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

Bug#712428: marked as done ([reporting]: lab has 44k binary packages, but the log references only 41k)

Your message dated Thu, 14 May 2015 09:29:14 +0200
with message-id <55544ECA.9030505@thykier.net>
and subject line Re: [reporting]: lab has 44k binary packages, but the log references only 41k
has caused the Debian Bug report #712428,
regarding [reporting]: lab has 44k binary packages, but the log references only 41k
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.)


-- 
712428: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=712428
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems
--- Begin Message --- 
Package: lintian
Version: 2.5.13
Severity: normal

In short:

"""
$ cat logs/lintian.log | grep -E -o 'N: Processing binary package \S+' | sed 's/^N: Processing binary package //g'  | wc -l
41221
$ wc -l laboratory/info/binary-packages 
44108 laboratory/info/binary-packages
"""

Both I and Jakub Wilk were expecting these two numbers to be
identical, but for some reason they are not.  The difference is 2887,
which is close to but not quite equal to the number of the number
binary packages that appear twice or more in the lab (e.g. once for
experimental and once for sid):

"""
$ cat logs/lintian.log | grep -E -o 'N: Processing binary package \S+' | sed 's/^N: Processing binary package //g'  | sort | uniq -c | grep -v ' 1 ' |wc -l
2951
"""[1]

I have not got a good clue on why this happens yet.  Removing them
from the lab makes harness reimport them[2] and while lintian
processes a lot of groups after that the difference remains the same
AFAICT.

~Niels

[1] Of the 2951, only 7 of them appears 3 times (and no package
appears more than that) at the time of measuing.

[2] TBH I had lintian remove all binaries with the names of packages
appearing in the lab but not in the log.  (i.e. I didn't include
versions in my removal queries).

--- End Message ---
--- Begin Message --- 
Version: 2.5.31

On Sat, 15 Jun 2013 23:35:52 +0200 Niels Thykier <niels@thykier.net> wrote:
> Package: lintian
> Version: 2.5.13
> Severity: normal
> 
> In short:
> 
> """
> $ cat logs/lintian.log | grep -E -o 'N: Processing binary package \S+' | sed 's/^N: Processing binary package //g'  | wc -l
> 41221
> $ wc -l laboratory/info/binary-packages 
> 44108 laboratory/info/binary-packages
> """
> 
> Both I and Jakub Wilk were expecting these two numbers to be
> identical, but for some reason they are not.  The difference is 2887,
> which is close to but not quite equal to the number of the number
> binary packages that appear twice or more in the lab (e.g. once for
> experimental and once for sid):
> 
> [...]

After the harness refactoring to use a state cache, we now got:

$ grep -E -o 'N: Processing binary package \S+' logs/lintian.log \
  | sed 's/^N: Processing binary package //g'  | wc -l
78203
$ grep binary: harness-state/state-cache | grep -v '\*'  | wc -l
78203


Accordingly, I am pleased to close this bug as fixed! :)

Thanks,
~Niels

--- End Message ---
Reply to: