Ticket #284: 0001-Work-around-for-the-random-rm-deaths.patch

File 0001-Work-around-for-the-random-rm-deaths.patch, 1.0 KB (added by cmetcalf, 4 years ago)

Fix for this issue without --spew or global usepty=0 hacks

  • buildbot/slave/commands.py 

    From b079eedb1ba987705f7c4e57aa6bce5b445d5e8b Mon Sep 17 00:00:00 2001
From: Chad Metcalf <cmetcalf@archrock.com>
Date: Mon, 3 Nov 2008 20:03:09 -0800
Subject: [PATCH] Work around for the random rm deaths

Work around for bug reported in http://buildbot.net/trac/ticket/198
Specifically a solution for http://buildbot.net/trac/ticket/284
---
 buildbot/slave/commands.py |    2 ++
 1 files changed, 2 insertions(+), 0 deletions(-)

diff --git a/buildbot/slave/commands.py b/buildbot/slave/commands.py
index 32a8fdb..ca96bea 100644
    a b class SourceBase(Command): 
    13921392        command = ["rm", "-rf", d] 
    13931393        c = ShellCommand(self.builder, command, self.builder.basedir, 
    13941394                         sendRC=0, timeout=self.timeout) 
     1395        # Work around for Bug # 
     1396        c.usePTY = False 
    13951397        self.command = c 
    13961398        # sendRC=0 means the rm command will send stdout/stderr to the 
    13971399        # master, but not the rc=0 when it finishes. That job is left to