Your message dated Sun, 15 Jun 2008 18:30:02 +1000
with message-id <[EMAIL PROTECTED]>
and subject line Re: Bug#486016: gc_approx fails with Not_found, exits with 
return code 1
has caused the Debian Bug report #486016,
regarding gc_approx fails with Not_found, exits with return code 1
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 [EMAIL PROTECTED]
immediately.)


-- 
486016: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=486016
Debian Bug Tracking System
Contact [EMAIL PROTECTED] with problems
--- Begin Message ---
Package: approx
Version: 3.0.0~bpo40+1
Severity: normal

Hi

Today gc_approx started failing.  This is reproducible for me.

Any ideas? (Other than filesystem corruption?  I will try an fsck later).

Thanks for your work in Debian.

Andrew V


# /usr/sbin/gc_approx -v -f
# /var/cache/approx/debian-marillat/dists/lenny/main/source/Sources.gz
# /var/cache/approx/debian-marillat/dists/lenny/main/binary-i386/Packages.gz
....
# /var/cache/approx/debian/dists/etch/contrib/source/Sources.gz
# /var/cache/approx/debian/dists/etch/contrib/binary-i386/Packages.gz
# /var/cache/approx/debian/dists/etch/contrib/binary-amd64/Packages.gz
# 
/var/cache/approx/debian/dists/lenny-proposed-updates/non-free/source/Sources.gz
Not_found


$ tail -n20 approx_bug.strace
close(3)                                = 0
write(2, "# /var/cache/approx/debian/dists"..., 83) = 83
gettimeofday({1213314895, 90134}, NULL) = 0
rt_sigaction(SIGINT, {SIG_IGN}, {SIG_DFL}, 8) = 0
rt_sigaction(SIGQUIT, {SIG_IGN}, {SIG_DFL}, 8) = 0
rt_sigprocmask(SIG_BLOCK, [CHLD], [], 8) = 0
clone(child_stack=0, flags=CLONE_PARENT_SETTID|SIGCHLD, 
parent_tidptr=0xbffa80f4) = 15073
waitpid(15073, [{WIFEXITED(s) && WEXITSTATUS(s) == 0}], 0) = 15073
rt_sigaction(SIGINT, {SIG_DFL}, NULL, 8) = 0
rt_sigaction(SIGQUIT, {SIG_DFL}, NULL, 8) = 0
rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0
--- SIGCHLD (Child exited) @ 0 (0) ---
open("/var/cache/approx/debian/dists/lenny-proposed-updates/non-free/source/Sources.14906.090133905",
 O_RDONLY|O_LARGEFILE) = 3
fcntl64(3, F_SETFD, FD_CLOEXEC)         = 0
_llseek(3, 0, [0], SEEK_CUR)            = 0
unlink("/var/cache/approx/debian/dists/lenny-proposed-updates/non-free/source/Sources.14906.090133905")
 = 0
read(3, "Package: flashplugin-nonfree\nPri"..., 4096) = 1555
close(3)                                = 0
write(2, "Not_found\n", 10)             = 10
exit_group(1)                           = ?


$ ll /var/cache/approx/debian/dists/lenny-proposed-updates/non-free/source/
total 8
-rw-r--r--  2 approx approx 104 2008-04-15 18:21 Release
-rw-r--r-- 20 approx approx 867 2008-06-10 03:40 Sources.gz



-- System Information:
Debian Release: 4.0
  APT prefers stable
  APT policy: (500, 'stable')
Architecture: i386 (i686)
Shell:  /bin/sh linked to /bin/bash
Kernel: Linux 2.6.18-6-686
Locale: LANG=en_AU.UTF-8, LC_CTYPE=en_AU.UTF-8 (charmap=UTF-8)

Versions of packages approx depends on:
ii  adduser                3.102             Add and remove users and groups
ii  bzip2                  1.0.3-6           high-quality block-sorting file co
ii  curl                   7.15.5-1etch1     Get a file from an HTTP, HTTPS, FT
ii  libc6                  2.3.6.ds1-13etch5 GNU C Library: Shared libraries
ii  libpcre3               6.7+7.4-3         Perl 5 Compatible Regular Expressi
ii  lsb-base               3.1-23.2etch1     Linux Standard Base 3.1 init scrip

approx recommends no packages.

-- no debconf information



--- End Message ---
--- Begin Message ---
Hi Eric

On Friday 13 June 2008 12:34, Eric Cooper wrote:
> > Today gc_approx started failing.  This is reproducible for me.
> >
> > Any ideas? (Other than filesystem corruption?  I will try an fsck
> > later).
>
> I doubt there's any problem with your filesystem, probably just a bug
> in my code.  Does it still fail after doing an "apt-get update" from a
> client that uses lenny-proposed-updates?  How about after running
> update_approx on the server?
>
> Also, can you send me your approx.conf file?  I tried looking at
>     debian/dists/lenny-proposed-updates/non-free/source/Sources.gz but
> it seems to be empty ...  If you still have a non-empty one,
> please send that too.

Attached debian/dists/lenny-proposed-updates/non-free/source/Sources.gz.

However a proper look at it has shown me what the problem is.  The file 
contents actually match 
etch-proposed-updates/non-free/binary-i386/Packages.gz. 

What I think happened is this.  A while ago I replaced a disk on my home 
server with a larger disk,  Because I didn't have room in the server to 
temporarily add the new disk, I added it to my desktop (which already had a 
backup of all the relevant files), and did a cp -a from my backups to the 
new disk, then dropped it in the server.  

My backups are managed by dirvish, which hardlinks unchanged files to the 
corresponding file from yesterday's backup.  In addition I occasionally 
runa script to hardlink other duplicate files together.  At the time I did 
the cp -a lenny-proposed-updates/non-free/source/Sources.gz appears to have 
been linked to {sarge,etch,lenny}-proposed-updates/{contrib,non-free}/
{binary-i386,binary-amd64,source}/{Sources,Packages}.gz.  (It appears to 
have been the 20 byte file you get from touch foo ; gzip < foo > foo.gz).  

Mostly I don't actually use *-proposed-updates.  Last week I updated a vm 
that does use etch-proposed-updates, and that resulted in 
{etch,lenny}-proposed-updates/{contrib,non-free}/
{binary-i386,binary-amd-64,source}/{Packages,Sources}.gz having the wrong 
content.

I have fixed things locally, and closed the bug.  Approx could prevent this 
by downloading files to a temp name, and then move them into place.  

Thanks for your work in Debian.
Andrew

Attachment: Sources.gz
Description: GNU Zip compressed data


--- End Message ---

Reply via email to