@ewwhite Second, what is with people that use a half-space in racks? That makes me start hugging myself and rocking back and forth while reciting multiplication tables.
@ewwhite Heh, wow. Do they call you "Edmund's comet" - only once every few years does he get close enough to the heat of the server racks for his contrail to be visible.
@ewwhite I count like 50u used between both, so you'll need to figure something out if you want to get them down to one rack. Can be done if you sell them on new servers to replace the beasties they have. =)
the client I'm referring to in Philadelphia will be able to get 2 ESX hosts, all of the Windows/VMWare/Symantec, Nexenta and two storage units for $30k
@ewwhite depends if you could benefit from the memory, that's the main benefit really - the extra cores is nice but the new disk lights and iLO4 aren't worth the change on their own
@JoelESalas we buy BL460c's generally - and the Gen8's max at 512GB, G7's was 384
@ewwhite we actually put 256GB (16x16GB) into our boxes but we tend not to get close to over committing on vCPUs or memory but for test/dev platforms we don't care about performance so the extra memory capacity comes in handy
Yeah - we generally give our VMs 8GB, so even if they filled it all up on the host we'd still be fine for ~32 VMs/host - which given they only have 32 threads is a bad idea anyway when most of our VMs have =>2 vCPUs
No, we have a Hyper-V cluster, clusters don't start without AD. So you must have at least one AD controller not in the cluster or else the first time the cluster goes offline it will not be able to start again.
So we diligently have two AD boxen outside the cluster. Lost power this morning, one AD controller has hardware issues now, and the other thinks it's 2011.
So cluster no starty.
Fortunately I knew about this type of problem and know how to resolve it... It's a lot of fricking work however. And I'm pretty pissed about the borked AD.
@MDMarra I don't know what it's deal is, but it thinks I'm putting in the wrong password and the computers are complaining about no trust relationship...
@JoelESalas I prefer an operating system that works...
@MDMarra OS X chokes on high disk I/O -- if you make the machine swap or tie the disk up with something intensive your interactive performance (at least in the GUI) goes to shit
@JoelESalas I never trust my computers. they're back-stabbing conniving little Ferengi boxen...
(note to self: the next time a server acts up, throw something at it and call it a toad-faced troll)
Our 'management' interface is currently on the Management0/0 port, which is 100MBPS. I would like to get Gigabit speeds on this interface so that would mean moving it to one of the GigabitEthernet 0/x ports.
Is there a way to move the subinterface that we have created under the Management0/0 por...
Well, the mgmt port on ASA units is a 10/100 port. He's doing this because he wants management on a Gigabit port. I'm not aware of any advantage to doing so, but maybe I'm missing an interesting application.
Any of the ports can do management functions. In older PixOS versions you couldn't remove the management-only state on the management interface. Apparently you can with the newer versions, making it just another 10/100 interface. The 5520 has 4 GbE interfaces, not sure why he isn't using them appropriately.
Not sure why someone who doesn't know what they're doing is mucking around with a $5k Cisco router either, which is probably doing something mission critical to boot...
@voretaq7 Are you familiar with Gore and do any of their products compete with any of yours? Just curious. I found that a guy I know works for Gore and he in essence measures aortas.
I was under the impression that a router was any device that could route packets from one subnet to another. And that Firewalls were Routers with the ability to selectively deny routing. And NATs were Routers with the ability to lie about the source of the packets in a organized fashion (see also, organized crime).
@rnxrx many people would say that a device that is a simple packet filter without the ability to track state is not a 'Firewall'. That is some argued that Ipchains was not a firewall, but Iptables/Netfilter was.
@BartSilverstrim In my day, routers cut wood, routing tables were where you put your coffee mug when you came into the shop for the morning and bits were the metal things you fastened with a chuck.
@BartSilverstrim you can enable trace on shutdown, but the logs are all gobbly gook and take too long to read through for my tastes... I'm just being lazy after a rough day.
@WesleyDavid It is amusing how many woodworking terms apply to computer technology.
It just seems like a really odd set of questions. I would guess some developer created the job description, which makes me think this is a new position that hasn't been around before.
@JoelESalas can't be that much training. They are using macs after all. :p Don't they just have to know where the on button is, and boot off the install disk so they can wipe the drive and re install when things are broke, and how to package your equipment up and send it off to be fixed by someone else.
This looks totally legit: https://docs.google.com/spreadsheet/viewform?formkey=dE5PelF1Z1luVUhaMVJpTFFGS1M1Mmc6MQ "We have just noticed that you have exceeded your email Database limit of 500 MB quota and your email IP is causing conflict because it is been accessed in different server location."
@wfaulk not me, but my firewalls block outgoing port 53 requests tcp&udp except from my internal DNS servers. So if a system had invalid DNS settings inside the network, they would get nowhere.
@JoelESalas Most computers use a RTC based on a crystal oscillator. Quartz in particular is extremely predictable based on the crystallographic axes dimensions. Basically, if you cut a chunk of quartz at a very precise thickness and stick electrodes to either side it will always be the same frequency for that thickness. Manufacturing defects happen in every industry, and some crystals are "off", which causes the RTC to drift.
does professional devops (server administrator) use webmin to manage their server ?
or is it only for unprofessional linux users ?
and is it only for ubuntu(.deb), fedora/redhat(.rpm) ?
thanks !
$ lsof /dev/urandom
COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME
xfce4-ses 1787 to 15r CHR 1,9 0t0 8199 /dev/urandom
applet.py 1907 to 9r CHR 1,9 0t0 8199 /dev/urandom
scp-dbus- 5028 to 10r CHR 1,9 0t0 8199 /dev/urandom
firefox 6603 ...
does professional devops (server administrator) use webmin to manage their server ?
or is it only for unprofessional linux users ?
and is it only for ubuntu(.deb), fedora/redhat(.rpm) ?
thanks !
