I used to be on a team managing hundreds of 2k3-2k12 boxes and the ones on 60 gig C: drives were always a pain with about 15 megs left at any time.
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
If I could figure out how to make linux active directory controllers I'd go that route, but they're a pain in the ass to manage, I'd rather stick with windows for all of that
not a doctor, not a lawyer, examples I use may not be fully researched so don't take out of context plz, don't @ me
We deploy a bunch of Dell zero clients for users to access VDI desktops (horizon).
My question is, can some module of Solarwinds track these assets?
Can SCCM?
I haven't dealt with SCCM in years and don't have much access to our SCCM or Solarwinds. Our infrastructure provider is saying there is no way to asset track these zero clients outside of hardcopy records and spreadsheets, which sounds suspect to me at best.
Do the boxes have a dedicated MIB, does this MIB contain some sort of unique identifier? If so, then yes! NPM will do that.
If not, do they have a plain text configuration file you can download through SSH or something? If yes, then NCM
If not, do they have a web interface/web api you can query? If so, then SAM
If not, then sort of, you can use their MAC addresses as their identifier and add a custom property for some sort of internal asset number? Then you are still looking at NCM
This will monitor that they are up and down, their IP addresses that that is probably that. You could also add in custom properties for like, warranty date and such if that isn’t automatically handled by SAM or NCM (SAM handles warranties for Server platforms, NCM handles EOL etc. for network devices)
If you want to know where they are on your Network and the users on each over time then you could also add in UDT.
So, the answer is; what do you want to track and what can the devices provide?
Apothe0sis on
0
Apothe0sisHave you ever questioned the nature of your reality?Registered Userregular
Yea, 100 is what I use for Windows now as well. I have some of my physical boxes here on 60 and it's a magic trick keeping those things from being full.
We deploy a bunch of Dell zero clients for users to access VDI desktops (horizon).
My question is, can some module of Solarwinds track these assets?
Can SCCM?
I haven't dealt with SCCM in years and don't have much access to our SCCM or Solarwinds. Our infrastructure provider is saying there is no way to asset track these zero clients outside of hardcopy records and spreadsheets, which sounds suspect to me at best.
Do the boxes have a dedicated MIB, does this MIB contain some sort of unique identifier? If so, then yes! NPM will do that.
If not, do they have a plain text configuration file you can download through SSH or something? If yes, then NCM
If not, do they have a web interface/web api you can query? If so, then SAM
If not, then sort of, you can use their MAC addresses as their identifier and add a custom property for some sort of internal asset number? Then you are still looking at NCM
This will monitor that they are up and down, their IP addresses that that is probably that. You could also add in custom properties for like, warranty date and such if that isn’t automatically handled by SAM or NCM (SAM handles warranties for Server platforms, NCM handles EOL etc. for network devices)
If you want to know where they are on your Network and the users on each over time then you could also add in UDT.
So, the answer is; what do you want to track and what can the devices provide?
I haven't found a MIB yet no, but I'm still looking.
Don't think so, but hadn't though about that angle, will investigate.
They do have web interfaces, but it basically spits out the IP/MAC and lets you do firmware upgrades.
I don't even care about status, we are just looking for a list. It looks like UDT will track the MAC of any device connected to a monitored switch, so I think that route will actually get us enough. We really just want a list to check off vs what accounting has in their fixed asset database (because they suck at keeping up with changes we submit to them and instead of making them better, IT of course gets to change our process in order to make up for someone's shortfall - hell today walking in the door I was interrogated about if Friday was a holiday and why it wasn't, because IT here means general question answer person).
I tried using network discovery against known IPs but while it does find the node, it can't talk because the zero doesn't have WMI or SNMP (but I will dig some more today for a MIB).
EDIT: I managed to get network discovery to find the zero clients (after manually changing SNMP on each zero to match what is allowed on our network (as opposed to the default crap). It does't really produce accurate results, but we can get the MAC address of each from the interface details, so it's good enough for me. Thanks for the info.
Most major vendors have some sort of lights-out management. HP has iLO, Dell has iDRAC, Fujitsu has IRMC.
Dell's is by far the best.
Would buy again.
Genuine question: why would you think this? iDRAC is an over underengineered dumpster fire with half implemented automation options.
iLO actually has decent support in Ansible and doesn't require a server reboot for every single change.
