opmanager | Enterprise IT Management, Network performance management, IT Servicedesk, Desktop Management, Datacenter Management, Server Management, Log Analysis and Security Management, Network Tools, ManageEngine Blogs

Quite often we get this query from our evaluators and customers. We already have this data exposed in the OpManager knowledgebase; however I thought it would be nice to do a blog on it as well.

To analyze bandwidth consumption, we ran OpManager in an isolated environment and monitored the devices as listed below

1. 100 Servers (includes service monitoring like Web, MySQL, FTP, SMTP, HTTPS)
2. 3 Switches (each had 24 Ports).
3. 3 Routers (each had 5 Interfaces).
4. 2 Firewalls (each had 3 interfaces).
5. 2 Printers

Servers are monitored through WMI with default monitors (CPU, Memory & Disk) associated.

Monitors like CPU, Memory and Interfaces Rx & Tx Traffic/ Utilization/ Errors/ Discards are associated to Network Devices (like Switches, Routers, Printers and Firewalls) by default.

Monitoring Interval is set to 1 minute for all monitors and the status polling is also set to 1 minute.

Here is the report for you...

UDP Traffic: 5225 bps

TCP/ WMI Traffic: 39435 bps

ICMP Traffic: 1980 bps (18 bps per device)

So, OpManager generates a total traffic of approximately 47Kbps within the LAN in order to monitor 100 Servers, 3 Routers, 3 Switches, 2 Firewalls and 2 Printers.

Hope you find this information useful. 

One more info…Recently, we brought out a new SNMP tutorial for beginners.  Do check out the animated gifs to explain SNMP communication for various commands like GET, TRAP and INFORM.

- Kalvin

We found something interesting with our network monitoring software - OpManager. Last week one of my colleagues, who was working on a customer's issue found that SNMP requests sent from ManageEngine products did not get any responses. We did all the basic steps - verifying whether SNMP is enabled on the devices, SNMP credentials etc. - and found everything was perfect. We then tried to send SNMP requests from our MIB browser, a SNMP troubleshooting tool, but no luck. The customer had a MIB Browser from another vendor and we tried sending the SNMP requests from that MIB browser and we got the response from the device.  After spending sometime on the issue, we found the SNMP requests sent from OpManager were getting dropped at the server itself (failed to reach the devices).

That reminded us of the few issues we had with the firewall and anitivirus. We checked the firewall and antivirus on the server. Firewall was disabled, but Symantec antivirus was running. Once we stopped the antivirus, we could receive responses from the devices for the SNMP requests sent from OpManager server. We were bewildered, at the same time very eager to know about why SNMP requests sent from ManageEngine applications are blocked, when the same from the other MIB browser could pass through. We analyzed both the MIB Browsers in-depth. The difference was the platform that they were built on. Our's were built on Java, while the other one was built on .Net. We did some quick research and found that antiviruses block SNMP requests when sent from Java based applications.

Have you folks come across the same? Feel free to share your experiences and thoughts.

This is the second time we had an issue with antivirus. The first one was with McAfee. It didn't allow MySQL of OpManager 8 to start. We did some changes in the McAfee policies to get it working. Please check this link if you haven't checked it before. If you face any such issues, try disabling the firewall and anti-virus during startup which should work.

Pravin

Quite a few cases of outages in the recent past: the Gmail outage caused by overloaded servers, Microsoft Sidekick’s data loss from a faulty server upgrade, Twitter’s DoS attacks etc.. Most of these cases could’ve been prevented if a more cautious approach was taken, the top guy considering all chances of failure and setting up backup facilities just in case something does go wrong. But then one can never be 100% disaster-proof.

It is times like these that IT teams truly appreciate a network fault management system having an integrated helpdesk:

• Quick access to device configuration records & knowledge base
• Automatic creation and assigning of trouble-tickets
  
• Immediate information on affected business services etc.

“I used a lot of products in the past but I don’t remember a product that was as useful as ServiceDesk Plus to administer all the problems that I have in my network and also the OpManager.. Sometimes when I have a server that’s unavailable, I can act before a user lets me know that I had a problem.” Jesse Gusmao Ferreira, from Hospital Samaritano, Sao Paulo, Brazil.

