Difficult decision, App Manager or Op Manager
Greetings,
I'm new to your line of products as I'm currently running a demo of both your App Manager and Op Manager. More then likely I'll end up purchasing both as combined they have what I need, but I'm bummed as it really seems these two products need to be merged into one main product. They have numerous overlap BUT each product also has a unique feature that the other one is really missing. Maybe this is a little early, but here is my wish list for your future road-map of the things I like/need out of each.
1) OpManager has alert escelation. I really need this in App Manager as well. Having just an alert when something is a warning and then when it is critical is not enough. When I'm on vacation, I want our team at the office handling things first. If they do not respond within a certain amount of time, then I want to be notified by SMS page so I can hop on-line and deal with it. It appears that OpManager has this feature, but again, I don't see it in App Manager.
2) Monitor Groups: The reason I really like App Manager is that I can group together systems that form a complete SLA/Business services solution. It helps with Root Cause Analysis and it also helps to let our staff know just what business services are effected when an individual piece of the puzzle fails. Unfortunately this is not available in OpManager. I would simply do everything in App Manager, but again, it misses some of the things that OpManager has (such as specific Router reports, escalation, etc.)
3) I would really like multi-tiered monitor groups. I have seen this come up in the forums before and the question has always been "why?" The examples given were not very good as they could be solved with the current dependency options, so let me give an example where the dependency's don't work.
If I want to have a high level Monitor Group called "Internet", this will naturally be composed of numerous other groups that more then likely represent clustered hardware. For example, I would like to have a sub-group called "Firewalls" where I have a cluster of firewalls, and also a group called "ISP Links" and probably one called "Internet Routers". If any one or more of these sub-groups has a failure of one of the cluster members, the main Internet group would only be a warning. Though I had a failure of multiple equipment, Internet traffic is still running, thus just a warning. However, if one of the sub-groups went critical (means we lost a cluster all together) then we obviously lost all Internet traffic, and as such at that time the main Internet group needs to go critical.
The reason that the current dependencies don't work in this scenario is as follows... You would have to put all six of these monitors under one Monitor Group. There is no way to say that you could have up to 3 of these fail (say one Internet router, one ISP Link and one Firewall) and it would only be a warning, but if two of these failed (say both ISP Links) then it is critical. Having sub-groups of main groups allows for this. Of course none of this is available in Op Manager which is a bummer.
4) I can't find any tracking of alerts that shows what action was taken. In other words, I would really like a rolling history that shows that when a threshold was broken, and XXX action was taken (i.e. shows proof that an email was sent.) It then needs to show a history of when escelation kicked in and then took whatever action that was associated with. Basically I'm looking for a rolling log when an alert is tossed of exactly what Op Manager or App Manager has done.
5) Looking for an option to setup a good old fashion PING test without having to try an SNMP read first. It appears that if I put in an IP address, it will try SNMP and a ton of other services on the device. Even if SNMP is not found, it seems to except it by doing just a ping test. Doing just a generic PING really needs to be an option from the get-go as many times I already know if SNMP is an option, and if not, I would just like to setup a dedicated PING test without having to wait for it to time-out on SNMP and other tests. Unfortunately under the App Manager, it doesn't appear that doing a PING test is even an option as if it can't use one of its' Monitor Types, it just errors out and you can't monitor via PING. Maybe there is another solution, but as a newbie, I'm looking for the "PING" monitor type.
6) Rolling User Notes. In another product I'm evaluating of similar price, they have a nice simple "Notes" section where Admin's can type in important notes for other people that log on. I could forsee using this to have Admin's notify other Admin's of pending maintenance, or maybe a recent upgrade performed that might effect other systems, etc. Your applications do have nice sections to type in notes for alerts, etc. which works well, but it doesn't work for just the generic notes of what is happening system wide that might effect other people.
I guess that's good for now. I'll return back to my demo testing to try and figure out which product I want to use for which system based on what is most important to me, Monitor Groups, escelation, PING's, etc. Again, these two products really need to be merged into one to make on fantastic product. Seperately they are good, but together they would be GREAT!
