Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Bug 1202858 – Restarting squid results in deleting all files in hard-drive

bugzilla.redhat.com

Post reply on HN
Print view View original

71–80 of 179 posts

Re: Bug 1202858 – Restarting squid results in deleting all files in hard-drive

#71

Post by userbinator »

a13xb wrote: post #22
kevan wrote: post #11

Earlier quoted context omitted.

Agreed, I should've elaborated. All it takes is something like this in the init script without checking if the variable is empty: rm -rf "$STEAMROOT/"*

I feel like it would be a frighteningly common bug. I remember one like this from 2011 [1]. Install/packaging/utility scripts usually do not get as much attention and testing as the application code itself. [1] https://github.com/MrMEEE/bumblebee-Old-and-abbandoned/issue...

I'd say the fact that these bugs only very occasionally happen - relative to the huge number of shell scripts out there that are being executed every day - that it's not really "frighteningly common". You only hear about the ones that fail.

Re: Bug 1202858 – Restarting squid results in deleting all files in hard-drive

#72

Post by Erwin »

cellularmitosis wrote: post #40
fermigier wrote: post #2

Scary. Even more scary is the fact that the bug has been open for a week, one person has confirmed 100% reproducibility, and no one seems to care at Red Hat. Isn't deleting peoples hard drives a big no no ?

Ahh Redhat, the distro which chose to symlink a bunch of binary lib files from Apache into the etc directory. "Why is grepping /etc taking so long? Binary files in /etc?!? WTF?!?" Coming from Debian, Redhat seems to make a lot of irk-worthy choices.

I find the /etc/httpd/logs symlink more annoying. If you want to grep through your Apache configuration you have to explicitly grep through conf and conf.d otherwise just going to /etc/httpd and doing a grep -r you're searching through gigs of Apache logs.

Re: Bug 1202858 – Restarting squid results in deleting all files in hard-drive

#73

Post by hk__2 »

JoshTriplett wrote: post #55
jedisct1 wrote: post #47

My good old trick to mitigate that is: touch /-@ I also always do it in my home directory: touch ~/-@ That's the first thing I do on a new host. When accidentally running rm -f *, the command expands to -@ first, which is not a valid option and makes the command fail before doing any harm rm: illegal option -- @ usage: rm [-f | -i] [-dPRrvW] file ...

That won't actually help with rm -rf /* , only with rm -rf * in / or $HOME.

Would it work with `touch "/ -@"` (not the space)?

Re: Bug 1202858 – Restarting squid results in deleting all files in hard-drive

#74

Post by guardian5x »

lpsz wrote: post #32
kevan wrote: post #11

Earlier quoted context omitted.

Agreed, I should've elaborated. All it takes is something like this in the init script without checking if the variable is empty: rm -rf "$STEAMROOT/"*

And this is why it is important to write something like set -eu on top of your bash scripts -- execution will stop on errors (non-zero retvals) and on undefined variables.

I wonder why set -eu is not the default setting.

Re: Bug 1202858 – Restarting squid results in deleting all files in hard-drive

#75

Post by raverbashing »

lambeosaurus wrote: post #52
jedisct1 wrote: post #47

My good old trick to mitigate that is: touch /-@ I also always do it in my home directory: touch ~/-@ That's the first thing I do on a new host. When accidentally running rm -f *, the command expands to -@ first, which is not a valid option and makes the command fail before doing any harm rm: illegal option -- @ usage: rm [-f | -i] [-dPRrvW] file ...

As someone who uses the commandline a lot but isn't exactly a wizard, why does this work?

More specifically, it is expanded on the shell

Example:

$ echo /*

/bin /boot /dev /etc /home (...)

Re: Bug 1202858 – Restarting squid results in deleting all files in hard-drive

#76

Post by coldtea »

mraison wrote: post #49
jiffytick wrote: post #35

Earlier quoted context omitted.

or check the variable before using it, like any other programming language: [[ "$VAR" ]] && rm -rf "$VAR/*" I think most of these issues stem from the fact that most developers that write shell scripts don't actually understand what they're doing, treating the script as a necessary annoyance rather than a component of the software.

> like any other programming language some real-world programming languages don't have undefined variables :)

Since the variable he shows is used in a string interpolation, it doesn't have to be undefined.

Being the emptys string "" would work just as well.

Re: Bug 1202858 – Restarting squid results in deleting all files in hard-drive

#78

Post by vbezhenar »

wodenokoto wrote: post #70
gmac wrote: post #62

Earlier quoted context omitted.

That's kind of scary — in that case I guess I should avoid creating a file named -rf.

That is really interesting. Can someone knowledgeable about the shell expand on this? I don't dare test it on my machine.

Shell session to demonstrate (DO NOT DO IT IN THE DIRECTORY WITH IMPORTANT FILES):

  $ touch important
  $ chmod 400 important
  $ rm *
  override r--------  vbezhenar/staff for important? n
  $ touch -- -rf
  $ ls -l
  total 0
  -rw-r--r--  1 vbezhenar  staff  0 Mar 24 14:35 -rf
  -r--------  1 vbezhenar  staff  0 Mar 24 14:35 important
  $ rm *
  $ ls -l
  total 0
  -rw-r--r--  1 vbezhenar  staff  0 Mar 24 14:35 -rf
  $ rm -- -rf

Re: Bug 1202858 – Restarting squid results in deleting all files in hard-drive

#79

Post by theothertom »

wodenokoto wrote: post #70
gmac wrote: post #62

Earlier quoted context omitted.

That's kind of scary — in that case I guess I should avoid creating a file named -rf.

That is really interesting. Can someone knowledgeable about the shell expand on this? I don't dare test it on my machine.

I've not tested it, but it should expand just like anything else. The effect would broadly be that running "rm *" in the directory would recurse into subfolders without warning.

Re: Bug 1202858 – Restarting squid results in deleting all files in hard-drive

#80

Post by rwmj »

fermigier wrote: post #2

Scary. Even more scary is the fact that the bug has been open for a week, one person has confirmed 100% reproducibility, and no one seems to care at Red Hat. Isn't deleting peoples hard drives a big no no ?

It's already fixed, plus there are many private comments which you cannot see (but I think you see "missing" comment numbers). Also the product in question is not released yet. It's good this was found, but no customer would have been affected unless they were using an alpha.
Post reply on HN