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

Bug#616064: marked as done (Hash Sum mismatch even DOSes apt-cache)

Your message dated Wed, 1 Mar 2017 20:24:15 +0100
with message-id <20170301192415.hixkdp2lajxgrxnz@crossbow>
and subject line Re: Hash Sum mismatch catch-all bug
has caused the Debian Bug report #624122,
regarding Hash Sum mismatch even DOSes apt-cache
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact owner@bugs.debian.org
immediately.)


-- 
624122: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=624122
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems
--- Begin Message --- 
Package: apt
Version: 0.8.11.5
Severity: wishlist

Gentlemen, the problems mentioned in
http://article.gmane.org/gmane.linux.debian.user/402702
http://aptosid.com/index.php?name=PNphpBB2&file=viewtopic&t=774
often:
1. knock out aptitude, and even apt-cache show, search. It doesn't know about any packages anymore.
1a. Can't work offline now as my lists are damaged.
2. it's like a DOS attack, can't hardly use any apt service because my
lists are now messed up.
3. we end up just putting the US site into source.list.d and say forget
about the mirrors.

Why can't mirror refreshing be done in the proper order so the end user
doesn't see any partial states?

--- End Message ---
--- Begin Message --- 
Hi,

I am hereby closing #624122 and its merges as it was never tracking
a specific issue, but catched all mails from users which encountered an
error involving the words "Hash Sum mismatch".

Over the years we have dealt with at least three different bugs in here
which would have been better if they would have been in new fresh bugs
rather than expanding on an existing but ultimatively unrelated bug
– not counting the initial "bug" of having repository updates not being
atomic, but over the last 6 years a lot has improved on that front as
well.

deity, ftpmasters, mirror and other teams have worked on whats dubbed
"by-hash" for years, which resulted in various smaller and bigger
improvements rolled out either universially by now and/or will be
included in the stretch release (depending on if server, client or both
need to support something).


That work seems to pay off as (ignoring spam) this bug class was pretty
silent in the last few months – and if someone wants to talk about it
the reporter will be better of starting with a new bug as going through
the buglog of a 6years and counting open bugreport intermixed with
unrelated messages and spam is kinda daunting…

In other words: It has taken 3 years, but now is the day I was talking
about: "One day, I will have the mental strength to close it…".


Triggered by koike on IRC btw who were asking about the 1y old "will
eventually be a different message for this" (talking about weak hashes)
which we have figured out was resolved a few months later.


Thanks to all reporters and to everyone working on these issues!


Best regards

David Kalnischkies

Attachment: signature.asc
Description: PGP signature


--- End Message ---
Reply to: