<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 12pt;
font-family:Calibri
}
--></style></head>
<body class='hmmessage'><div dir='ltr'>Thank you for the information. <br><br>Can you expand on the memory problems for large mutable files - in particular, what kind of problems, and what is considered "large" - I've always found that to be a very personal judgement call, whether it's in the MB or the PB.<br><br>FYI: 768MB of RAM on the VM results in the same level of success that 4096MB did, so the critical point is somewhere between 512MB and 768MB. <br><br><div>> Date: Wed, 25 Sep 2013 17:38:41 +0100<br>> From: davidsarah@leastauthority.com<br>> To: deepside@hotmail.com<br>> CC: tahoe-dev@tahoe-lafs.org<br>> Subject: Re: [tahoe-dev] Precise Puppy (linux) tahoe-lafs 1.10.0 initial report<br>> <br>> On 25/09/13 03:49, Garonda Rodian wrote:<br>> > Results of python ./setup.py trial - out of memory below. <br>> > First, Is there a way to run a "storage node only" trial, or an "introducer node only"<br>> > trial, or a "gateway node only" trial, etc?<br>> <br>> I'm afraid not; the tests are not organised according to that dimension.<br>> <br>> > Second, does failing the trial mean that it's unlikely to be stable running for real in<br>> > any particular node config?<br>> <br>> Not necessarily; the tests require significantly more memory than actual usage.<br>> The ticket for this is <https://tahoe-lafs.org/trac/tahoe-lafs/ticket/1476>, although<br>> it's short on detail.<br>> <br>> Typically a Tahoe storage process will take no more than around 64 MiB, although there<br>> are some memory problems with handing large mutable files.<br>> <br>> -- <br>> Daira Hopwood ⚥<br>> <br></div> </div></body>
</html>