seattletestbed / resource Goto Github PK

@vladimir-v-diaz notes that Mac_BSD_resources.py contains lots of out-commented code.

Let's clean things up!

The benchmark script called by the Seattle installer expects a vesselinfo file (usually created by the custom installer builder for the user or the clearinghouse) so that it knows how to partition resources into vessels. The code there intends to inform the user if the file is malformed, but breaks if the file is missing at all. In the latter case, the installer log will contain the following:

New installation, beginning benchmark.
Traceback (most recent call last):
  File "seattleinstaller.py", line 1755, in perform_system_benchmarking
  File "/home/john/hello/trunk/custominstallerbuilder/html/static/installers/base/seattle/seattle_repy/benchmark_resources.py", line 543, in main
IOError: [Errno 2] No such file or directory: 'vesselinfo'
New installation, beginning benchmark.
Traceback (most recent call last):
  File "seattleinstaller.py", line 1755, in perform_system_benchmarking
  File "/home/john/hello/trunk/custominstallerbuilder/html/static/installers/base/seattle/seattle_repy/benchmark_resources.py", line 543, in main
IOError: [Errno 2] No such file or directory: 'vesselinfo'

To give a better indication of what went wrong, the except clause should catch IOError and treat that separately from the other exceptions the code reading in the vesselinfo file can raise when the file is malformed.

It might be a good idea to add probable causes for this error (à la "Trying to install a baseinstaller"?).

Thanks to @aot221 for reporting this issue!

The Mac/BSD resource measurement script expects a sysctl key named kern.ipc.maxsockets to find the maximum number of sockets that a Seattle install should use. On OS X 10.10, this key does not always exist, and the mechanics of when it exists or is missing are unknown currently.

The error is seen during install,

$ ./install.sh
System benchmark starting...
sysctl: unknown oid 'kern.ipc.maxsockets'
Benchmark failed for insockets resource: unable to find max number of sockets
Benchmark failed for outsockets resource: unable to find max number of sockets
The above benchmarking error(s) occurred.
If you choose to continue anyways then default values will be used for failed benchmarks.
Failed.
This install cannot succeed either because required installation info is corrupted or resources are insufficient.
Please email the Seattle project for additional support, and attach the installer_benchmark.log and vesselinfo files, found in the seattle_repy directory, in order to help us diagnose the issue.

David A. Herrera from Universidad Antonio Nariño in Colombia helped developing and testing a patch, which essentially goes and preinitializes the number to 512 (the number found on my OS 10.6.8 box).

Before sending a PR, I would appreciate any Mac users to chime in and report the value found on their machines (if it exists!) by sending the output of

uname -v
sysctl kern.ipc.maxsockets

Thanks everyone for contributing!

The bandwidth code right now is specific to client or server, and only allows a client to check upload speed. We want to have one set of code, written in repy, that would allow one to check both upload and download speed.

Mac_BSD_resources.py uses sysctl hw.physmem to find the size of RAM. Unfortunately, the info is off by a factor of 4 on a MacBook Pro 8,2 running Mac OS X 10.6.8 on which I'm testing this. (At least we are underestimating the size, so this only hurts the Seattle install, but not the resource donor.)

sysctl hw.memsize gives the correct memory size, 8 GB here.

See also https://discussions.apple.com/thread/1753088?start=0&tstart=0

We should test what the different hw. keys return on a couple of machines, and then decide on a fix.