg%22 but it did not work. If i go to mysql and check the msg field in the syslog table it is just empy (or just has %). All my devices are cisco devices running IOS. Any ideas? Regards Vilberg

2 1

process list WMI
by Simon Smith 13 Aug '14

13 Aug '14

Hi all, just a quick question is the currently a WMI module that checks out for number of running Processes for a windows machine? or even a table I can look at to see the currently running processes on observium? I’ve seen your able to get ahold of the processes via WMI just not sure if the is one already integrated? if the isn’t, can anyone point me in the direction as to create the module ? regards Simon

2 1

Internet 512k global routes
by KENNETH D MILLER II 13 Aug '14

13 Aug '14

With the Internet hitting 512k routes, I was looking in observium for the IPv4 prefixes for our Brocade XMR routers. I'm getting a draw error on all prefix updates. graphs/type=bgp_updates is working ok. graphs/type=bgp_prefixes_ipv4unicast or any of the prefix graphs are not. I've tried on observium version 5728 and CE version 5229. Is anyone else using XMR, CES/CER, or MLX routers getting the same thing? Thanks, Ken

1 0

Description in CentOS Interfaces
by Andres Pugawko 13 Aug '14

13 Aug '14

Hi ! Somebody know if it is possible to add description in CentOS interfaces? There is a script for adding description in Debian interfaces, Is there another for centOS or somebody developed one? Regards Andres Pugawko Network Engineer Description: cid:image001.jpg@01CEAF1A.D081C8A0 Cámara Argentina de Internet – CABASE IXP Suipacha 128 - Piso 3° F (C1008AAD) Buenos Aires - Argentina Tel: (54-11) 4326-0777 int 9112 Cel: 11-6915-7866 <http://www.cabase.org.ar/> www.cabase.org.ar

2 1

Billing
by Mike Grant 13 Aug '14

13 Aug '14

I have the subscription version and have the billing working. What I am trying to do is work out if I can only bill certain traffic. i.e. I only want to charge for international traffic at a $ per MB rate but charge nothing for local traffic. I'm using a Mikrotik Cloudcore router, the international data goes out one gateway and the domestic out another. So I really only want to bill the data for a port that goes in and out one particular gateway. Anyone doing anything like this? Thanks, Mike.

4 5

Defunct processes
by kai＠angry-monk.com 13 Aug '14

13 Aug '14

Hey, Ever since I installed observium I've been getting lost of defunct "sh" processes showing up in my process list. I'm running Ubuntu Linux. 14.04 LTS Linux ***** 3.13.0-32-generic #57-Ubuntu SMP Tue Jul 15 03:51:08 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux I'm running the latest version of the Open source edition (installed 2 days ago) and I see this in my process list all the time since the install. www-data 27725 0.0 0.0 0 0 ? Z 10:37 0:00 [sh] <defunct> www-data 28164 0.0 0.0 0 0 ? Z 10:42 0:00 [sh] <defunct> www-data 28589 0.0 0.0 0 0 ? Z 10:47 0:00 [sh] <defunct> www-data 28989 0.0 0.0 0 0 ? Z 10:52 0:00 [sh] <defunct> www-data 28991 0.0 0.0 0 0 ? Z 10:52 0:00 [sh] <defunct> www-data 29398 0.0 0.0 0 0 ? Z 10:57 0:00 [sh] <defunct> www-data 29810 0.0 0.0 0 0 ? Z 11:02 0:00 [sh] <defunct> www-data 29820 0.0 0.0 0 0 ? Z 11:02 0:00 [sh] <defunct> www-data 30656 0.0 0.0 0 0 ? Z 11:12 0:00 [sh] <defunct> www-data 31062 0.0 0.0 0 0 ? Z 11:17 0:00 [sh] <defunct> www-data 31074 0.0 0.0 0 0 ? Z 11:17 0:00 [sh] <defunct> www-data 31903 0.0 0.0 0 0 ? Z 11:27 0:00 [sh] <defunct> www-data 32698 0.0 0.0 0 0 ? Z 11:37 0:00 [sh] <defunct> www-data 32700 0.0 0.0 0 0 ? Z 11:37 0:00 [sh] <defunct> www-data 32722 0.0 0.0 0 0 ? Z 11:37 0:00 [sh] <defunct> www-data 32732 0.0 0.0 0 0 ? Z 11:37 0:00 [sh] <defunct> Any Suggestions? Kai

1 0

Re: [Observium] Billing
by Anthony Polselli 12 Aug '14

12 Aug '14

Netflow Sent from my Android phone using TouchDown (www.nitrodesk.com) -----Original Message----- From: Mike Grant [mikeg(a)itcl.co.nz] Received: Tuesday, 12 Aug 2014, 2:04PM To: observium(a)observium.org [observium(a)observium.org] Subject: [Observium] Billing I have the subscription version and have the billing working. What I am trying to do is work out if I can only bill certain traffic. i.e. I only want to charge for international traffic at a $ per MB rate but charge nothing for local traffic. I’m using a Mikrotik Cloudcore router, the international data goes out one gateway and the domestic out another. So I really only want to bill the data for a port that goes in and out one particular gateway. Anyone doing anything like this? Thanks, Mike.

1 0

Sensor_value alerts do not process
by Alain Gyger 12 Aug '14

12 Aug '14

I'm using SVN version 5742 and all of my alerts work except those that are supposed to check a sensor_value metric. I'm able to check and get emails for alerts on processor_usage, toner_current, and mempool_perc, to name a few, so I know the system is working. When I run the alerter.php script against one of the hosts indicated under the processor_usage alert and I get "1 Status: OK" messages. But when I run alerter.php against a host indicated under the sensor_value alert for Battery Capacity or for APC UPS Temperature I get "-1 Unknown status". The graphs for these sensors are fine, the MySQL table sensors-state lists the correct values as well. My configuration for temperature looks like this: Entity Type: Temperature Delay: 1 Status: 0/0/0/0/2 Check Conditions: Require any condition Tests: sensor_value gt 40 Device Match: type match power Entity Match: sensor_descr match Temp* The alert checker seems fine, but it looks like the tests are failing when they are run. Has anyone else had the same issues with sensor_value alerts? Or is this feature not quite ready to be used?

1 0