Likewise, I have a customer that has two separate locations… and we have/use the format “customer-gw.location.domain.net” so when one of those devices has an outage, it just shows up as “customer-gw”, so we have to drill down into the device and find out where that device actually is…. Is it customer-gw.location1 or customer-gw.location2?
…Ron
From: observium [mailto:observium-bounces@observium.org] On Behalf Of Justin Miller Sent: Tuesday, January 24, 2017 1:34 PM To: Observium Network Observation System observium@observium.org Subject: Re: [Observium] hostname shortening bug
Tom,
I have this problem as well, and it might be more helpful if the behavior shows the first 11 characters of the whole hostname instead of shorting it to just the first delimiter.
I use names like af1.pry-214d, af1.byrd-osa, and af1.trc-rcs. The first two both get shortened to af1 and I can’t tell them apart in some places, while the latter is always shown correctly. I’d rather see af1.pry-214... and af1.byrd-os… (or similar) than af1 and af1.
This is in the p2p radios list under ports:
In All devices it shows fine.
Justin
On Jan 24, 2017, at 4:43 AM, Tom Laermans <tom.laermans@powersource.cx mailto:tom.laermans@powersource.cx > wrote:
Sebastian,
It's weird that you think this is "fully random". You are using different naming schemes for 20% of your switches yet somehow that escaped you... :-)
Observium splits on DNS label ('.'-separated), there is a maximum hostname length to show in many places (to keep things readable).
It is therefore logical that one switch is fra4.b and the other is fra4-b-cs-ex4200, as fra4.b.e2-switch1 would exceed the maximum length of 12. I guess it is a bug we don't shorten the other one to fra4-b-cs-e, which is likely to be caused by not using a . in the name at all...
Tom
On 01/24/2017 09:49 AM, Sebastian Klute wrote:
Hi Adam,
I've attached some screenshots to show you the differences mentioned. Hope you got everything you need.
Kind regards, Sebastian
Am 20.01.2017 um 13:00 schrieb observium-request@observium.org mailto:observium-request@observium.org :
Message: 2 Date: Thu, 19 Jan 2017 15:48:07 +0000 From: "Adam Armstrong" mailto:adama@memetic.org adama@memetic.org To: "" mailto:observium@observium.org observium@observium.org Subject: Re: [Observium] hostname shortening bug Message-ID: mailto:7c316156-63ac-4680-89c3-59c3f773d281@getmailbird.com 7c316156-63ac-4680-89c3-59c3f773d281@getmailbird.com Content-Type: text/plain; charset="utf-8"
Hi,
Please provide screenshots of all relevant information so we know what you're talking about. :>
adam. On 19/01/2017 15:46:34, Sebastian Klute mailto:sk@accelerated.de sk@accelerated.de wrote: Dear Team,
we have noticed since a couple of days that some switches are shorten in the name. In detail: At the overview page under the section 'Status Warnings and Notifications' we got several switches shorten in name. For example the name is 'fra4.b.j1-switch1.acc' and in the given section it is shorten to 'fra4.a' and thats it. This behaviour is not occuring at other switches. Only at about 20% of our infrastructure this strange shortening appears. Fully random, no particularly order or chronological order to notice. I cannot get an idea why this is happening.
The switches are added through hostnames as usual. First I thought this is caused by the dots in the name that they are somehow noticed as a kind of DNS suffix and that the names are shorten by this. But at other switches named by the same scheme this is not the case.
Do you got an Idea why this is happening or what we can do against it?
Kind regards, Sebastian
