Yea, the `main` Docker image has issues with being behind release, there's an open PR to fix it.
On Tue, Nov 21, 2023 at 7:38 AM Elliott Balsley <[email protected]> wrote: > Nice, that sounds like it might fix it. I’m running the docker image > prom/snmp-agent:main which was published 2 days ago. But I don’t know which > commit that refers to. > > I read the description of this new patch: > first attempt to match the lookup node so that the OID matches that of > the metric in all but the last branch. > In case no match is found, it reverts to the old behavior. > > Would it be better to attempt this recursively, so if it fails to find the > name in the last branch, then it looks one branch up, and so on? I’m > curious, if it were implemented that way, would there ever be a need to > specify multiple MIBDIRS? > > On Mon, Nov 20, 2023 at 21:43 Ben Kochie <[email protected]> wrote: > >> Have you updated your generator recently? We added a patch that tries to >> use OIDs closer to the original index first. >> >> See: https://github.com/prometheus/snmp_exporter/pull/828 >> >> There's also now support for multiple files in the exporter, as well as >> different input filenames in the generator. You could do something like >> this: >> >> MIBDIRS="mibs/base:mibs/vendorA" >> generator --generator-path="generator-vendorA.yml" >> --output-path="snmp-vendorA.yml" >> MIBDIRS="mibs/base:mibs/vendorB" >> generator --generator-path="generator-vendorB.yml" >> --output-path="snmp-vendorB.yml" >> snmp_exporer --config.file="snmp-*.yml" >> >> Longer-term, I want to add the ability to specify MIBDIRS pre module in >> the generator. This way you can put the conflicting MIBs in different dirs >> and still have one config. >> >> On Tue, Nov 21, 2023 at 4:18 AM 'Elliott Balsley' via Prometheus Users < >> [email protected]> wrote: >> >>> Aha! I found the problem by looking closer at the generated snmp.yml. >>> This PDU manufacturer OID is *1418*. I have another MIB installed from >>> a different manufacturer *1718* which also has an OID called >>> outletName. So the generator config I shared before actually produces >>> incorrect snmp.yml as below. The solution I found is to be explicit in the >>> lookup like this: >>> lookups: >>> - source_indexes: [outletIndex] >>> lookup: 1.3.6.1.4.1.1418.6.5.1.2 >>> >>> Is this expected behavior and is there a better way to handle it? >>> >>> Incorrect snmp.yml: >>> - name: outletStatus >>> oid: 1.3.6.1.4.1.1418.6.5.1.6 >>> type: gauge >>> help: ' - 1.3.6.1.4.1.1418.6.5.1.6' >>> indexes: >>> - labelname: outletIndex >>> type: gauge >>> lookups: >>> - labels: >>> - outletIndex >>> labelname: outletName >>> oid: 1.3.6.1.4.1.1718.3.2.3.1.3 >>> type: DisplayString >>> >>> On Mon, Nov 20, 2023 at 6:41 PM 'Elliott Balsley' via Prometheus Users < >>> [email protected]> wrote: >>> >>>> I'm trying to configure snmp-exporter with a PDU, using a lookup to >>>> label each outlet with its human-readable name (outletName) by referencing >>>> its outletIndex. But when I try to do this, the outletName label is an >>>> empty string. Is there something wrong with my generator config? Example: >>>> >>>> >>>> >>>> >>>> * walk: - 1.3.6.1.4.1.1418.6.5 lookups: - >>>> source_indexes: [outletIndex] lookup: outletName* >>>> >>>> This is how the metrics look. >>>> *outletStatus{outletIndex="0",outletName=""} 1* >>>> >>>> Without any lookups, the outletName appears correct: >>>> *outletName{outletIndex="0",outletName="1-MacStudio"} 1* >>>> >>>> Below is how the raw SNMP data looks: >>>> >>>> % snmptable -v2c -c netflixsnmp -m IBOOTPDU-MIB 10.37.155.163 >>>> 1.3.6.1.4.1.1418.6.5 >>>> SNMP table: IBOOTPDU-MIB::outletTable >>>> >>>> outletIndex outletName outletInitialState outletCycleTime >>>> outletControl outletStatus outletActualStatus >>>> 1 1-MacStudio last 10 >>>> on on on >>>> 2 2-Mezzo-1 last 10 >>>> on on on >>>> 3 3-UltraStudio last 10 >>>> on on on >>>> 4 4-MP-60 last 10 >>>> on on on >>>> 5 5-Mezzo-2 last 10 >>>> on on on >>>> 6 6-Core110f last 10 >>>> on on on >>>> 7 Outlet-7 last 10 >>>> on on on >>>> 8 8-Roku last 10 >>>> cycle on on >>>> >>>> I'm not sure if this matters but I noticed the outletIndex values start >>>> at 1, while the OID table indexes start at 0. Could that be the problem >>>> here? >>>> I've attached the MIB. I also opened a Github issue with more details >>>> in case this is a bug: >>>> https://github.com/prometheus/snmp_exporter/issues/1039 >>>> >>>> -- >>>> You received this message because you are subscribed to a topic in the >>>> Google Groups "Prometheus Users" group. >>>> To unsubscribe from this topic, visit >>>> https://groups.google.com/d/topic/prometheus-users/MPuTk5yDsAg/unsubscribe >>>> . >>>> To unsubscribe from this group and all its topics, send an email to >>>> [email protected]. >>>> To view this discussion on the web visit >>>> https://groups.google.com/d/msgid/prometheus-users/19aca381-36e0-4e05-8044-6aecda223da6n%40googlegroups.com >>>> <https://groups.google.com/d/msgid/prometheus-users/19aca381-36e0-4e05-8044-6aecda223da6n%40googlegroups.com?utm_medium=email&utm_source=footer> >>>> . >>>> >>> -- >>> >> You received this message because you are subscribed to the Google Groups >>> "Prometheus Users" group. >>> To unsubscribe from this group and stop receiving emails from it, send >>> an email to [email protected]. >>> To view this discussion on the web visit >>> https://groups.google.com/d/msgid/prometheus-users/CALajkdhdvtgpQz5iq_7cFtyvYNc8hCzqxE29kEUGmuCsoZ42Ng%40mail.gmail.com >>> <https://groups.google.com/d/msgid/prometheus-users/CALajkdhdvtgpQz5iq_7cFtyvYNc8hCzqxE29kEUGmuCsoZ42Ng%40mail.gmail.com?utm_medium=email&utm_source=footer> >>> . >>> >> -- You received this message because you are subscribed to the Google Groups "Prometheus Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/prometheus-users/CABbyFmqrRzS%2BTEMrvrp%2BGoe5Cz0A3V-z5CCaD_tSoUP9HgNYnA%40mail.gmail.com.
