Answering inline: On Aug 13, 2009, at 10:26 AM, msarro wrote:
> Greetings everyone! > We're currently looking into monitoring solutions for our particular > problem. We're a windows shop. Currently we're using a system called > Next9, however there are a number of issues. Any new hardware needs > to be implemented by their developers, at expense to us. Further > their licensing is prohibitively expensive. > > We currently have 40 customer sites, but are growing on average 1-2 > sites a week. Our SLA stipulates that we have to monitor their > equipment, but we are quickly becoming overwhelmed doing this > manually. Our largest site deployments have Next9 implemented, the > rest have to be checked by hand. > > Our most common issues are disks filling up, servers with general > hardware issues (dead hard disk, etc - things that would be picked > up by windows logs), servers not being up-to-date, etc. We would > like to be able to monitor this automatically for all of our > customer sites. > > Now, is there a way to do this in a centralized manner (ie a single > server at our site)? For instance, suppose we have sites A through > Z. If we receive an SNMP trap from site A and site Z at the same > time, would the server at our site be able to tell that that trap > originated from site A and not site Z? Our servers follow the same > naming convention at every site. If you have network access to everything at the other locations, you can just create Locations for each site no problem. If the different sites have the same IP addresses and/or the same hostnames, you'll probably need Enterprise, which supports duplicate IPs and hostnames and allows you to roll multiple Zenoss installations into a single global dashboard. > We would also like to monitor virus scanner status if possible. > Internally we would like to be able to monitor all of our systems. If the virus scanner exposes some sort of metric you can watch (SNMP, WMI, a URL for screen scraping), then a performance template would probably work. > Would we be able to monitor tape backup status? Like, a last run > time, etc. This isn't incredibly necessary since we can tell by disk > usage, but it would be very helpful. Again, assuming the metrics are exposed you can watch disk usage and arrange events based on time of last run and missed runs. > Would we be able to monitor things like pagefile size, etc? Yup. > What requirements would the customer need to provide? Would they > need to open up SNMP? Assuming their networks aren't openly accessible, then you're looking at probably needing a VPN connection or Zenoss Enterprise. > Are there any other implementations like this, where you'd be > aggregating data from numerous customer sites - not just from the > same autonomous system? There most definitely are Core and Enterprise installations that are widely distributed geographically. Like this: http://www.flickr.com/photos/zenoss/2870405569/ > Any help you could provide would be very much appreciated. Thank you! > > We're also looking into Nagios and solar winds, however Nagi doesn't > include automatic discovery, and solar winds is extremely expensive. Thanks, Matt Ray Zenoss Community Manager community.zenoss.com [email protected] _______________________________________________ zenoss-users mailing list [email protected] http://lists.zenoss.org/mailman/listinfo/zenoss-users
