[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: wifi failed after hibernate

Freddy Freeloader wrote:
I can see where the current manner of configuration be sort of a hassle if you use your laptop at work with a wireless connection, but I don't see how making wpa_supplicant easier to use overall is worth getting as sarcastic as you got though. You're obviously running Sid or Etch, and wireless is under heavy development, so what do you expect?
Well, there were two things, specifically that annoyed me.

First, I didn't get any warning message during the upgrade. The NEWS.Debian file says, in all caps "IF YOU WERE USING WPASUPPLICANT BEFORE, THEN THIS UPGRADE PROBABLY BROKE YOUR CONFIGURATION". If it's important enough to put in all caps (and it is) and if you know that it's going to *break* the previously-working configs of a majority (ie, over 50%... which is what I infer from their use of "PROBABLY" instead of "POSSIBLY") of the installed base, then I think some kind of mention of this in the preinst or postinst script would have saved a lot of hassle for people.

Secondly, once I had discovered why my wifi suddenly didn't work, I was actually eager to switch over to the new method... until I read the part about how, if you needed roaming capability, then you had to switch back to the old way. Think about that for a moment. If you're planning on using your wifi for *roaming* (which I estimate is a vast majority), then you've got to switch back to using the old /etc/init.d/wpasupplicant script.... which is now *GONE* and your old /etc/wpasupplicant.conf file... which is now *GONE*.

So, to summarize:
- They didn't notify the user that the package was going to break the existing config.
- The new way didn't do what most people needed it to do.
- The upgrade deleted files that were needed to go back to the old way.

- Joe

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature

Reply to: