[tahoe-dev] Precise Puppy (linux) tahoe-lafs 1.10.0 initial report

Garonda Rodian deepside at hotmail.com
Wed Sep 25 02:49:19 UTC 2013


Initial report from a brand new user on a quest to build minimum expense grids; this is the virtual side of the experiment.

First, let me say the quickstart guide is great; if I get farther along, I may attempt to build a "quick config" menu system/wizard/installer to reduce the need for editing text files for the average user.

VMWare Player 6.0.0, VH 10

NO vmware tools installed

Precise Puppy 5.7.1 (new ISO, md5 c4999c4bd8ca3a8fc935389c2667f848   sha512 d2fa5b6aa599e69a9335c598b7d4bd2311be67c0a27559a9aade3cc9c3f442b03654b3b68564fea5a6304e490aa68ef1f7cebb5611cc7b901ae14fd81c24d92d)

uname -a:
Linux PrecisePuppyBase 3.9.11 #1 SMP Sat July 27 19:40:54 GMT-8 2013i686 i686 i386 GNU/Linux

Intel i7 CPU, 1 vProc, 1vCore

512MB virtual SATA drive, "heavy encryption" persistence file, ext4

Used Puppy Package Manager (GUI) to install python 2.7.3 [gcc 4.6.3]
Build successful


Results of python ./setup.py trial - out of memory below.  
First, Is there a way to run a "storage node only" trial, or an "introducer node only" trial, or a "gateway node only" trial, etc?  
Second, does failing the trial mean that it's unlikely to be stable running for real in any particular node config?

384MB of RAM assigned: FAILED 

512MB of RAM assigned: OK or SKIPPED until 
(hand typed, so forgive any typos)
allmydata.test.test_system
  Connections
    test_rref ...
  SystemTest
    test_debug_trial ...
    test_filesystem ... Node._startService failed, aborting
[Failure instance: Traceback: <type 'exceptions.OSError'>:[Errno 12] Cnanot allocate memory
/usr/lib/python2.7/threading.py:524:__bootstrap
/usr/lib/python2.7/threading.py:551:__bootstrap_inner
/usr/lib/python2.7/threading.py:504:run
--- <exception caught here> ---
/root/spot/Downloads/allmydata-tahoe-1.10.0/support/lib/python2.7/site-packages/Twisted-12.2.0-py2.7-linux-i686.egg/twisted/python/threadpool.py:167:_worker
and so on
and so forth
/root/spot/Downloads/allmydata-tahoe-1.10.0/src/allmydata/util/iputil.py:224:_synchronously_find_addresses_via_config
and so on
/usr/lib/python2.7/subprocesses.py:679:__init__
etc.

4096MB RAM: all OK or SKIPPED; just watching HTOP I didn't see RAM go over 313MB used or so until the test set it failed before, which point it jumped up to 393MB total RAM used, then later on to 531MB total RAM used , but I don't have a real peak.
  End results with 4096MB assigned to the VM: PASSED (skips=7, expectedFailures=3, successes=1124)

P.S. If I'm lucky, the Raspberry Pi has completed its trial run, though if this is the RAM requirement, I'm not holding out much hope.

P.P.S. Has anyone tried tahoe-lafs on the Debian (Ubuntu, if you must) build for an ODROID-U2 or ODROID-XU?  Those look like the logical next step up in the small ARM space.

P.P.P.S. I don't see where to give a storage node either a limit (not reserved[_free]_space, but maximum[_used]_space), or a location.
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://tahoe-lafs.org/pipermail/tahoe-dev/attachments/20130925/17b08720/attachment.html>


More information about the tahoe-dev mailing list