Community discussions

 
rodolfo
Long time Member
Long time Member
Topic Author
Posts: 543
Joined: Sat Jul 05, 2008 11:50 am

two ideas for dude

Sun Mar 18, 2012 9:17 am

idea1: when the dude sends notify email message, it must group messages from last 3 minutes in a single email. This is necessary especially when a subtree of the network seems to go down because of single point to point and a rid of message are sent.

idea2: set automatically to unack the routers that return up. now, if an acked router return up, then it return down, it goes in acked status, avoiding notification up>down
rodolfo
IZ0UQV
 
fbsdmon
newbie
Posts: 29
Joined: Thu Dec 29, 2011 11:37 pm

Re: two ideas for dude

Sun Mar 18, 2012 11:03 am

1: You can solve this by using parents (see device setup). If a parent device goes down, you will not receive notifications for the children devices.

2: I'm not sure if I got this right, but you are complaining that if you Ack a device it doesn't alert you that it came back up? IF so, don't ack it :) There is nothing wrong with a device being red :)
Ack/Unack is ment to be used to alert other members of your team that someone has seen this error , acknowledge it and started working on it. Or when you are testing something and don't want errors to pop-up for everybody.
 
rodolfo
Long time Member
Long time Member
Topic Author
Posts: 543
Joined: Sat Jul 05, 2008 11:50 am

Re: two ideas for dude

Sun Mar 18, 2012 1:14 pm

thanks

1. ok, but I have two dude server in my network and the configuration could be complicate to maintain. is better to have 1 email with 100 rows than 100 email with one row (the same for sms!)

2. this is what happen:
- if device goes down it become RED
- I ACK and it become BLUE
- if the device return in line it become GREEN (but it does not unack itself!)
- if the device return down it does not become red, but BLUE because of in the precedent step I havent Unack'ed it.
This is the problem: I hope the device Unack automatically itself when change from ack>up, otherwise my collegue think it is also maged.
rodolfo
IZ0UQV
 
fbsdmon
newbie
Posts: 29
Joined: Thu Dec 29, 2011 11:37 pm

Re: two ideas for dude

Sun Mar 18, 2012 4:04 pm

1. No need for 2 dude servers. You can use one of them as a dude agent. You do that by setting up a special user called agent (see groups tab from admins section) on one of the dude servers, then add it as an agent (agent tab under global settings) on the other dude server. Then in device settings, select the dude agent (not dude server 'default') for each device that is to be puled by the agent.
- Then parenting will work and you will be getting one email per outage. Also, I setup e-mail notification only on once service per device, so I don't get 10 messages per device, but one!

2. This is normal as far as i know. Again, the purpose of ack is to inform others that you are working on a problem or the problem will not be resolved soon and you dont want red on your map. After your resolve the problem, you should unack. Ack/Unack is a manual thing!
Maybe the developers can add feature where you can choose to auto unack, but from experience, even other software use the same logic like dude.
 
Sob
Forum Guru
Forum Guru
Posts: 4808
Joined: Mon Apr 20, 2009 9:11 pm

Re: two ideas for dude

Sun Mar 18, 2012 7:30 pm

People forget to do things. So they ack the device, work on it, fix it and before they have chance to unack it, some other urgent task pops up and in the end they never get to unacking the device. And no one notices, because the device is nice and green. And half a year later, when the device goes down again, it's still acked, which is of course not good. So auto unack option would be nice. Even better than simple "unack on up" would be "unack when up for <configurable time>".
People who quote full posts should be spanked with ethernet cable. Some exceptions for multi-topic threads may apply.
 
User avatar
geoffsmith31
Member Candidate
Member Candidate
Posts: 151
Joined: Fri Nov 05, 2010 6:08 am
Location: Toowoomba, Australia

Re: two ideas for dude

Tue Mar 20, 2012 9:59 pm

I have the similar issue with Ack/Unack situation. On a map whith multiple devices, each with multiple services it is necessary to Ack anything that is down. For 2 reasons: 1. so that other members of our team know that someone else is aware of the issue and is working on it. 2. to stop the "down" service from masking other failures on that device/map (ie. If the device is already showing orange due to disk being over 90% full (non-critical) you will not see a change in device status if a critical service dies unless the original issue has been Ack-ed).

The necessity to manually UnAck devices when issues are fixed becomes a problem in a very busy IT environment with multiple team members. Quite often in our environment someone will forget to UnAck the issue they have fixed and then next time it happens, we miss it. Having Acks automatically Unacked when the device/probe goes Up would certainly solve this issue for us. There are other circumstances where we have mobile decices that go Up/Down as part of normal use. We keep these Acked so that they do not present as red when they are disconnected. So we need a checkbox or something per device to indicate if Ack-ed outages should be automatically UnAck-ed.
 
Bivvy
newbie
Posts: 32
Joined: Sat Feb 04, 2017 1:36 am

Re: two ideas for dude

Wed Oct 17, 2018 11:58 pm

Agree - the need to manually UnAck is a significant problem. Either need
- Ack to be cleared automatically when service comes back up again or
- A way to tell which devices are still Ack'd.

Who is online

Users browsing this forum: No registered users and 7 guests