From: Brian Warner <warner@allmydata.com>
Date: Thu, 15 Nov 2007 21:55:00 +0000 (-0700)
Subject: test_mutable: workaround: use more peers to avoid random test failures.
X-Git-Url: https://git.rkrishnan.org/junkers?a=commitdiff_plain;h=8a2736480a045e73e0a4dc1373f2300eac166508;p=tahoe-lafs%2Ftahoe-lafs.git

test_mutable: workaround: use more peers to avoid random test failures.

The underlying issue is recorded in #211: one corrupt share in a query
response will cause us to ignore the remaining shares in that response, even
if they are good. In our tests (with N=10 but only 5 peers), this can leave
us with too few shares to recover the file.

The temporary workaround is to use 10 peers, to make sure we never get
multiple shares per response. The real fix will be to fix the control flow.

This fixes #209.
---

diff --git a/src/allmydata/test/test_system.py b/src/allmydata/test/test_system.py
index 2a6bfcfe..91f7d9ee 100644
--- a/src/allmydata/test/test_system.py
+++ b/src/allmydata/test/test_system.py
@@ -311,6 +311,7 @@ class SystemTest(testutil.SignalMixin, unittest.TestCase):
         NEWERDATA = "this is getting old"
 
         d = self.set_up_nodes()
+
         def _create_mutable(res):
             c = self.clients[0]
             log.msg("starting create_mutable_file")