https://jira.whamcloud.com This file is an XML representation of an issue en-us 9.4.14 940014 05-12-2023 https://jira.whamcloud.com/browse/LU-6625 Lustre <p>the online-manual on symlink() tells that symlink() should set errno EEXIST if specified link-path already exists. But it actually sets ENOENT when the last ward of the link-path is "/".</p> <p>It looks to me that the last ward is "/" is awkward. But the directory specified as link-path actually exists so I think we should set errno EEXIST. <br/> Moreover ext3 returns EEXIST.</p> <p>This patch modifies a if-statement in ll_lookup_nd for client to look up MDS properly and for symlink() to set errno EEXIST.</p> <p>By the way, you can see the difference of errno between lustre and ext3 with the attached trivial test program </p> LU-6625

symlink() sets wrong errno, ENOENT

Bug Major Resolved Not a Bug Niu Yawei Hiroya Nozaki patch Thu, 21 May 2015 07:47:40 +0000 Thu, 4 Jun 2015 06:44:55 +0000 Thu, 4 Jun 2015 06:44:55 +0000 0 3 <p>Hiroya Nozaki (nozaki.hiroya@jp.fujitsu.com) uploaded a new patch: <a href="http://review.whamcloud.com/14902" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/14902</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-6625" title="symlink() sets wrong errno, ENOENT" class="issue-link" data-issue-key="LU-6625"><del>LU-6625</del></a> llite: symlink() sets wrong errno, ENOENT<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: 1<br/> Commit: ac5fd86f91923259007a2d1984c44f104f300550</p> <p>The reason of kernel returning -ENOENT for such case is: </p> <p>see lookup_create():</p> <div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"> <pre class="code-java"> /* * Special <span class="code-keyword">case</span> - lookup gave negative, but... we had foo/bar/ * From the vfs_mknod() POV we just have a negative dentry - * all is fine. Let<span class="code-quote">'s be bastards - you had / on the end, you'</span>ve * been asking <span class="code-keyword">for</span> (non-existent) directory. -ENOENT <span class="code-keyword">for</span> you. */ <span class="code-keyword">if</span> (unlikely(!is_dir &amp;&amp; nd-&gt;last.name[nd-&gt;last.len])) { dput(dentry); dentry = ERR_PTR(-ENOENT); } </pre> </div></div> <p>I don't think we need to fix this in Lustre, and actually -ENOENT does make sense to me as the comment said above. <img class="emoticon" src="https://jira.whamcloud.com/images/icons/emoticons/smile.png" height="16" width="16" align="absmiddle" alt="" border="0"/></p> <p>BTW: Because local filesystem needn't do revalidation, ext3 returns -EEXIST for such case.</p> <p>Andreas Dilger (andreas.dilger@intel.com) uploaded a new patch: <a href="http://review.whamcloud.com/15110" class="external-link" target="_blank" rel="nofollow noopener">http://review.whamcloud.com/15110</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-6625" title="symlink() sets wrong errno, ENOENT" class="issue-link" data-issue-key="LU-6625"><del>LU-6625</del></a> obdclass: remove malloc fault injection<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: 1<br/> Commit: 0ec71e5838903eafd54421496ebc5a4a5c924593</p> <p>hm, sounds reasonable.<br/> OK, thank you !! I'm afraid to say but could you please close this ticket as not a bug ??</p> <p>Thank you, Hiroya.</p> Development Rank 1|hzxdrz: Rank (Obsolete) 9223372036854775807 Severity