Hy Adam,
In real live it looks like this if everything is "good":
<<<md>>> Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] [raid4] [raid10] md1 : active raid1 sdb2[1] sda2[0] 59559808 blocks super 1.2 [2/2] [UU]
md0 : active raid1 sdb1[1] sda1[0] 1950656 blocks super 1.2 [2/2] [UU]
unused devices: <none>
Criterions for Yellow Alarm (my opinion): One or more "U" is replaced with "_" or "recovery" is found
Criterion for Red Alarm (my opinion): Device is marked as inactive and has "_"s
Look at: https://raid.wiki.kernel.org/index.php/Mdstat
Thanks Richard
-- Richard Müller | Geschäftsführer
teamix GmbH Südwestpark 43 90449 Nürnberg
Tel.: +49 911 30999 0 | Fax: +49 911 30999 99 mail: rm@teamix.de | web: http://www.teamix.de | blog: http://blog.teamix.de
Amtsgericht Nürnberg, HRB 18320 | Geschäftsführer: Oliver Kügow, Richard Müller
***Save the Date: Besuchen Sie uns auf der it-sa Messe in Nürnberg vom 06.10.15 – 08.10.15 auf Stand 12-205***
-----Ursprüngliche Nachricht----- Message: 1 Date: Sun, 23 Aug 2015 12:26:07 +0100 From: "Adam Armstrong" adama@memetic.org To: "" observium@observium.org Subject: Re: [Observium] "Monitoring" Linux Softraid State? Message-ID: ea384ce9-cb37-4664-83e1-95356caeab5e@getmailbird.com Content-Type: text/plain; charset="utf-8"
Is all of the required information returned in the md section?
It'd probably just take a bit of code to parse the output and inject the required entries into the status system.
adam. On 23/08/2015 10:31:37, Richard M?ller rm@teamix.de wrote: Hy *, ? I just noticed that observium_agent returns a <<<md>>> section: ? 1.)??? Question: is there actual a way of monitoring the state of linux softraid devices in Observium, I just do not know? Documentation / google is quite silent about this question. :-)
2.)??? If no, how can this be implemented? Idea #1: Introduce a ?(red|yellow) Alarm? state even for Unix devices, as most common network devices have. So this technique can be also used for different kinds of alarms (Raid, Network, ECC, ?) a.?????? Yellow Alarm: Raid degraded b.????? Red Alarm: Raid broken :-) ? Idea #2: Is this something, which can be implemented ?outside Observium? (e.g. by some ?geniuos? snmpd-configuration lines in snmpd.conf) and Observium just uses the normal ?(yellow|red) Alarm? way, which is already implemented? Sorry, I am not an SNMP-Expert, so I cannot answer this for myself. ? Thanks Richard --