What? iDRAC doesn't require server reboots. You can change it freely and it never reboots the actual server. Shit, you can push out full version firmware updates and not reboot the server. The DRAC itself will restart, but not the server. It's literally 1 click upgrading without server reboots.
With all the functionality the Enterprise version gives you, along with the fact that it ties in to OpenManage, and OpenManage has vCenter Plugin, I don't see what the problem with iDRAC is. Have you used an iDRAC 7 or 8?
How many CPUs/Cores and how much memory should I dedicate to my MySQL server? It's currently storing about 6 gigs (it is growing quickly though), it seems to tack on about 2-3 gigs a year. (Do you think 60 gigs of hard drive is good too?)
Rule of thumb for MSSQL if you have the resources is 1:1 memory to database size, 1gig for the OS, 1gig for the SQL application itself. I don't know how well that translates to MySQL. If you have a total of 6 gigs of database, I would give it 8-12 gig of ram to account for growth.
Also make sure you edit the settings of your VMs and enable CPU and RAM hotplugging. That lets you add cpus and ram to the VM on the fly without a reboot.
Since you were talking about remote access earlier, you can do some port redirection on your firewall to give you access to your iDRAC. And the SSL HTML5 interface for ESX can be made public facing too.
How many CPUs/Cores and how much memory should I dedicate to my MySQL server? It's currently storing about 6 gigs (it is growing quickly though), it seems to tack on about 2-3 gigs a year. (Do you think 60 gigs of hard drive is good too?)
Rule of thumb for MSSQL if you have the resources is 1:1 memory to database size, 1gig for the OS, 1gig for the SQL application itself. I don't know how well that translates to MySQL. If you have a total of 6 gigs of database, I would give it 8-12 gig of ram to account for growth.
Also make sure you edit the settings of your VMs and enable CPU and RAM hotplugging. That lets you add cpus and ram to the VM on the fly without a reboot.
Since you were talking about remote access earlier, you can do some port redirection on your firewall to give you access to your iDRAC. And the SSL HTML5 interface for ESX can be made public facing too.
way ahead of you broskies
not a doctor, not a lawyer, examples I use may not be fully researched so don't take out of context plz, don't @ me
IIS redirects are so fucking soul crushing and time consuming to do vs nginx/apache/anything else. Why microsoft? Why does your GUI have to be such shit and direct editting your config files always ends in disaster due to your retarded formatting rules? Related, my entire day today will be doing redirects
Instead of actually doing project planning like I really need to be doing, today I am documenting/reporting everything we have/haven't tried to do to fix outlook authentication popups because it happens once every like 5 months to one user and this time it happened to an executive and news made its way to the wrong place and now it is being made out that this is a daily occurence and our exchange is broken.
So, gathering ticket info, documentation, steps taken, working with the help desk guy on putting everything into a report.
This is almost as bad/dumb as the domain controller almost blowing up last night.
IIS redirects are so fucking soul crushing and time consuming to do vs nginx/apache/anything else. Why microsoft? Why does your GUI have to be such shit and direct editting your config files always ends in disaster due to your retarded formatting rules? Related, my entire day today will be doing redirects
IIS is like
the upper management is pretty sure they deprecated it
but there's one guy left maintaining it
he actually got fired like 15 years ago, but is still getting paid from a HR error, Office Space style
no one's sure what building he works out of so IIS's corpse shambles on
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
I'll move this to a new thread if it becomes too much of a diversion, but - any of you guys manage an AWS organization? I'm trying to set things up for a startup company I'm a co-founder of, and I'm kind of completely confused between their accounts vs IAM users with respect to incorporating policies, would be useful to ping some hopefully simple questions
If anyone has a good, 'explain this to me like I'm 5' tutorial link I'd accept that as well
Param (
[string]$Computer = (Read-Host Remote computer name),
[int]$Days = (Read-Host Number of days to search)
)
cls
$Result = @()
Write-Host "Gathering Event Logs, this can take awhile..."
$ELogs = Get-EventLog System -Source Microsoft-Windows-WinLogon -After (Get-Date).AddDays(-$Days) -ComputerName $Computer
If ($ELogs)
{ Write-Host "Processing..."
ForEach ($Log in $ELogs)
{ If ($Log.InstanceId -eq 7001)
{ $ET = "Logon"
}
ElseIf ($Log.InstanceId -eq 7002)
{ $ET = "Logoff"
}
Else
{ Continue
}
$Result += New-Object PSObject -Property @{
Time = $Log.TimeWritten
'Event Type' = $ET
User = (New-Object System.Security.Principal.SecurityIdentifier $Log.ReplacementStrings[1]).Translate([System.Security.Principal.NTAccount])
}
}
$Result | Select Time,"Event Type",User | Sort Time -Descending | Out-GridView
Write-Host "Done."
