[PacketRat]

I have had an issue with one of my sites having blocked threads every 2-3 days which is causing email to stop flowing. Was wondering what everyone was doing to monitor your BES's. Anyone have the router service stoping and restarting using a script?

thoughts?

thanks

[acnst]

Do you use a tool like Microsoft Operations Manager? This allows you to get notifications when certain events appear in the eventlog. Beside this there tools tailored just for monitoring, e.g. www.bnator.com - Home.

[hookedOnBB]

You can configure the BES to have the controller restart the agents based on the number of health check failures ( wait counts).
To monitor the BES and users for hung threads and other issues you can try out BoxTone.

[acnst]

Before you buy any new product - if you do not have any yet in place - I would recommend to take enough time to evaluate the products and check which one fits your needs best.

[J.W.]

I'm going to see how Servers Alive does with it. It has the ability to send to SMS, so if BES craps out, I still get the alert.

I'll let you know how it goes.

[Krissy]

I use perl to back up the event log every 10 mintues then, I wrote a script to read the file and export/email the results if a string is found (i.e. a certain error code). It's been working pretty well for me, and works much more efficiently than BES alert because it really doesn't hit my servers resources and I can run it any machine if I want as long as it has access to my POA.

EDIT - I could really use this same idea to send via SMS as well. I think I might add that in today. Thanks for the Idea JW ;)

[Khue]

I have MOM monitoring specific services that I've noticed trends with as well as SQL alerts for DB issues. MOM can kill and restart services and then restart the BB specific services. Pretty easy to setup once you spend 9million hours figuring out MOM (not a big fan of it).

[TheBesWizard]

What specific entries are you looking for in the event logs and/or log files. It seems that many of you have your own home grown solution to stability problems and I would like to compile a list of just what you are doing and what problems are occuring.

tbw in waterloo

[mahoward]

I have an hourly script that runs off my secondary workstation, which grep's the number of "NoteID.*DELIVERED" lines (Domino) in the current day's log file, then compares that number with the number from an hour prior. If there is no change then I get a page. This is a down and dirty way to determine if something is getting in the way of message delivery, without having to rely on discerning hung threads, etc., which never quite worked right for me anyway.

I disable the BESAlert service everytime I install/upgrade any of my servers. Because of that I avoided the entire "BESAlert" taking up 100% CPU issue awhile back...

[#james.power]

We use NetIQ. They have an additional add on designed for monitoring BES specifically. It monitors anything from your SRP connection to pending mails

[Rich Y]

How do you find NetIQ... I've been testing an evaluation install, and find it quite complicated?! What are the scripts you find most usefull?

[qc_metal]

Originally Posted by Khue

I have MOM monitoring specific services that I've noticed trends with as well as SQL alerts for DB issues. MOM can kill and restart services and then restart the BB specific services. Pretty easy to setup once you spend 9million hours figuring out MOM (not a big fan of it).

Khue - I hear you there... I think they should have called it 'MOM in law'. It is about that frustrating to work with.

In any case, we are using Intellipool Network Monitor here (link-> Network Monitoring, Server Monitoring, Performance Monitoring - Intellipool). Easy to use, and the price is very good. Also, I think there is OpManage (link-> Network Monitoring Software - ManageEngine OpManager), which can monitor up to 20 devices for free.

Oh yeah... I almost forgot, TOTALLY free monitoring tool:

The Dude (I don't know what is up with the name...) - link-> MikroTik Routers and Wireless

Rob

[hookedOnBB]

Originally Posted by mahoward

I have an hourly script that runs off my secondary workstation, which grep's the number of "NoteID.*DELIVERED" lines (Domino) in the current day's log file, then compares that number with the number from an hour prior. If there is no change then I get a page. This is a down and dirty way to determine if something is getting in the way of message delivery, without having to rely on discerning hung threads, etc., which never quite worked right for me anyway.

I disable the BESAlert service everytime I install/upgrade any of my servers. Because of that I avoided the entire "BESAlert" taking up 100% CPU issue awhile back...

That is nifty way of checking for issues. An improvement would be to check it against the messages delivered the same hour every week. This would give a even better indication of issues. We have seen that email tends to follow an hourly pattern for any given day. Friday 3-4 pm patterns are entirely different from Friday 4-5 pm.