Reputation
Next privilege 125 Rep.
Vote down
Badges
1
Newest
 Supporter
Impact
0 people reached

  • 0 posts edited
  • 0 helpful flags
  • 1 vote cast
Aug
21
comment How do you add a Windows environment variable without rebooting?
The above comment describes the client OS's, but the same more or less applies to servers. I'm not sure if the shutdown with a reason dialog supports the exit explorer feature though, never had the opportunity to test it, and don't want to muck with any of the servers in my control.
Aug
21
comment How do you add a Windows environment variable without rebooting?
Uhh. Please don't kill windows explorer unless one of its processes are hung. Instead, open the shutdown windows dialog and cancel it while holding ctrl+alt+shift. That will cleanly exit windows explorer. In Vista+ the shutdown windows dialog is harder to find (but still present, at least through 7 (unsure about 8 and 10)), so there is a second method. ctrl+shift+right-click in an empty part of the start menu and choose exit explorer. In 8 the exit option is the same, but you use the taskbar not the start menu.
Aug
21
awarded  Supporter
Mar
13
comment Message ALL users before shutdown
Unixes vary. On OS X (and Linux) you want shutdown -r +5 "message", for a shutdown in 5 minutes. (No -t, which specfies the amount of time to let programs exit gracefully.) You can vary the time, but 1 minute is the minimum. A longer time may be better because it actually gives users a chance to react, or perhaps even contact the admin to abort the shudown. On Solaris you do: shutdown -y -g 30 -i 6 "message" for a 30 second shutdown.
Jan
7
comment Using Y power cables in a datacenter
I think the "PSU" in point 3 should also be "PDU", resulting in the points meaning: you consider the risk of cable or PDU failure to be much smaller than the risk of PSU failure which sounds about right to me. I've seen many more PSU failures than cable failures. I've not done much rack-based work, so I can't really comment on PDU failure rates.
Dec
9
comment What are the different widely used RAID levels and when should I consider them?
Raid 1000000 would require a minimum of 128 disks, but it would provide 64 disks worth of storage space, it was have the same worst case write performance as Raid 1, and any 2 adjecent drive failures would kill the array. You were descrbing Raid 0111111, which would have pretty good reliability (Raid 11111110 would have must better reliability on avergage.)
Oct
26
comment Why would certain network switches stop working, others are fine?
@Aaron: To follow up on Evan's reply. All ports on a hub share a collision domain, and are either passive devices, or may amplify signals back to full strength as part of the process. A repeater is a device that acts much like a hub, any packet that comes in will be rebroadcast on all other ports. However each such port has a seperate collision domain and repeaters will rebroadcast in the event of a collision. Many so-called Hubs are actually repeaters.
Aug
16
comment Which Is a Good Starting Point For Beginner? Apache, Nginx or Cherokee?
I was not misreading your answer, but I saw that it could be misread, so commented to clarify. Server side includes is not included in many people's defintions of static, depite not really being dynamic either. It would also be easy to misinterpert "or it proxies", as requiring a fully compliant HTTP 1.1 server to host the dynamic content, which is not the case, since it can also proxy to FastCGI services. Since those misreadings were possible, I thought clarifying them in a comment may be worthwhile.
Aug
10
comment Which Is a Good Starting Point For Beginner? Apache, Nginx or Cherokee?
Just to clarify though, nginx does have native server-side includes, and supports FastCGI, so it is not purely proxy-or-nothing. It can also be set up to serve static content, pass anything it could not serve onto annother server, and replace any 404 or similar errors with a server-side redirection.