The Precise Pangolin has reached end of life, so this bug will not be
fixed for that release
** Changed in: apt (Ubuntu Precise)
Status: New => Won't Fix
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
Sorry for the slow reply. I wasn't quite sure if precise is needed or
not. Attached is a debdiff with the backport of the http/https refactor
that David did. The diff is a bit scary because it contains a lot of
churn/shuffle of the http/https code.
Note that the debdiff needs a bit of cleanup
This bug should indeed be fixed in trusty with the upload that Brian
identified. If a backport for precise is needed I can look into that.
** Changed in: apt (Ubuntu)
Status: New = Fix Released
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is
It looks to me like this may be resolved in Trusty due to the following
changelog entries:
apt (0.9.12) unstable; urgency=low
My problem with https is that I hadn't the infrastructure set up to test it –
and code needs testing, especially if it ends up on millions of systems.
Others might be good enough to write (or apply) bugfree code without testing,
but I am not.
So what I do with patches I can't test is that I
I'm also unconvinced of this patch. It drops the If-Range logic
altogether which means that it requests a range without specifying what
the mtime of the partial file was. It also does not use If-Modified-
Since in this case. So you'd still end up with a file that's partially
corrupted if the file
The current code completely corrupts things. It's definitely got a bug.
Dropping If-Range entirely at least gets *correct* behavior.
David, people *have* read your previous comments, *and responded to them*.
If you don't want to do https stuff, then apt needs a maintainer who
does.
So, since
Ping, need some movement/response to OP comments.
Thanks!
Adam
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1157943
Title:
apt-get update fails hash checks on https repositories when file size
David, sorry that ping was meant for you to try and unblock this issue.
Thanks again
Adam
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1157943
Title:
apt-get update fails hash checks on https
Could you guys please refrain from pinging me for comments without
reading my previous comments?
Not only does this response now eat up all the time I have available for
APT this week, its also highly demotivating to have the public record of
two @canoncial.com people that they don't have to read
David, any opinion on Thomas's previous comment ? Jury is still out on
wether or not the patch proposed is adequate.
Thank you,
...Louis
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1157943
Title:
David, #6 means that servers SHOULD send a 416. But apt should support
both, because it cannot rely on servers sending a 416.
#7, you have misunderstood. Yes, there is newer contents, and the server
returns the current complete contents of the file in this case, and apt
is *not replacing the file
#6: SHOULD means that there are servers outwhere behaving like that. So we have
to support both. Not that we can ignore should-clauses.
#7: Do I really have to comment that? I am not that good at explaining jokes,
but lets try:
1. APT downloads a part of a file.
2. replaces the file on the
David, could you provide your thoughts in response to comment #6 and #7?
Thank you,
Adam
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1157943
Title:
apt-get update fails hash checks on https
** Also affects: apt (Ubuntu Precise)
Importance: Undecided
Status: New
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1157943
Title:
apt-get update fails hash checks on https repositories
David, are you happy with this patch?
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1157943
Title:
apt-get update fails hash checks on https repositories when file size
changes
To manage
(assuming David means me: Better ask Michael Vogt as he is debugging
another https problem at the moment)
That said, step 4 is incorrect, Range answers with an error code of 416. The
behavior described is that of Range with If-Range.
I haven't tested this at all, but browsing documentation makes
David, servers are not required to send 416; it's only a SHOULD.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1157943
Title:
apt-get update fails hash checks on https repositories when file size
The behavior in 4 is not the If-Range + Range behavior. The latter is
about what happens when the contents have changed; the behavior in 4 is
not about changed content, but about a range request which starts after
the end of the file on the server.
--
You received this bug notification because
Oh, and the bug is particularly pathological. Because after it happens,
the index file is in the staging area is now always larger than what's
on the server, it won't self correct.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
** Patch added: apt-fix.diff
https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1157943/+attachment/3586456/+files/apt-fix.diff
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1157943
Title:
The attachment apt-fix.diff seems to be a patch. If it isn't, please
remove the patch flag from the attachment, remove the patch tag, and
if you are a member of the ~ubuntu-reviewers, unsubscribe the team.
[This is an automated message performed by a Launchpad user owned by
~brian-murray, for
22 matches
Mail list logo
