{5} Assigned, Active Tickets by Owner (Full Description) (20 matches)

List tickets assigned, group by ticket owner. This report demonstrates the use of full-row display.

FlorianFesti

Ticket Summary Component Milestone Type Created
Description
#46 Improve db4 settings (cache sizes) and syncing behaviour rpm rpm-4.10.0 defect 03/05/09

Right now creating the Basenames index takes quadratic time (over # of common basenames). Caching should be able to hide this. First tests show that there are several hours that can be saved for really big installs.

pmatilai

Ticket Summary Component Milestone Type Created
Description
#67 Do not open database when querying package rpm defect 06/10/09

When querying a package, rpm tries to open a database. However there is IMHO no need to have working database when querying file on disk. And rpm also gives the response, just also prints some errors.

$ rpm -qp --queryformat '%{VERSION}\n' ~/archivemail-0.6.1-3.ae.src.rpm error: cannot open Packages index using db3 - No such file or directory (2) error: cannot open Packages database in /var/lib/rpm warning: /home/joey/archivemail-0.6.1-3.ae.src.rpm: Header V3 DSA signature: NOKEY, key ID 4e6c9f5e 0.6.1

(Forwarded from Debian BTS - http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=451441)

#97 Rebuild rpm database causes segfault rpm defect 09/19/09

I've been running into problems when trying to rebuild the rpm database on my system after KPackageKit fails. I've tried the RPM database recovery to repair the database and when I get to the last step to rebuild the database, rpm -v --rebuilddb a segfault is returned. I currently have rpm 4,7,1 installed on fedora 11.

In the kernel the following is reported:

2009-09-19 08:57:49 localhost kernel rpm[8990] general protection ip:364f91c8fe sp:7fffdddeb010 error:0 in libdb-4.7.so[364f800000+16d000]

In my latest rpmpkgs file the following is listed:

error: cannot open Packages database in /var/lib/rpm error: cannot open Packages database in /var/lib/rpm error: cannot open Packages index using db3 - (-30974) error: db4 error(-30974) from dbenv->failchk: DB_RUNRECOVERY: Fatal error, run database recovery error: db4 error(-30974) from dbenv->failchk: DB_RUNRECOVERY: Fatal error, run database recovery rpmdb: Thread/process 3167/140619021432560 failed: Thread died in Berkeley DB library rpmdb: Thread/process 3167/140619021432560 failed: Thread died in Berkeley DB library

In addition I'm attaching a backup of my rpm database, and the output from rpm -vv --rebuilddb.

#121 rpmbuild: should --target really set %_target? rpm defect 01/09/10

--target is used to specify the architecture for which a package should be generated; incidentally it also sets $RPM_OPT_FLAGS which often has things like -mcpu=i686, for example. However, --target sets %_target, which actually looks wrong. By the autotools definitions, "host" is actually the parameter for selecting the arch of the program, while "target" refers to the code produced by that program in turn. Needless to say that "target" is therefore only in use for compilers and alike, standard programs like strace only have host (where it is supposed to run on) and build (where the compiler runs).

Thus, given an rpmbuild strace.spec --target=sparcv9 would still be %configured with --build=sparc64-unknown-linux --host=sparc64-unknown-linux, which is not what was wanted.

Under these considerations, I think rpmbuild's --target should set %_host instead. (Or a --host option needs to be added, whichever is more reasonable.) What do you think?

#157 Permissions not attached to files in chroots rpm defect 05/06/10

Installation with rpm --root yields:

/srv/nfs/base32# linux32 rpm --root $PWD -Uhv cups-client-1.3.11-4.5.1.i586.rpm Preparing... ########################################### [100%]

1:cups-client warning: user lp does not exist - using root( 26%)

warning: group lp does not exist - using root ########################################### [100%]

And that despite lp existing in both /etc/{passwd,group} and /srv/nfs/base32/etc/{passwd,group}. Looks non-trivial. Any idea what could be done?

#186 rpm-4.8.1 mips: no isa_name, isa_bits rpm defect 10/18/10

rpm-4.8.1 build for Fedora 13 mips gives this error: 

error: Macro %__isa_name has empty body
error: Macro %__isa_bits has empty body

Attached is a patchlet to fix this.

