We have had a long standing issue that is now at bubbling point as we have been caught out a few times by clients.
the scenario is that we have a template for servers that deploys standard monitor sets - disk space / memory / cpu etc.
These all get deployed no problem but when we check monitorlog it might show one counter only and nothing from any other monitor set.
sometimes the solution is to run an update scan by list (which we are now having to do near weekly) or sometimes it has been to update the agent.
the issue we have is that we don't know till the client / engineer raises it as "did you know we only have 1% free on this disk"
When we check monitor sets all are applied as they should be but then in monitor log nothing or perhaps as I said a random one or two counters.
The other issue we had the other day was one of the drives was sat there saying waiting for data but C drive was fine, in this case the D drive had run out of space.
We have over 500 servers being monitored so I don't have hours to be sat checking each one to see "is the monitoring working"
Does anybody else have this issue or have seen it and got a fix for it?
Do people have a way to manage the uncertainty that kaseya monitoring presents?
any other commetns welcome :o)
Oh and kaseya know about my issue and while they feel my pain have no solution for me.
Could it be a server performanc issue?
I must be all alone on this one - I find it hard to believe...
I have seen similar things, but normally I am able to get it going. I think it is down to the different monitors that the different editions of windows use. Whenever I apply a new monitor set I check ten minutes after to make sure it is picking something up.
I'll have to wait until I see it again before I can post on how I got it working, if it is even the same issue.
I'm very dubious over how the monitor sets actually work and am very cautious around them. As an example I have a set that should warn me if disk space drops below 15% - Object = Logical Disk, Counter = % Free Space, Instance = *ALL etc. When I look at the monitor that's been assigned to the machine it is looking at C, D, E, F & G. Which never relates to the actual disks in the machine it's looking at so most counters are sitting - Waiting for data ...
This in turn means I would need to setup individual sets depending on each machines configuration and as I have nearly 3000 servers it's just not practical.
What I have done to check that the sets are actually working is use one of my sandbox servers and put the thresholds so that it shoudl show me something and in the main it does, I'm just concerned with the amount of extra work I need to put in.
We have a SharePoint solution that interrogates the data from the ksubscribers db and we use that for disk space as it's usually pretty reliable providing that we are getting regular audits.
Nothing worse than when a customer tells you that they have a problem that you should have spotted though.
@Michael - If the machine does not have a C,D, E, F or G drive, Kaseya will ignore those drives and only monitor the drives the machine has. This was actually an enhancement from the v5 product and was added precisely so you wouldn't have to create multiple monitor sets for each drive combination. In other words, in the old version of monitoring if you assigned a monitor set that monitored C,D & E to a machine that only had C & D, it would constantly alarm on the E drive. In this version, Kaseya will automatically ignore E because it's not on the machine.
The problem we have is that sometimes the E: for a server is fixed-disk that needs to be monitored, othertimes not. Maybe E: for me is a removable drive that is always full? Maybe it isn't anything ... until I plug in my thumb drive in that only has a couple megs of free space and I send off alarms. Or maybe E: on another server is some sort of recovery or utility partition that has little to no free space. Not sure how it would react with CDs ... or mapped drives to servers that have quotas ... etc. etc.
Is there any way to tell Kaseya "Only look at my fixed disk drives"? I realize we could customize the monitor set in these situations, but that is another pain in the rear.
Max the issue here is not the c / d / e issue - This is normally resolved by an update scan by list.
our issue is that we apply a load of monitor sets - so say Monitor Disk queue lengths / Disk space C drive / Exchange information store started etc.
So we apply it fine via template but then we discover that when you click on monitor log it only shows maybe one of the counters instead of the 10 you applied
We are now running update scan's by list every week to try to alleviate the issue.
our main point with Kaseya (support etc) has been - we can fix it if we know about it but without knowing about it we are screwed.
We have over 500 servers currently being monitored - on average each server would have at least 15 individual items being monitoring from counters to services to processes so how do we police that these are all working.
We also have seen the "Waiting for Data" but this would be on valid counters / disks - this happened last week when a servers D ran out of space - in Kaseya under monitor log - Disk space Drive D - waiting for data.
I've had issues where I go in and the monitor sets are still applied but have not been collecting any data. I usually catch this when something goes wrong and I want to review the logs, and *poof* nothing is there. It's really sporadic and I haven't been able to reproduce reliably, I also haven't submitted a ticket yet but will when I get some time.