Sergio,

No idea is ever stupid....but I think you may be up to something.

My WUP system is currently running WUP and Denika(MRTG) reports on the same server.

I cannot move Denika reports out of the WUP server until my new fiscal yr begin.

So the my WUP server is doing a lots of work....I think this problem with maintenance periods  may only happens on servers that are this kind of busy...just a tought.

But since Blackout periods is working flawless...I will keep on using it until I can move some of the load out of the WUP server.

Thanks

Hi,

same problem here. I have the database on a separate SQL server and for most of the devices, I also get those up-messages after maintainance even though they are dependent on down-states.

However, I tested it with both WUP and the database on one laptop and everything works just fine. I'm curious to find out if that has something to do with it.

I also informed Ipswitch tech support about this, if they can help me out I'll post it here.

Ramin

Yup, interesting point here ! I must say that I have deployed until now 4 different WUP setups and on all of them the sql server is on the same system.

Could it be now that there is a time difference between the WUP system and the SQL system, and that somewhere WUP is not consistent with its time management -working sometimes with the WUP system time and other times with some timestamp that the SQL server would have ?

Sergio,

on your WUP deployments are you using the Default SQL from Whatsup or are you using MS SQL 2000?.

In my case the SQL database is running on MS SQL 2000 on the same server...just wondering if the type of Database has anything to do it and the load on the server too.

Regards

All four are using :

* Hp Proliant ML350 with Xeon 3.xxGhz;

* 1 Gig Ram;

* Windows 2003 Standard edition;

* Raid-1 array for system, Raid-5 array for data;

* MSDE 2000.

The setup seems pretty overkill but the reason for that is that we are planning to do lots of data acquisition (history graphs for snmp values etc) in the near future. I will then probably migrate to Sql server but currently MSDE is more than enough. Besides, hardware is so damn cheap these days that I don't hesitate to oversize my systems...

The biggest tweak I made on the setup is that, since I have plenty of Ram, I set MSDE so that it can use 256MB ram instead of the default 64 that WUP sets.

Of these four setups, the biggest one monitors approx 500 devices, with a total of about 2000 monitors (snmp active monitors, nt services, snmp traps). Even with that, ram usage is quite low and cpu only spikes a little bit when I ask for reports, otherwise the system is pretty much idle.

I don't think it has anything to do with a time mismatch or the server being too busy. Because it only happens to certain devices.

I deleted one of those devices that sends up messages after maintenance mode and recreated it. After that it worked just fine. I'm guessing that for devices I created at a certain time, some flag wasn't set correctly and now it has a problem with the down dependency on the up.

I just did this with another device that misses a poll every once in a while, doesn't give me an up message then, but does give me an up after maintenance and it also fixed the problem for that device. Can anyone else reproduce this behavior?

I'm also wondering if anyone that has the database on a separate SQL server tried reinstalling WUP leaving the original database intact and still has this problem.

Ramin

This problem will be resolved in SP1a, when that is actaully relased.....it is in beta testing right now...

I was reading some notes about SP1a on this forum....and this is one of the problem that will get fixed hopefully by that release. I switched to Blackout for the time being.

Regards