#43 RFE: Add tags to save the circumstances under which the package got installed rpm enhancement 03/05/09

Add "hdr" manip. code for key/val pairs that yum wants

  • from_repo
  • note
  • installed_for (groupinstall+group/dep+dep/explicit - enough?)
#154 Automatic Provides and Requires for python modules rpm rpm-4.10.0 enhancement 03/22/10

Currently, there is no automatic provides/requires for python modules in RPM, similar to what perl modules have. For example:

  • Provides(SQLAlchemy)
  • Requires(setuptools)

The pythondeps.sh script only handles python's ABI version dependency.

I've written a new pythondeps.sh script to do that. It is based on the latest Git version (2010-02-09) and adds the scanning of module provides and requires from the python metadata (egg-info file and directory when setuptools is used). The details of the script and a technical analysis by Toshio Kuratomi is available here : http://lists.fedoraproject.org/pipermail/python-devel/2010-March/000230.html

The script itself is available here : pythondeps.sh.

I've written a small test suite (shell-based) here. It was originally intended to test versionned requirements, but to keep the script simple we decided not to handle versions at the moment, at least until the python versionning scheme stabilizes.

Please consider this script for inclusion. I am, of course, very open to discussion on the idea or the implementation.

#848 RFE: quiet flag for scriptlets rpm enhancement 09/11/11

There's a lot of unnecessary /dev/null cruft in countless specfiles whose purpose is to redirect stderr/out of scriptlets to /dev/null.

Having a quiet flag that'd automatically redirect scriptlet stderr and stdout to /dev/null or suppress them some other way would provide an easier way to accomplish that, as well as to use programs that always output something as scriptlet programs with -p yet have their output quiet.

For example %post -Q -p /some/thing. (-q instead of -Q would be more consistent with %setup, but it seems to be already taken for scriptlets by some recent changes.)

#849 RFE: no-error flag for scriptlets rpm enhancement 09/11/11

There's a lot of unnecessary "exit 0" or "

 :" and the like cruft in countless specfiles whose purpose is to avoid scriptlets exiting with non-zero exit codes.

Having a no-error flag that'd make the exit status always 0 would provide an easy way to accomplish that, for example %post -e -p /bin/false. It'd also provide a way to do this for scriptlet programs without a script body, for example the /bin/false example above, or %post -e -p /sbin/ldconfig etc.

#850 export header start and header end positions as a part of RPM header rpm enhancement 09/13/11

Some tools (like YUM) work with RPM headers.
They need to know when the header start and header end within a file (it's part of createrepo at least).
I think that custom RPM file parsing is a really bad idea and RPM should provide this information.

I'm mostly interested in python bindings.
Can you export the information, preferably as a part of the RPM header?

See the _get_header_byte_range method in:
http://yum.baseurl.org/gitweb?p=yum.git;a=blob;f=yum/packages.py;h=f72c068200587288a7c42371f7dacb2faba60991;hb=refs/heads/yum-3_2_X

#15 rpmfcELF: Assertion `s != ((void *)0)' failed rpm task 12/11/08

I'm trying to build an RPM. I previously ran into problem where the build would fail due to failure to brp-strip the binaries. The error message was something like:

/usr/bin/strip: Unable to recognise the format of the input file / home/mydir/RPM/BinaryName

So I added exit 0 to the top of brp-strip, and got past that error. But now when I get to the end of the %file section I get this error:

rpmbuild: rpmfc.c:1079: rpmfcELF: Assertion `s != ((void *)0)' failed.

google search has so far been fruitless. Can anyone help me out with this?

Thank you,

Robert

#21 message : diff: illegal option -- d rpm defect 01/14/09

We've been using rpm for about 10 years on a variety of systems : Linux of course, but also Solaris and HP-UX. We are currently in the process of upgrading rpm on Solaris from version 4.0.4 to version 4.4.2.3. I came across the following error message : rpmbuild --quiet -bb <mypackage>.spec diff: illegal option -- d

GNU diff is not installed on our system and we don't plan to as we try to be as close as possible to a standard Solaris install. I have found the following - and hopefully temporary - workaround : I'm using a custom macros file with a '%_ _check_files' tag referencing a modified 'check-files' script (which does a mere diff instead of diff -d).

Using rpm on non-Linux systems was criticized in the past by some people here. If we have to install other packages to do so, I'm afraid that we'll have to quit.

