yellow team mailing list archive

[Gary Poster <gary.poster@xxxxxxxxxxxxx>]

We're not out of the woods yet.

After an initial failure because I did not choose a machine with enough 
RAM, I chose a machine that most definitely did have enough RAM 
(c1.xlarge vs. m2.4xlarge fwiw; these are the only two choices with 8 
cores that I found in the docs).  From that and a bit more digging on 
the second very very big machine, it seems that 8 processes running LP 
need a machine with at least 10.5G RAM.  Good to know.
So, I started a test run.  They ran successfully for about two and a 
half hours and then everything...stopped.  It's the read/select loop 
pairing that Benji and I have seen before.  So, we need to figure that 
out and fix it.
I've attached a zip of the test lists and a zip of the results that I 
got before everything hung, as found in the build/.testrepository/tmp... 
file.
So...that's the next thing to tackle.

Gary

Attachment: results.zip
Description: Zip archive

Attachment: testlists.zip
Description: Zip archive