Community discussions

MUM Europe 2020
 
complete2006
Member Candidate
Member Candidate
Topic Author
Posts: 235
Joined: Tue Feb 07, 2006 7:18 pm

Problems are back since actual beta

Thu Apr 19, 2007 2:28 pm

After updating to actual beta the problems of unreal outages are here again. I see that a probe is down for example "ping" or "dns". It stays down for hours until I do a manual reprobe. What is wrong?
 
complete2006
Member Candidate
Member Candidate
Topic Author
Posts: 235
Joined: Tue Feb 07, 2006 7:18 pm

Thu Apr 19, 2007 9:32 pm

Many devices are red,red,red. But they are still alive! There is a bug!

Is nobody here who has the same problems?
I got the feeling that I am on the MT blacklist.
@Normis, Uldis: Am I not worth to answer?
 
firebat
Member
Member
Posts: 390
Joined: Mon Apr 11, 2005 8:38 am

Thu Apr 19, 2007 10:46 pm

Yes, same problem here. :(

I was asking in another thread of how to send notifications for elements when only all probes are down. I added a second probe to my elements in addition to Ping to query for specific OIDs so that if the ping failed, the unit wouldn't be flagged as down. Still awaiting and answer. The is with 2.0 R12 I was going to upgrade to v3 Beta5 but there are a lot of functions to tweak as a result of the changes made so I have been holding off. Right now, I turned off email notifications as I just get flooded with emails for unreal outages.
 
uldis
MikroTik Support
MikroTik Support
Posts: 3428
Joined: Mon May 31, 2004 2:55 pm

Fri Apr 20, 2007 11:26 am

We are unable to reproduce the problem. What is your 'Probe Timeout'
and 'Probe Interval' values for services you have this problem? What kind of
probes they are using? What are settings for probes? We appreciate your input
to help find those bugs.
 
complete2006
Member Candidate
Member Candidate
Topic Author
Posts: 235
Joined: Tue Feb 07, 2006 7:18 pm

Mon Apr 23, 2007 1:04 pm

@uldis
This issue happens on differnt probes

ping
dns
smtp
disk
cpu
..
.

It comes and it goes. Today I got no false alarms. On Friday and before I got them in mass. In the prevoius beta I thougt the problem was fixed case the phenomen was extremly rare.

I wonder what the difference is between automatic poll (which will let the probe down) and an manual reprobe (this brings the device up).

The configuration:

Windows Server 2003 (SP2) with Terminal services. The dude client is started from terminal session.

Regards
Thomas
 
winkelman
Member Candidate
Member Candidate
Posts: 235
Joined: Wed Aug 16, 2006 5:00 pm
Location: Amsterdam, The Netherlands

Mon Apr 23, 2007 5:41 pm

I've had such a thing happen on my network when there where network problems.

It turned out that one Token Ring device (connected through a bridge to the Ethernet-network) was faulty and generating 'beacon errors'. In turn, this resulted in a high number of broadcasts on the Ethernet network. While this only caused some small problems on the Ethernet network, it made The Dude completely flip.

Perhaps you have got some device on your network that is causing 'broadcast storms', this seems to cause the Dude to flip-out.
 
firebat
Member
Member
Posts: 390
Joined: Mon Apr 11, 2005 8:38 am

Mon Apr 23, 2007 11:40 pm

My issue is only with Pings failing. All other probes are OK. I have an L3 network and the problem is happening on all subnets. I even see the issue with a UPS that is directly connected to the same MT router as the NMS (different port/subnet...same router....NMS and UPS are the only device on the two subnets).

Who is online

Users browsing this forum: thalesramos60 and 9 guests