}
Else
{ Write-Host "Problem with $Computer."
Write-Host "If you see a 'Network Path not found' error, try starting the Remote Registry service on that computer."
Write-Host "Or there are no logon/logoff events (XP requires auditing be turned on)"
}
It prompts you for a remote host name and a number of days back you want to search and then dumps out a sort-able table of date/time and usernames for logon/logoff events on that host:
I'll move this to a new thread if it becomes too much of a diversion, but - any of you guys manage an AWS organization? I'm trying to set things up for a startup company I'm a co-founder of, and I'm kind of completely confused between their accounts vs IAM users with respect to incorporating policies, would be useful to ping some hopefully simple questions
If anyone has a good, 'explain this to me like I'm 5' tutorial link I'd accept that as well
I hesitate to outright speak for anyone here, but I don't remember anyone who frequents the thread doing any AWS stuff.
They have an IAM intro here and if this video is any indication, it looks like it's trying to be pretty ELI5
HOW AM I ONLY NOW LEARNING ABOUT Out-GridView?????????
THIS SHIT WAS IN POWERSHELL 2????
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
you'd thing it was like a different version of Format-Table
not a whole fucking GUI interface
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
Ticket received from a client just now, on Thursday afternoon:
We are upgrading <software> on <server> on Friday evening. This will require an update on all of the client machines once it's been completed. We will need this done prior to start of business on Monday morning.
I hate when someone asks me how to fix something in Excel and on the outside I'm nodding along and on the inside I'm like "Are you fucking shitting me, I didn't even known Excel could DO that"
I hate when someone asks me how to fix something in Excel and on the outside I'm nodding along and on the inside I'm like "Are you fucking shitting me, I didn't even known Excel could DO that"
If I had a dollar for every time that happened to me I'd be retired.
I'm lucky in that our boss will back us up on people bringing their excel bullshit to us.
Oh your 2 gig spreadsheet with links to 100 other spreadsheets isn't working right and you want us to fix it?
Hmm let me check, yup, making excel spreadsheets is literally your entire job, this one's on you buddy.
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
Posts
I used to be on a team managing hundreds of 2k3-2k12 boxes and the ones on 60 gig C: drives were always a pain with about 15 megs left at any time.
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
If I could figure out how to make linux active directory controllers I'd go that route, but they're a pain in the ass to manage, I'd rather stick with windows for all of that
XBL:Phenyhelm - 3DS:Phenyhelm
Do the boxes have a dedicated MIB, does this MIB contain some sort of unique identifier? If so, then yes! NPM will do that.
If not, do they have a plain text configuration file you can download through SSH or something? If yes, then NCM
If not, do they have a web interface/web api you can query? If so, then SAM
If not, then sort of, you can use their MAC addresses as their identifier and add a custom property for some sort of internal asset number? Then you are still looking at NCM
This will monitor that they are up and down, their IP addresses that that is probably that. You could also add in custom properties for like, warranty date and such if that isn’t automatically handled by SAM or NCM (SAM handles warranties for Server platforms, NCM handles EOL etc. for network devices)
If you want to know where they are on your Network and the users on each over time then you could also add in UDT.
So, the answer is; what do you want to track and what can the devices provide?
the "no true scotch man" fallacy.
I only sleep on a 700 thread count. anything less and I might as well be sleeping on sandpaper.
it is so good
I haven't found a MIB yet no, but I'm still looking.
Don't think so, but hadn't though about that angle, will investigate.
They do have web interfaces, but it basically spits out the IP/MAC and lets you do firmware upgrades.
I don't even care about status, we are just looking for a list. It looks like UDT will track the MAC of any device connected to a monitored switch, so I think that route will actually get us enough. We really just want a list to check off vs what accounting has in their fixed asset database (because they suck at keeping up with changes we submit to them and instead of making them better, IT of course gets to change our process in order to make up for someone's shortfall - hell today walking in the door I was interrogated about if Friday was a holiday and why it wasn't, because IT here means general question answer person).
I tried using network discovery against known IPs but while it does find the node, it can't talk because the zero doesn't have WMI or SNMP (but I will dig some more today for a MIB).
