Bug 85504
| Summary: | rpm produces segmentation fault when installing or removing some packages | ||||||
|---|---|---|---|---|---|---|---|
| Product: | [Retired] Red Hat Linux | Reporter: | Keith Lea <keith> | ||||
| Component: | rpm | Assignee: | Jeff Johnson <jbj> | ||||
| Status: | CLOSED WORKSFORME | QA Contact: | Mike McLean <mikem> | ||||
| Severity: | high | Docs Contact: | |||||
| Priority: | medium | ||||||
| Version: | 8.0 | CC: | pekkas | ||||
| Target Milestone: | --- | ||||||
| Target Release: | --- | ||||||
| Hardware: | i686 | ||||||
| OS: | Linux | ||||||
| Whiteboard: | |||||||
| Fixed In Version: | Doc Type: | Bug Fix | |||||
| Doc Text: | Story Points: | --- | |||||
| Clone Of: | Environment: | ||||||
| Last Closed: | 2003-03-22 20:18:47 UTC | Type: | --- | ||||
| Regression: | --- | Mount Type: | --- | ||||
| Documentation: | --- | CRM: | |||||
| Verified Versions: | Category: | --- | |||||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||
| Cloudforms Team: | --- | Target Upstream Version: | |||||
| Embargoed: | |||||||
| Attachments: |
|
||||||
|
Description
Keith Lea
2003-03-03 20:30:30 UTC
Hmmm, are you using LDAP passwords? If so, you need to run nscd to avoid a segfault that affects statically linked binaries like /bin/rpm. I don't know what LDAP is, so no, I'm not using LDAP passwords. (And there's no administrator who would've installed such a thing for me - this is my own machine.) Are you running nscd? If not, can you start "/sbin/service nscd restart" and repeat? no, it was not running. after starting it I still get the same segfault with the same backtrace running rpm -e freetype-utils. OK. What's in /etc/nssswitch.conf? Try files first, make sure that all the users in the freetype package are known. What glibc? passwd: files nisplus shadow: files nisplus group: files nisplus hosts: files nisplus dns bootparams: nisplus [NOTFOUND=return] files ethers: files netmasks: files networks: files protocols: files nisplus rpc: files services: files nisplus netgroup: files nisplus publickey: nisplus automount: files nisplus aliases: files nisplus -- [root@localhost root]# rpm -q glibc glibc-2.3.1-51 -- I think you're assuming I know more about rpm and ns* than I do. What does "Try files first, make sure that all the users in the freetype package are known" mean? Ah, sorry. The backtrace from getpwnam on is trying to look up a user name for a file through glibc (see man 3 getpwnam). The dl_* routines are an indication that something (like LDAP, one means to look up users, NIS is another, /etc/passwd aka files is a 3rd way) is fubar. A segfault often happens with /bin/rpm, a statically linked binary, which is more sensitive to structure changes (there are a few) and linkage through helper libraries. So, you have files first, which should use /etc/passwd, but there appears to be a further lookup through nisplus. What happens if you edit /etc/nssswitch.conf and delete "nisplus" on the passwd/shadow/group lines? Save a copy, duh ... same problem. do I need to restart anything after editing it? OK. No nothing needs restarting.
An strace should give a hint as to what file is being loaded, and
from that I might be able to guess what is happening.
Can you try
strace -f -o /tmp/xxx the_command_here
and attach /tmp/xxx? Thanks.
Created attachment 90469 [details]
strace output for rpm -e freetype-utils
here you go.
please help me Jeff, I still can't install or remove packages at all :( OK. The strace indicates nscd running with successful
root lookup.
Can you try attaching the output of
strace -o /tmp/xxx rpm -evv freetype-utils
The -vv will help me figger where the program is segfaulting.
if you want the /tmp/xxx too, I can attach that, but i assume it's the same as last time... D: opening db environment /var/lib/rpm/Packages joinenv D: opening db index /var/lib/rpm/Packages rdonly mode=0x0 D: locked db index /var/lib/rpm/Packages D: opening db index /var/lib/rpm/Name rdonly mode=0x0 D: opening db index /var/lib/rpm/Pubkeys rdonly mode=0x0 D: read h# 663 Header sanity check: OK D: ========== DSA pubkey id 219180cddb42a60e D: read h# 819 Header V3 DSA signature: OK, key ID db42a60e D: ========== --- freetype-utils-2.1.2-7 D: opening db index /var/lib/rpm/Requirename rdonly mode=0x0 D: closed db index /var/lib/rpm/Pubkeys D: closed db index /var/lib/rpm/Requirename D: closed db index /var/lib/rpm/Name D: closed db index /var/lib/rpm/Packages D: closed db environment /var/lib/rpm/Packages D: opening db environment /var/lib/rpm/Packages joinenv D: opening db index /var/lib/rpm/Packages create mode=0x42 D: getting list of mounted filesystems D: sanity checking 1 elments D: computing 4 file fingerprints D: computing file dispositions D: opening db index /var/lib/rpm/Basenames create mode=0x42 D: ========== --- freetype-utils-2.1.2-7 D: erase: freetype-utils-2.1.2-7 has 4 files, test = 0 D: opening db index /var/lib/rpm/Name create mode=0x42 D: read h# 819 Header V3 DSA signature: OK, key ID db42a60e If that's all the output, then instance #819
is probably damaged.
Can you give me a pointer (i.e. URL, attachments won't work)
to a tarball of you database
cd /var/lib
tar czvf /tmp/rpmdb-85504.tar.gz rpm
and I'll zap record #819
note that freetype-utils isn't the only package that crashes on erase. http://www.kano.net/in/rpmdb.tar.gz Hmmm, you database looks fine, signatures/digests
are being verified, so it's impossible that the
data is corrupt or modified.
Here's what I see for, say, freetype-utils:
...
D: read h# 819 Header V3 DSA signature: OK, key ID db42a60e
D: +++ h# 484 Header V3 DSA signature: OK, key ID db42a60e
D: adding "freetype-utils" to Name index.
D: adding 4 entries to Basenames index.
D: adding "System Environment/Libraries" to Group index.
D: adding 9 entries to Requirename index.
D: adding "freetype-utils" to Providename index.
D: adding "/usr/bin/" to Dirnames index.
D: adding 9 entries to Requireversion index.
D: adding "2.1.2-7" to Provideversion index.
D: adding 1 entries to Installtid index.
D: adding 1 entries to Sigmd5 index.
D: adding "dd89f5af8c5406f45bfcb2a0360a1fdedc008c01" to Sha1header index.
D: adding 4 entries to Filemd5s index.
So there's something else local to your machine that's causing
the problem
Let's see if this problem can't be bracketed. You're currently
running rpm-4.1-1.06. Can you try rpm-4.1-9 packages from
ftp://people.redhat.com/jbj/test-4.1
You'll have to install manually, so do the following:
cd /var/tmp
<download all the packages, don't forget popt>
mkdir xxx
cd xxx
for i in ../{rpm,popt}-*.rpm
do
echo $i ---
rpm2cpio $i | cpio -dim
done
find . -type d -exec chmod 755 {} \;
tar cf - . | (cd /; tar xvf -)
and see if same segfault?
If segfault then there's something wrong with your pam
modules is my guess.
everything works now :) thanks a ton. is there anything else I should do? If you added "private" to /etc/rpm/macros, then you might want to remove. Otherwise, maybe a --rebuilddb. You ought to figger what happened, however. If a manual rpm2cpio "fix"ed, then your files were damaged somehow. We noticed the same problems with RHL8 rpm. The same procedure for replacing RPM worked fine for us. |