Graphs not working over 400MB of traffic

[aib]

as we already discussed, you have to use a key which contain ifHCInOctets

in your trigger I see ifInOctets, which is NOT working with high-speed interfaces.

Did you do it by mistake or for special purpose?

[ptera]

ifHCOutOctets

Cannot implode expression "8*{Template SNMP Network Device 64bit Interfaces:ifHCOutOctets[{#SNMPVALUE}].last()}/{Template SNMP Network Device 64bit Interfaces:ifSpeed[{#SNMPVALUE}].last()}>9.0". Incorrect item key "ifHCOutOctets[{#SNMPVALUE}]" provided for trigger expression on "Template SNMP Network Device 64bit Interfaces".

[ptera]

I have it configured this way

Key
ifInOctets[{#SNMPVALUE}]

SNMP OID
IF-MIB::ifHCInOctets.{#SNMPINDEX}

[ptera]

So I changed...

Key
ifInOctets[{#SNMPVALUE}]
to
Key
ifInHCOctets[{#SNMPVALUE}]

The
SNMP OID is still
IF-MIB::ifHCInOctets.{#SNMPINDEX}

[ptera]

Immediate triggers on connections that are not congested...

ONB-OSPF.ptera.net
.ONB-OSPF.ptera.net Interface GigabitEthernet1/5 Outbound Congestion 2014-11-14 13:40:12 3s No
.ONB-OSPF.ptera.net
.ONB-OSPF.ptera.net Interface Vlan202 Outbound Congestion 2014-11-14 13:40:12 3s No
.CET-ColoSW.ptera.net
.CET-ColoSW.ptera.net Interface GigabitEthernet0/2 Outbound Congestion 2014-11-14 13:40:11 4s No
.CET-ColoSW.ptera.net
.CET-ColoSW.ptera.net Interface GigabitEthernet0/3 Outbound Congestion 2014-11-14 13:40:11 4s No
.CET-BGP.ptera.net
.CET-BGP.ptera.net Interface GigabitEthernet0/2 Outbound Congestion 2014-11-14 13:40:08 7s

[PortalNET]

Hi guys

i know this is a very old thread, it´s not my intention to be a post digger.... but this was the most similar problem topic i have found so far....


i have a strange issue, using zabbix 4x...


i have the following scenario, 1gb link connection on our small ISP,

we have triggers to graph our pppoe-clients traffic monitoring..

it connects on the device via SNMP v2

when we add the host and set it up properly it connects and takes around 30mins to update and create all graphs from the radius pop server machine, it loads all the graphs for all the clients, link interfaces, cpu, etc.etc.etc.

after around 6 to 8h of graphing, everything stops..........giving me blank results...i do not think its related to link speed, cause our link goes over 400mbps easily... and all the other interfaces NICs on the radius server machine are working...

we monitor graph the main internet link interface also which is on the same device, and it works just fine, including CPU etc.etc.etc..

just the users pppoe-graphs that stop working after a half- a dauy or so.... with no alarm triggering functions.

i am kind of lost .... will try to check the logs for any kind of hints possible..

all our fiber optics structure network of over 60km in the region.. we monitor pings below 1 to 2ms... so don´t think it might be connection losses...

only thing similar to the initial post is our 1gb connection NIC on the zabbix server...

its running 64x bits verion zabbix 4.x


As shown on attachment, the Link graph working fine... but pppoe client login stops working, and ppoe connection is active for the last 72h without interruption

[PortalNET]

The graphs below show the issue.