I'm new to your line of products as I'm currently running a demo of both your App Manager and Op Manager. More then likely I'll end up purchasing both as combined they have what I need, but I'm bummed as it really seems these two products need to be merged into one main product. They have numerous overlap BUT each product also has a unique feature that the other one is really missing. Maybe this is a little early, but here is my wish list for your future road-map of the things I like/need out of each.
1) OpManager has alert escelation. I really need this in App Manager as well. Having just an alert when something is a warning and then when it is critical is not enough. When I'm on vacation, I want our team at the office handling things first. If they do not respond within a certain amount of time, then I want to be notified by SMS page so I can hop on-line and deal with it. It appears that OpManager has this feature, but again, I don't see it in App Manager.
2) Monitor Groups: The reason I really like App Manager is that I can group together systems that form a complete SLA/Business services solution. It helps with Root Cause Analysis and it also helps to let our staff know just what business services are effected when an individual piece of the puzzle fails. Unfortunately this is not available in OpManager. I would simply do everything in App Manager, but again, it misses some of the things that OpManager has (such as specific Router reports, escalation, etc.)
3) I would really like multi-tiered monitor groups. I have seen this come up in the forums before and the question has always been "why?" The examples given were not very good as they could be solved with the current dependency options, so let me give an example where the dependency's don't work.
If I want to have a high level Monitor Group called "Internet", this will naturally be composed of numerous other groups that more then likely represent clustered hardware. For example, I would like to have a sub-group called "Firewalls" where I have a cluster of firewalls, and also a group called "ISP Links" and probably one called "Internet Routers". If any one or more of these sub-groups has a failure of one of the cluster members, the main Internet group would only be a warning. Though I had a failure of multiple equipment, Internet traffic is still running, thus just a warning. However, if one of the sub-groups went critical (means we lost a cluster all together) then we obviously lost all Internet traffic, and as such at that time the main Internet group needs to go critical.
The reason that the current dependencies don't work in this scenario is as follows... You would have to put all six of these monitors under one Monitor Group. There is no way to say that you could have up to 3 of these fail (say one Internet router, one ISP Link and one Firewall) and it would only be a warning, but if two of these failed (say both ISP Links) then it is critical. Having sub-groups of main groups allows for this. Of course none of this is available in Op Manager which is a bummer.
4) I can't find any tracking of alerts that shows what action was taken. In other words, I would really like a rolling history that shows that when a threshold was broken, and XXX action was taken (i.e. shows proof that an email was sent.) It then needs to show a history of when escelation kicked in and then took whatever action that was associated with. Basically I'm looking for a rolling log when an alert is tossed of exactly what Op Manager or App Manager has done.
5) Looking for an option to setup a good old fashion PING test without having to try an SNMP read first. It appears that if I put in an IP address, it will try SNMP and a ton of other services on the device. Even if SNMP is not found, it seems to except it by doing just a ping test. Doing just a generic PING really needs to be an option from the get-go as many times I already know if SNMP is an option, and if not, I would just like to setup a dedicated PING test without having to wait for it to time-out on SNMP and other tests. Unfortunately under the App Manager, it doesn't appear that doing a PING test is even an option as if it can't use one of its' Monitor Types, it just errors out and you can't monitor via PING. Maybe there is another solution, but as a newbie, I'm looking for the "PING" monitor type.
6) Rolling User Notes. In another product I'm evaluating of similar price, they have a nice simple "Notes" section where Admin's can type in important notes for other people that log on. I could forsee using this to have Admin's notify other Admin's of pending maintenance, or maybe a recent upgrade performed that might effect other systems, etc. Your applications do have nice sections to type in notes for alerts, etc. which works well, but it doesn't work for just the generic notes of what is happening system wide that might effect other people.
I guess that's good for now. I'll return back to my demo testing to try and figure out which product I want to use for which system based on what is most important to me, Monitor Groups, escelation, PING's, etc. Again, these two products really need to be merged into one to make on fantastic product. Seperately they are good, but together they would be GREAT!