701 reputation
510
bio website sourcefrog.net
location Sydney, Australia
age 38
visits member for 4 years, 6 months
seen Dec 15 at 6:09

May
27
comment Is it important to reboot Linux after a kernel update?
You should really separate production hosting (fairly exposed and security-critical, should get updates right away) from running git repos (probably only trusted users but must be secure) and general screen sessions. VMs are cheap!
May
27
awarded  Excavator
May
27
revised Is it important to reboot Linux after a kernel update?
clearer titles
May
27
suggested approved edit on Is it important to reboot Linux after a kernel update?
Feb
6
awarded  Necromancer
Feb
6
awarded  Yearling
Oct
24
comment Linux: how to restore config file using apt-get/aptitude?
Does that actually restore missing configuration files?
Oct
15
answered How to forbid non-UTF-8 filenames?
Aug
27
awarded  Notable Question
Aug
10
comment How do I write man pages?
Another good option is ronn, which reads the more-widely-used Markdown text markup language.
Jun
2
comment Need an IP Range for google's CDN to set up a firewall
This is true, but... It's not the number of employees, but the amount of third party content hosted by Google, such as docs, YouTube and blogger. A proxy probably us smarter than IP based restrictions if you want to restrict outbound requests.
Sep
18
comment PPTP VPN Problem
It might also be a firewalling problem on the client's local network - try connecting from somewhere else if you can.
Aug
18
comment linux LVM mirror vs. MD mirror
@toor the third volume is used to store a write-ahead log for the mirror, basically to keep track of what needs to be resynced after an abrupt stop. On MD, this is normally stored in a reserved part of the two disks.
Jun
1
awarded  Yearling
Feb
14
awarded  Popular Question
Jan
30
awarded  Revival
Oct
17
answered linux LVM mirror vs. MD mirror
Sep
29
comment 100% uptime for a web application
@ChrisLively All the more reason to have a mature understanding of risk then. The dominant paradigm for safety engineering is probabilistic risk assessment. There are systems that could kill (not just annoy) thousands of people and they still have a low, hopefully well understood, but non-zero probability of failure.
Jul
28
awarded  Scholar
Jul
28
accepted Causes of sudden massive filesystem damage? (“root inode is not a directory”)