Package: lam4 Version: 7.1.1-3 Severity: normal
When an lam MPI program crashes on our AMD64 multiprocessor machine (dual core dual opteron), lamclean or even lamhalt and lamboot fail to fully clear shared memory semaphores, as can be seen with ipcs. Eventually, shared memory is exhausted and lam fails to run on N processors, although it may still run on N-1 processors. It is necessary to use ipcrm to clear the semaphores before lam works properly again. It would be desirable that lamclean or lamboot should be able to do this without ipcrm or running some cleanup script. Not sure if this is an upstream problem - I only have a Debian multi-processor machine available to test. Thanks. Cliburn -- System Information: Debian Release: 3.1 APT prefers stable APT policy: (500, 'stable'), (50, 'unstable'), (50, 'testing') Architecture: amd64 (x86_64) Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.12-1-amd64-k8-smp Locale: LANG=en_US, LC_CTYPE=en_US (charmap=ISO-8859-1) Versions of packages lam4 depends on: ii debianutils 2.8.4 Miscellaneous utilities specific t ii dpkg 1.10.28 Package maintenance system for Deb ii libc6 2.3.2.ds1-22 GNU C Library: Shared libraries an ii libgcc1 1:3.4.3-13 GCC support library ii libstdc++5 1:3.3.5-13 The GNU Standard C++ Library v3 Versions of packages lam4 recommends: ii lam-runtime 7.1.1-3 LAM runtime environment for execut -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
