Once again, it goes to the orginal set-up. It does not sound like you have any thing near an 'everyday' linux box - or more to the point - the appropriate IT personnel to assemble a linux box of any type.
Let's call a spade a spade - we have had our share also of IT personnel that shouldn't have built the linux box either. We learned quickly those boxes wouldn't stay up from the get-go. It truly sounds like your people are starting with insufficient foundations.
As for attacks -our servers stay under attack numbering tens of hostile hits per second - and the generalities of uptime, our experience is that a proper linux build stays up until we decide to bring it down (6-7-8-9-months uptime), usually these days to pop in more cheap ram.
btw, our linux server to windows server ratio is 6:1
WE have found windows-centric IT personel a) MUST tinker with servers - they can't help themselves. b)pressed in the trenches, we've had some admit they really view Windows OS's as the equivalent of job security c)have tendencies to over-install services on initial linux server-builds - under a "this sounds neat" to put in - (which results in something they can tinker with)
These behavior patterns result in loose-server builds, and constant tinkering.
When we deploy a tightly-built and configured linux server - and we can keep tinkerers away from it, it stays up.
In other words, if it ain't broke don't fix it.
If you're exercising oversight over IT funtionality, you need to determine how much of the IT problem solving is caused by the IT department itself.
If your IT guys are having problems keeping servers up on a daily basis, they are being configured from the start the wrong way and/or your IT guys need to stop tinkering with them every day.
Slashdot Mirror
Once again, it goes to the orginal set-up. It does not sound like you have any thing near an 'everyday' linux box - or more to the point - the appropriate IT personnel to assemble a linux box of any type. Let's call a spade a spade - we have had our share also of IT personnel that shouldn't have built the linux box either. We learned quickly those boxes wouldn't stay up from the get-go. It truly sounds like your people are starting with insufficient foundations. As for attacks -our servers stay under attack numbering tens of hostile hits per second - and the generalities of uptime, our experience is that a proper linux build stays up until we decide to bring it down (6-7-8-9-months uptime), usually these days to pop in more cheap ram. btw, our linux server to windows server ratio is 6:1
WE have found windows-centric IT personel a) MUST tinker with servers - they can't help themselves. b)pressed in the trenches, we've had some admit they really view Windows OS's as the equivalent of job security c)have tendencies to over-install services on initial linux server-builds - under a "this sounds neat" to put in - (which results in something they can tinker with)
/or your IT guys need to stop tinkering with them every day.
These behavior patterns result in loose-server builds, and constant tinkering.
When we deploy a tightly-built and configured linux server - and we can keep tinkerers away from it, it stays up.
In other words, if it ain't broke don't fix it.
If you're exercising oversight over IT funtionality, you need to determine how much of the IT problem solving is caused by the IT department itself.
If your IT guys are having problems keeping servers up on a daily basis, they are being configured from the start the wrong way and