On 5/15/06, <b class="gmail_sendername">Tim Peters</b> &lt;<a href="mailto:tim.peters@gmail.com">tim.peters@gmail.com</a>&gt; wrote:<div><span class="gmail_quote"></span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
[Tim]<br>&gt; &quot;Something went wrong&quot; here on Windows.&nbsp;&nbsp;The Windows buildbot slaves<br>&gt; other than mine eventually died with:<br>&gt;<br>&gt; &quot;&quot;&quot;<br>&gt; ...<br>&gt; test_tarfile<br>&gt;<br>&gt; command timed out: 1200 seconds without output
<br>&gt; &quot;&quot;&quot;:<br>&gt;<br>&gt; I was working on my box at the time, and it became dramatically<br>&gt; unusable:&nbsp;&nbsp;took about two minutes just to swap enough of the OS back<br>&gt; in to _bring up_ Task Manager to see what was going on. ...
<br><br>OK, I fixed that, but unfortunately the checkin comment was correct:<br><br>&nbsp;&nbsp;&nbsp;&nbsp;The Windows buildbot slaves shouldn't swap themselves to death<br>&nbsp;&nbsp;&nbsp;&nbsp;anymore.&nbsp;&nbsp;However, test_tarfile may still fail because of a<br>
&nbsp;&nbsp;&nbsp;&nbsp;temp directory left behind from a previous failing run.<br>&nbsp;&nbsp;&nbsp;&nbsp;Windows buildbot owners may need to remove that directory<br>&nbsp;&nbsp;&nbsp;&nbsp;by hand.<br><br>I did that on my box, and test_tarfile passes there now, but<br>test_tarfile is still failing on other Windows slaves; 
e.g.,<br><br><a href="http://www.python.org/dev/buildbot/all/x86%20W2k%20trunk/builds/694/step-test/0">http://www.python.org/dev/buildbot/all/x86%20W2k%20trunk/builds/694/step-test/0</a><br><br>...<br><br>WindowsError: [Error 183] Cannot create a file when that file already exists:
<br> 'c:\\docume~1\\trentm~1.act\\locals~1\\temp\\testtar.dir\\directory'<br>_______________________________________________<br><br>
</blockquote></div><br>
Ignorant question probably, but why can't the test just check for the directory first, and remove it if it exists?<br>
<br>
-Brett<br>