Closed GuidoWilden closed 4 months ago
Not sure if this one has slipped through the cracks.
Hi @GuidoWilden
do you have this line in your sysobject.ids ?
674.11000.5000.100.2 Dell NETWORKING
Anyway the following should help:
674.11000.5000.100.2.1.18 Dell NETWORKING S5212F-ON
I still have to check if I can improve new Dell MibSupport module to better support this kind of Dell devices.
Thank you @g-bougard , made that change but still didn't work as expected. The switch did get created in the databases all along but weirdly no ports, or only the management port. I now created the xml on command line and imported it, so all good. I have seen this before though. There is the odd switch that just will not import properly via the task triggered from the GUI. Run it on command line and all is fine.
Maybe you have configured a too short snmp timeout for netinventory task in GLPI and the device doesn't respond in time in that case during ports inventory.
Cranked it all the way up to 60 but still no ports.
That's really weird. Can you share the log part for such a device when run from service with debug2 enabled ?
Also I just pushed an update on Dell MibSupport module to enhance support for Dell os10 products. This means sysobjectid.ids file updates should not be required for other models.
Also the module will set ServiceTag as AssetTag, not sure this is really valuable or in the right place, tell me.
Maybe the xml's reveal something. Archive.zip
Rereading your previous post: these are the only two switches running OS/firmware 10.5.3.3. Maybe worth updating. We have more switches on v10 but they are running 10.5.4.0 and these came in perfectly fine.
In the log you posted earlier, I only saw the netdiscovery part, and this is the netinventory part which is really interesting. You can eventually create a job with a single ip to scan to limit the output.
Thank you for the tip. I have limited it to those two switches. glpi-agent.log.zip
Okay, this is a well-known problem which I think comes from glpi-inventory plugin: as few ips are reported during netdiscovery, the plugin doesn't register the right ip to use when netinventory is requested. Here the first one will be tried on 10.1.1.9 and the second on 10.1.1.10. As SNMP is not answering on these ips, you only have datas related to netdiscovery.
Is there any good reason these ips exist on these devices ?
We still tried to figure out this problem in the past but I don't remember exactly why @skita told me we can't make it works.
My guess is if you use GLPI-Agent ToolBox feature you'll obtain what you expect as it runs netinventory immediatly with the ip used for netdiscovery. But I don't know what could be done with glpi-inventory plugin at that time. Maybe you can manage to remove these ips from the device if they are not really used.
Thank you @g-bougard, I will enquire with our network folks regarding those IPs. If they can't be deleted I will try and use the Toolbox feature. Something I have yet to play with.
Here the details:
snmpwalk output: snmpwalk.txt.zip