Summary: | Crash when composing mail for recipient with known expired PGP key | ||
---|---|---|---|
Product: | [Unmaintained] kmail | Reporter: | Stephan Binner <binner> |
Component: | composer | Assignee: | kdepim bugs <kdepim-bugs> |
Status: | RESOLVED WORKSFORME | ||
Severity: | crash | CC: | esken |
Priority: | NOR | ||
Version: | unspecified | ||
Target Milestone: | --- | ||
Platform: | Compiled Sources | ||
OS: | Linux | ||
Latest Commit: | Version Fixed In: | ||
Sentry Crash Report: |
Description
Stephan Binner
2004-06-09 12:14:37 UTC
Stephan, you should really know that a backtrace would be extremly helpful. Since I can't reproduce the crash I can't fix it. I will post a backtrace when you help me to make valgrind print out more than its help on my system. ;-( I traced it now manually to matchingKeys.erase( std::remove_if( matchingKeys.begin(), matchingKeys.end(), NotValidTrustedEncryptionKey ), matchingKeys.end() ); in Kleo::KeyResolver::getEncryptionKeys(). --tool=addrcheck is most likely the option valgrind is trying to help you to :) Here we go: kmail: returned 2 keys ==11744== Warning: client syscall munmap tried to modify addresses 0xCAD0B991-0x82B3A88 ==11744== ==11744== Invalid read of size 4 ==11744== at 0x3D59CC20: fREe (in /opt/kde-head/lib/libkdecore.so.4.2.0) ==11744== by 0x3D59D015: free (in /opt/kde-head/lib/libkdecore.so.4.2.0) ==11744== by 0x3D0F78C4: gpgme_key_unref (key.c:324) ==11744== by 0x3D0E819F: GpgME::Key::~Key() (key.cpp:69) ==11744== Address 0x4816D274 is not stack'd, malloc'd or (recently) free'd *** KMail got signal 11 (Crashing) gpgme 0.3.16 / SUSE 9.1 > gpgme 0.3.16
I assume you are using KMail HEAD? How did you get it to compile against a 0.3-branch gpgme? Try again with >= 0.4.5, please.
> I assume you are using KMail HEAD? Yes. > How did you get it to compile against a 0.3-branch gpgme? That was a mindless mention of the distribution version which is not used. Actually in use is You are missing gpgme 0.4.5 or higher. Gpgme will be built statically from libkdenetwork/libgpgme-copy. that's also why above gpgme version has debug information. Sorry for confusion. :-) Stephan, does current cvs still crash for you? *** Bug 84028 has been marked as a duplicate of this bug. *** With same address and keys from above I just tried and it didn't crash. Till says "close it". :-) |