OpenSSI 1.0.0 released!!

Hi,

Sorry for the cross post. I came across this on OpenSSI website. I guess others may also be interested.

-aneesh

The OpenSSI project leverages both HP's NonStop Clusters for Unixware technology and other open source technology to provide a full, highly available Single System Image environment for Linux.

Feature list:
1. Cluster Membership
* includes libcluster that application can use
2. Internode Communication

3. Filesystem
* support for CFS over ext3, Lustre Lite
* CFS can be used for the root
* reopen of files, devices, ipc objects when processes move is supported
* CFS supports file record locking and shared writable mapped files
(along with all other standard POSIX capabilities
* HA-CFS is configurable for the root or other filesystems
4. Process Management
* almost all pieces there, including:
o clusterwide PIDs
o process migration and distributed rexec(), rfork() and
migrate() with reopen of files, sockets, pipes, devices, etc.
o vprocs
o clusterwide signalling, get/setpriority
o capabilities
o distributed process groups, session, controlling terminal
o surrogate origin functionality
o no single points of failure (cleanup code to deal with
nodedowns)
o Mosix load leveler (with the process migration model from NSC)
o clusterwide ptrace() and strace
o clusterwide /proc/<pid>, ps, top, etc.

5. Devices
* there is a clusterwide device model via the devfs code
* each node mounts its devfs on /cluster/node#/dev and bind mounts it
to /dev so all devices are visible and accessible from all nodes, but by default you see only local devices
* a process on any node can open a device on any node
* devices are reopened when processes move
* processes retain a context, even if they move; the context
determines which node's devices to access by defaul
6. IPC
* all IPC objects/mechanisms are clusterwide:
o pipes
o fifos
o signalling
o message queues
o semaphore
o shared memory
o Unix-domain sockets
o Internet-domain sockets
* reopen of IPC objects is there for process movement
* nodedown handling is there for all IPC objects
7. Clusterwide TCP/IP
* HA-LVS is integrated, with extensions
* extension is that port redirection to servers in the cluster is
automatic and doesn't have to be managed.
8. Kernel Data Replication Service
* it is in there (cluster/ssi/clreg)
9. Shared Storage
* we have tested shared FCAL and use it for HA-CFS
10. DLM
* is integrated with CLMS and is HA
11. Sysadmin
* services architecture has been made clusterwide
12. Init, Booting and Run Levels
* system runs with a single init which will failover/restart on
another node if the node it is on dies
13. Application Availability
* application monitoring/restart provided by spawndaemon/keepalive
* services started by RC on the initnode will automatically restart on
a failure of the initnode
14. Timesync
* NTP for now
15. Load Leveling
* adapted the openMosix algorithm
* for connection load balancing, using HA-LVS
* load leveling is on by default
* applications must be registered to load level
16. Packaging/Install
* Have source patch, binary RPMs and CVS source options;
* Debian packages also available via ap-get repository.
* First node is incremental to a standard Linux install
* Other nodes install via netboot, PXEboot, DHCP and simple addnode
command;
17. Object Interfaces
* standard interfaces for objects work as expected
* no new interfaces for object location or movement except for
processes (rexec(), migrate(), and /proc/pid/goto to move a process)

Yeah, that's we want, take buggy, unreliable, soon-to-be-removed-from-mainline code and put an entire clustering layer on top of it. Too bad someone is going to need to completely reimplement this "clusterwide device model".

Has someone said that devfs was going to be removed from 2.4? Or are you confusing 2.4 and 2.6 in addition to being nasty and argumentative about one nit in a very large project.

I assume it will be changed in 2.6, if you want to criticise you left off the name of the existing feature in 2.4 which should have been used instead.

   -bill davidsen ([email protected])
"The secret to procrastination is to put things off until the
 last possible moment - but no longer"  -me

I can already hear SCO's lawyers screaming "They are taking technology from UnixWare and incorporating in Linux! Let's sue them!!!"...

That said, this looks really interesting.

Regards: David Weinehall

 /) David Weinehall <[email protected]> /) Northern lights wander      (\
//  Maintainer of the v2.0 kernel   //  Dance across the winter sky //
\)  http://www.acc.umu.se/~tao/    (/   Full colour fire           (/

As briefly mentioned at last week's cluster summit, we'd like to try to integrate the Red Hat (nee Sistina) GDLM, want to give it a try?

"One DLM to rule them all, one DLM to mind them, one DLM to sync them all, and in the cluster, bind them"

Regards,

Daniel