How many of you have known it the minute your network monitor loses network connectivity? Or worse, how many of you knew through end user complaints (I’m sure in no mild words!)?

OpManager now sends an SMS to your mobile phone when the hosting server loses network connectivity. You also get an SMS alert when the OpManager server regains connectivity. Set-up is very simple - just key in your mobile number when configuring the SMS server in the Admin tab. Here's the list of supported modems for SMS notifications.



SMS on losing connectivity: “OpManager has detected that it has lost network connectivity and has suspended all monitoring. Monitoring will be automatically resumed once connectivity is established.”

SMS on re-establishing connectivity: “Network connectivity reestablished. Restarting Network monitoring services”

PS: You can also have two OpManager installations set-up in a fail-over configuration to ensure an always-monitored network environment. Check out OpManager's Network Monitoring Failover and Failback

Happy New Year everybody! Team OpManager wishes you all success and happiness throughout 2009.

We're all ready to talk network management after that hiatus on our blogs. Truth is we're suddenly finding ourselves doing a lot of work n a lot less of talk. Is that related to the recession? The papers, television, internet, radio - everyone's onto the recession. I can't wait for it all to get over - the pretty faces on TV look so sad when presenting the news!

Well, 2008 has been a good year for OpManager. A year that's made us put in the extra sweat to bring out additions in the product, in line with our goal of becoming THE complete IT infrastructure monitoring platform. The Distributed Edition of OpManager, now with VMware monitoring included, found customers within a month of product launch. We pushed ourselves in the network monitoring front by adding traffic flows monitoring and VoIP performance monitoring. 2009 is even going to see greater efforts spent in making the network admin's life a whole lot easier.

I'll sign off for now with a bit on OpManager 8 that's coming out very soon. OpManager 8 is sure to make network management a piece of cake with a number of options to customize the product to your monitoring needs. You just have to get a hold of it when it's out.

Thank you 2008. And 2009, here we come!

2008 - Highlights:

ManageEngine Euro Roadshow
Sept 8th - October 3rd.
Denmark - Sweden - France - Italy - Germany - Portugal - Poland - UK

We at ManageEngine are always game for opportunities that allow us to meet with the people who care about IT. This time round we thought we'd do something big to spend quality time with our customers, users, partners and any soul who cares about Networks, servers, applications, systems, the works. We would like to welcome you all to our ManageEngine Euro Roadshow conducted with the help of our partners throughout Europe from the 8th of September to October the third.

Check out the dates here: http://manageengine.adventnet.com/euroroadshow/agenda.html
Our speakers: http://manageengine.adventnet.com/euroroadshow/speakers.html
You can register here: http://manageengine.adventnet.com/euroroadshow/registration.html

We very look forward to meeting you!

Retail price of electricity in California is 11.53 cents per kW hour. If you run a Vostro 200 Mini in CA, fitted with Intel Pentium Dual Core E2160 processor and a 17” monitor non-stop for a year, you would burn 12264* kW hours. This will cost you $1414. Assume you have a customer in CA with 500 such Vostro Minis. He is going to burn $353, 505 for a five year period just on electricity for these desktops.

65Watts + 75Watts * 8760 hours / 1000 = 12264 kW hours
12264kW hours * 11.53 /100 * 500 * 5years = $353,505

The standard business hours 9AM-6PM Monday to Friday contribute just 27% of the overall available time. 
And in these 45 hours, reports indicate that, only 60% of the time the desktops are actively used. Rest goes in breaks, phone calls, meetings, and discussions. If someone enables strict power management ** for the above case, it will result in a savings of approximately $300k

 $353,505 * 0.73 = $258,058 (for non-working hours)
 $353,505 * 0.27 * 0.4 = $38,178 (for 40% idle time on office hours)
Total: $258,058+$38,178 = $296,236

*Thermal Design Power of E2160 is 65Watts. For monitors its 75Watts. 65+75*8760 hours /1000 =12264 kW hour. Refer Wikipedia for TDP of other processors.

** MSP Center enables you to implement power management across multiple customers’ right from your NOC. Checkout www.mspcenterplus.com for more details.