EDIT: I managed to get network discovery to find the zero clients (after manually changing SNMP on each zero to match what is allowed on our network (as opposed to the default crap). It does't really produce accurate results, but we can get the MAC address of each from the interface details, so it's good enough for me. Thanks for the info.
What? iDRAC doesn't require server reboots. You can change it freely and it never reboots the actual server. Shit, you can push out full version firmware updates and not reboot the server. The DRAC itself will restart, but not the server. It's literally 1 click upgrading without server reboots.
With all the functionality the Enterprise version gives you, along with the fact that it ties in to OpenManage, and OpenManage has vCenter Plugin, I don't see what the problem with iDRAC is. Have you used an iDRAC 7 or 8?
Rule of thumb for MSSQL if you have the resources is 1:1 memory to database size, 1gig for the OS, 1gig for the SQL application itself. I don't know how well that translates to MySQL. If you have a total of 6 gigs of database, I would give it 8-12 gig of ram to account for growth.
Also make sure you edit the settings of your VMs and enable CPU and RAM hotplugging. That lets you add cpus and ram to the VM on the fly without a reboot.
Since you were talking about remote access earlier, you can do some port redirection on your firewall to give you access to your iDRAC. And the SSL HTML5 interface for ESX can be made public facing too.
way ahead of you broskies
I prefer the thick client to the web interface in general, but the HTML5 interface is leaps and bounds better than the flash.
So, gathering ticket info, documentation, steps taken, working with the help desk guy on putting everything into a report.
This is almost as bad/dumb as the domain controller almost blowing up last night.
IIS is like
the upper management is pretty sure they deprecated it
but there's one guy left maintaining it
he actually got fired like 15 years ago, but is still getting paid from a HR error, Office Space style
no one's sure what building he works out of so IIS's corpse shambles on
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
Well seeing as I have no route into that network at all. . .
XBL:Phenyhelm - 3DS:Phenyhelm
Just like, plug a cable in. Problem solved, right?
No, no it does not.
the "no true scotch man" fallacy.
XBL:Phenyhelm - 3DS:Phenyhelm
fix'd
If anyone has a good, 'explain this to me like I'm 5' tutorial link I'd accept that as well
Param ( [string]$Computer = (Read-Host Remote computer name), [int]$Days = (Read-Host Number of days to search) ) cls $Result = @() Write-Host "Gathering Event Logs, this can take awhile..." $ELogs = Get-EventLog System -Source Microsoft-Windows-WinLogon -After (Get-Date).AddDays(-$Days) -ComputerName $Computer If ($ELogs) { Write-Host "Processing..." ForEach ($Log in $ELogs) { If ($Log.InstanceId -eq 7001) { $ET = "Logon" } ElseIf ($Log.InstanceId -eq 7002) { $ET = "Logoff" } Else { Continue } $Result += New-Object PSObject -Property @{ Time = $Log.TimeWritten 'Event Type' = $ET User = (New-Object System.Security.Principal.SecurityIdentifier $Log.ReplacementStrings[1]).Translate([System.Security.Principal.NTAccount]) } } $Result | Select Time,"Event Type",User | Sort Time -Descending | Out-GridView Write-Host "Done." } Else { Write-Host "Problem with $Computer." Write-Host "If you see a 'Network Path not found' error, try starting the Remote Registry service on that computer." Write-Host "Or there are no logon/logoff events (XP requires auditing be turned on)" }It prompts you for a remote host name and a number of days back you want to search and then dumps out a sort-able table of date/time and usernames for logon/logoff events on that host:
I hesitate to outright speak for anyone here, but I don't remember anyone who frequents the thread doing any AWS stuff.
They have an IAM intro here and if this video is any indication, it looks like it's trying to be pretty ELI5
LADIES
THIS SHIT WAS IN POWERSHELL 2????
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
you'd thing it was like a different version of Format-Table
not a whole fucking GUI interface
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
We are upgrading <software> on <server> on Friday evening. This will require an update on all of the client machines once it's been completed. We will need this done prior to start of business on Monday morning.
I find their timeline very unlikely to hold up.
He should have read the install guide.
If I had a dollar for every time that happened to me I'd be retired.
Oh your 2 gig spreadsheet with links to 100 other spreadsheets isn't working right and you want us to fix it?
Hmm let me check, yup, making excel spreadsheets is literally your entire job, this one's on you buddy.
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies