I am trying to monitor a new ESX 4 server, but it isn't being seen as a vmware server. It says it's a net snmp server. Any ideas? Do we just have to wait to monitor these servers? I am going to install 9.5 and hope...
We expect to have a Release Candidate for NPM 10 in weeks, not months. RCs can be run in production. Once we are confident of the RC (which is about quality, not time), we'll GA. So it's reallly soon, but we can't be more precise than that.
With ESX4, VMWare changed the SNMP implementation that was used in ESX 3.5. As such, ESX4 monitoring will not work with NPM 9.1 or 9.5. We are working with VMWare and looking to add support in a future release for ESX4
I must say this is a very big disappointment! I would have thought that with the vSphere release date and the NPM 9.5 release date being so close together, that we would have had this functionality in 9.5. It seems to me that being a VMware partner, this would have been something planned out better.
We are moving all our ESX servers to ESX 4 within the next month to take advantage of some of the improvements it offers over ESX 3.5. With that being said, how soon do we see this functionality again in NPM? I don’t want to have me ESX host unmonitored for long.
Unfortunantly as I mentioned in my previous post VMWare completely changed their SNMP implementation in version 4 and a lot of the data we used with 3.5 is no longer accessible with VMWare via SNMP. I would suggest opening a case with VMWare as well. We are working on resolving, but no timeline I can commit to yet.
Is there any update to a timeline for ESX4 support?
When will we see 4.0 support? Please give an idea.
With the purchase of Tek-Tools is it possible you will be offering their viritualization monitoring product as a stop gap measure to customers on active maintenance until you can finally finish building in the functionality? I've seen the beta and its nice but its not a GA release yet and until it is we cannot monitor some of our most critical systems properly.
I had a server crash last week becasue the datastore filled up. The VMware alerts are not very good.
We will have Orion native vsphere support very shortly. One of the goals of acquiring Tek-Tools was to go beyond what we have been doing and accelerate our virtualization management. Don't think short-term. The long term combination of technologies is the point. I can't say more than that, but stay tuned.
Hi Denny,
Will the Orion native vSphere support in Orion 10 support ESX4i as well as ESX4?
Thanks
John
Yes, it should work
Hi Denny,Will the Orion native vSphere support in Orion 10 support ESX4i as well as ESX4?ThanksJohn
I have a copy of the beta and I have ESXi 4 servers, there are some things that work and some that don't with ESXi.
VMware doesn't provide the same data in the snmp for ESX and ESXi. SolarWinds is still taking some info from SNMP.
It shows all my guests per host, the guest os, power state, the IP (if vmware tools is installed and the viclient shows it), server version, number of VMs, total RAM on host. It has charts showing the amount of RAM, Network, and Memory each guest is using.
In case it's of any use to anyone:
I've been playing around using the VIPerlToolKit to report on datastore usage (can be downloaded from sourceforge, and used with activeperl). The following script expects 3-4 parameters, Server Username Password DataStoreName Index (optional, starts at 0). You can specify the datastore via the name, or via the index (but you have to put in a dummy name). I'm calling it via APM Windows Script poller....
-----8<-----8<------
#!/usr/bin/perl use warnings; use VMware::VIM2Runtime;use VMware::VILib;my $args=$WScript->{Arguments};$ENV{'VI_SERVER'}=$args->item(0);$ENV{'VI_USERNAME'}=$args->item(1);$ENV{'VI_PASSWORD'}=$args->item(2);$ENV{'VI_DATASTORE'}=$args->item(3);if($args->count>4){ $ENV{'VI_INDEX'}=$args->item(4);}my $ok=99;my %opts = ( datastore => { type => "=s", variable => "VI_DATASTORE", help => "Datastore: Datastore name (req)", required => 1, }, index => { type => "=i", variable => "VI_INDEX", help => "Index: Datastore Index (opt)", required => 0, }, );Opts::add_options(%opts); Opts::parse(); Opts::validate();Util::connect();my $datastore_name = Opts::get_option('datastore');my $index=-1;my $count=0;if(Opts::option_is_set('index')){ $index=Opts::get_option('index');}my $views= Vim::find_entity_views(view_type => 'Datacenter');foreach $view (@$views){ #print Dumper ($view->datastore)."\n"; foreach my $ds (@{$view->datastore}) { my $obj=Vim::get_view(mo_ref => $ds); if(($index<0 && $obj->summary->name eq $datastore_name) || $count==$index) { $free=$obj->summary->freeSpace; $total=$obj->summary->capacity; $percentUtil=100-(($free/$total)*100); print "Message: ".$obj->summary->name."\n"; print "Statistic: ".$percentUtil."\n"; $ok=0; } $count++; }}Util::disconnect();$WScript->quit($ok);
I've been playing around using the VIPerlToolKit to report on datastore usage (can be downloaded from sourceforge, and used with activeperl). The followingscriptscript expects 3-4 parameters, Server Username Password DataStoreName Index (optional, starts at 0). You can specify the datastore via the name, or via the index (but you have to put in a dummy name). I'm calling it via APM Windows Script poller....
I've been playing around using the VIPerlToolKit to report on datastore usage (can be downloaded from sourceforge, and used with activeperl). The following
script expects 3-4 parameters, Server Username Password DataStoreName Index (optional, starts at 0). You can specify the datastore via the name, or via the index (but you have to put in a dummy name). I'm calling it via APM Windows Script poller....
If you're interested in monitoring how VMware (and everything else, for that matter) uses storage, you might check out the newest member of the SolarWinds' family: Storage Profiler and Virtual Profiler
http://www.tek-tools.com/virtual/endtoendmapping.php
having trouble calling this script... can you give an example of what your command line arguments look like?
Any word on this? Our VM environment is in the process of upgrading to 4.0 and everytime we move an ESX host off 3.5 I am baraged with the "Where's my stuff!?" questions.
Thanks,
Mike
Good news on this front! We are extremely close to wrapping up QA acceptance and going into RC. Stay tuned for the invite!
Do you have an approximate date? I really need to get ESX 4.0 support.
Thanks.
I'd like to get an invite if you're handing them out.
I would like to request an invite on this as well.
Thanks!
I just got notice that the RC is available to download! Yea!
I found the link to download RC earlier today, but the link is dead.
So the RC is out, however, we are doing a staged rollout in phases. If you want to get in the next round/wave, send me a private message via thwack with you SWID please.
So I guess having joined fairly recently, I can't send a PM. We'd like to get access to the RC.
I can't seem to send a PM to you. I really need to get this RC in place in the near future. We have been running blind for far to long. Please let me know what alternate method I can use to get in touch with you and get on the list.
Would you please email me a link to the RC? I can seem to private message you either.
DO NOT UPGRADE TO vSphere even with update 1! We upgraded 4 months ago and have down time every week. If my SAN has any sort of issue we lose connectivity in the entire environment every time. There are undocumented bugs every time we turn around.
DO NOT SEEK THE UPGRADE
DO NOT UPGRADE TO vSphere even with update 1! We upgraded 4 months ago and have down time every week. If my SAN has any sort of issue we lose connectivity in the entire environment every time. There are undocumented bugs every time we turn around.DO NOT SEEK THE UPGRADE
I think that's a pretty broad statement to make. I'm sorry to hear you've had so many issues, but we've been on vSphere 4 for quite a while (without U1, and now with), and have had no more issue than with 3.5. There are a lot of other factors that VMWare performance is dependant on (server and component hardware, SAN equipment and type (FIber, iSCSI?), networking hardware (including configs), and even host and guest settings, what type of server/switch/SAN redundancy do you have, etc.), and everyone's design has their own signature. I truly hope VM support gets your problems resolved, though, as I know first hand how much an impact a VM interruption can have!
Sorry to hear, FWIW we've had a lot of success with 4.0 and 4.0u1 ... FWIW, my experience with VMware tech support has been very good. Depending on your environment, SAN interruptions can cause a lot of issues with clustered ESX configurations. A production SAN shouldn't be experiencing intermittent issues routinely...
I really like vSphere 4, one thing that may help is to review the changes in the way shared storage has changed between 3.5 and 4, especially for iSCSI. There are some documents out there that will help you with your multipathing if you are experiencing issues with your SAN and losing connectivity to your VMDK's.