#94 No macro.c rpm defect 09/08/09

I recently tried to emerge rpm in gnetoo linux with a 2.6.6-r6 kernel and the emerge fails. It shows an error with macro.c (listed below). It was seuggested to look at the code for macro.c and see where the issue is. Using the find command nothing turned uo for macro.c. looks like the file is not present in the package.

macro.c: In function rpmGlob: macro.c:1700: warning: implicit declaration of function setlocale macro.c:1700: error: LC_COLLATE undeclared (first use in this function) macro.c:1700: error: (Each undeclared identifier is reported only once macro.c:1700: error: for each function it appears in.) macro.c:1700: warning: assignment makes pointer from integer without a cast macro.c:1703: error: LC_CTYPE undeclared (first use in this function) macro.c:1703: warning: assignment makes pointer from integer without a cast

#158 rpmbuild should error and/or toss out EMPTY provides/requires/conflicts/obsoletes entries rpm defect 05/13/10

https://bugzilla.redhat.com/show_bug.cgi?id=473084

somehow that pkg http://kojipkgs.fedoraproject.org/packages/opal/3.4.2/1.fc10/i386/opal-3.4.2-1.fc10.i386.rpm

has an empty provides entry. I suspect it came from the find-provides script.

This tripped up yum (this is now fixed in yum) but in general I think an empty provides is a bad idea.

#176 "rpmbuild -bp --force" silently skips all %patch directives rpm defect 08/12/10

In rpm 4.8.1 (and 4.7.2 and probably other versions), if you call parseSpec() with force=1 (as python/spec-py.c is inclined to do), parsePrep.c's doPatch code always returns the empty string. This is because of this line in parsePrep.c:

/* On non-build parse's, file cannot be stat'd or read. */

if (spec->force

checkOwners(fn)) goto exit;

This code seems to be clearly not as intended. It should instead read:

if (!spec->force && checkOwners(fn)) goto exit;

This is really screwing up my ability to execute steps of rpmbuilding from python scripts.

#17 RPMTRANS_FLAG_NOSIGNALS rpm rpm-4.10.0 enhancement 12/16/08

Currently (as far as I can see) librpm processes SIGINT-like signals itself (e.g. function rpmdbCheckSignals calls exit(EXIT_FAILURE)). It turns out that an librpm using application cannot react on signals somehow differently. Is it possible to include the capability of custom termination signals handling in the next version of RPM? It could be seen this way for an application:

rpmtsSetFlags(ts, RPMTRANS_FLAG_NOSIGNALS);

#54 RFE: use quilt for %patch management rpm enhancement 04/25/09

It would be nice when 'quilt' (http://savannah.nongnu.org/projects/quilt) is supported for managing the patchsets. When having packages with much patches, it is often difficulty to rediff them as multiple patches might affect one source file and the usual 'gendiff' will fail then.

'quilt' eases rediffing and generation of new patches very much (e.g. see its manpage).

To use 'quilt' a

| %patch1 -p 1 -b .foo

has to be translated to

| quilt import -p 1 %PATCH1 && quilt push

The '-b ...' is to be ignored. The conventional 'patch' method must stay but there should be a way to switch to 'quilt'.

#136 [PATCH] Extract interpreter dependencies from #!/usr/bin/env scripts. rpm enhancement 02/10/10

Here's one possible approach to extract interpreter dependencies from "#!/usr/bin/env foo" scripts, to be applied over the patches in #134 and #135.

The first added sed line finds /usr/bin/foo from "#!/usr/bin/env /usr/bin/foo" (however useless using /usr/bin/env that way might be) and should be uncontroversial.

The 2nd sed line and the path loop might be a bit controversial but is the first straightforward implementation that I thought of; for relative things like "foo" in "#!/usr/bin/env foo" it checks if the "foo" executable is in some of "rpm managed" dirs that are assumed to be in all users' $PATH and emits a dependency on the first found such one.

#822 RFE: add pager to rpm query output rpm enhancement 02/17/11

"rpm -qa"'s output use to be very long so I usually run this with less: rpm -qa | less. It would be nice if rpm query run pager automatically.

Attached patch runs pager according to PAGER environment variable if set or "less" with "FXR" flags.

Note: See TracReports for help on using and creating reports.