chpox provides transparent checkpointing and restarting of processes on Linux clusters. It was originally designed for recovering tasks that have a long execution time (i.e. numerical simulations) in case of system crashes, power failures, etc. It may work with openMosix, is SMP safe, does not require program recompiling/relinking, and supports virtual memory, regular open files, pipes, Unix domain sockets, current directory, and child processes.
|Tags||Clustering/Distributed Networks Recovery Tools|
|Operating Systems||POSIX Linux|
Release Notes: Support for Linux kernel version 2.6.17 was added. The user-space interface was improved. Shared file descriptors/tables are now supported. Some bugfixes were made.
Release Notes: This release fixes errors that prevented chpox from compiling with newer versions of binutils. Linux kernel version 2.4.32 is now supported.
Release Notes: Building for openMosix was fixed. Checkpoint and restart to and from pipes was fixed.
Release Notes: Support for the s390 and s390x architectures was added. A new dump file format was introduced. Chpox now correctly saves and restores processes that use GS and FS registers for TLS on the i386 architecture. Other bugfixes and security fixes were made.
Release Notes: Bugfixes related to locking orger and /proc entry handling were made. The module vmadump has been updated to one from bproc-4.0.0pre3. Support for x86-64 was included and support for SPARC was dropped.