Software
Marginal MGTnet links can cause boot failure
Typically, SiCortex systems are configured with one (SC648) or four (SC5832) gigabyte Ethernet ports that connect the SSP to management nodes in the system (m[0,2,4,6]n6 on the SC5832, and m0n6 on the SC648).
Compiler Overview
The following is a list of the compilers, libraries and tools that you can use on your SiCortex system. It does not contain information that nchanges frequently, such as our scientific libraries.
Handling oom/malloc Failures
Because we run individual nodes without swap space, memory demands must be met locally. To do this set the Linux tuning parameters, overcommit_memory and overcommit_ratio, as follows.
Lustre — Mounting Multiple Sicortex File Systems
Collision occurs when two nodes in different SiCortex systems that have the same network id try to mount the external Lustre file system at the same time. Such collisions cause the mount to fail with the message transport endpoint shutdown, which will appear in dmesg on the external server, and in the SSP’s syslog and the node console logs.
Lustre — Reading and Writing the Same File from Two Different Nodes
Writing a file from one node while simultaneously reading it from another causes the file lock to ping-pong rapidly between the two, degrading performance.
What does the Slurm error "User defined signal 1 message" mean?
I started a Slurm job on the SSP and it failed with the message: "User defined signal 1 message."
joe@ssp1 ~/ts_f90 $ srun -p sc0-comp1 -n 1 ts_real_misalign
srun: error: sc0-m0n3: task0: User defined signal 1
What commands is scgcc using"
How do I find out what commands the cross-compile wrappers, such as scgcc, are actually executing?
How do I enable auto-instrumentation?
The PathScale compilers provide a -finstrument-functions flag that is compatible with the GNU -finstrument-functions option and provides the same functionality.
What languages and tools does SiCortex support?
What languages and tools does SiCortex support?
Renaming Symbols with the PathScale Compiler
Renaming symbols is useful when you find you're trying to link to a library that uses a different underscoring convention than the one with which you're trying to compile. More generally, it will allow you to completely rename a symbol in the code you're compiling that is occasionally useful in working around a name collision.
Choosing the Right Copy Command
Do you ever wonder why the version of a directory copied to machine X doesn't work the same way on Y?
Compiler Update
What system code is not GPL or open source?
In general, only license-specific code will be not released as open source.
Is the patched kernel compatible with PVFS?
Why does SiCortex use Lustre?
What kernel patches are you using?
Random number generator
What is your binary compatibility?
Print to Stdout
Origin of SLURM
Node Boot Configuration
Cross-compile Wrapper Commands
Determining Output Host
Linux Kernel Space Usage
