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-4157 Lustre <p>Deleting files with the in-kernel client (3.12-rc7) is impossible. The rm command gets stuck at 100%CPU and is unkillable with an example call trace like the following:</p> <p>Oct 23 15:49:05 beo-05 kernel: [ 1361.539903] <span class="error">&#91;&lt;ffffffff815020ff&gt;&#93;</span> ? __schedule+0x2ff/0x8d0<br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.539907] <span class="error">&#91;&lt;ffffffff812567c4&gt;&#93;</span> ? radix_tree_next_chunk+0x1a4/0x210<br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.539911] <span class="error">&#91;&lt;ffffffff810d665a&gt;&#93;</span> ? find_get_pages+0xca/0x150<br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.539914] <span class="error">&#91;&lt;ffffffff810d666a&gt;&#93;</span> ? find_get_pages+0xda/0x150<br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.539917] <span class="error">&#91;&lt;ffffffff810e042d&gt;&#93;</span> ? pagevec_lookup+0x1d/0x30<br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.539921] <span class="error">&#91;&lt;ffffffff810e20f1&gt;&#93;</span> ? truncate_inode_pages_range.part.11+0xa1/0x630<br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.539925] <span class="error">&#91;&lt;ffffffffa09376e5&gt;&#93;</span> ? lmv_lock_match+0xf5/0x2d0 <span class="error">&#91;lmv&#93;</span><br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.539933] <span class="error">&#91;&lt;ffffffffa087dffb&gt;&#93;</span> ? ll_have_md_lock+0x14b/0x3e0 <span class="error">&#91;lustre&#93;</span><br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.539936] <span class="error">&#91;&lt;ffffffff810e26c1&gt;&#93;</span> ? truncate_inode_pages_range+0x41/0x50<br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.539939] <span class="error">&#91;&lt;ffffffff810e2750&gt;&#93;</span> ? truncate_inode_pages+0x10/0x20<br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.539947] <span class="error">&#91;&lt;ffffffffa089d4c7&gt;&#93;</span> ? ll_md_blocking_ast+0x447/0x650 <span class="error">&#91;lustre&#93;</span><br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.539951] <span class="error">&#91;&lt;ffffffff8120314c&gt;&#93;</span> ? fuse_request_send_nowait_locked+0x6c/0xd0<br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.539960] <span class="error">&#91;&lt;ffffffffa05dd177&gt;&#93;</span> ? ldlm_cancel_callback+0x67/0x190 <span class="error">&#91;ptlrpc&#93;</span><br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.539969] <span class="error">&#91;&lt;ffffffffa05e68aa&gt;&#93;</span> ? ldlm_cli_cancel_local+0x7a/0x3c0 <span class="error">&#91;ptlrpc&#93;</span><br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.539979] <span class="error">&#91;&lt;ffffffffa05e8fbd&gt;&#93;</span> ? ldlm_cli_cancel_list_local+0xdd/0x240 <span class="error">&#91;ptlrpc&#93;</span><br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.539989] <span class="error">&#91;&lt;ffffffffa05e9295&gt;&#93;</span> ? ldlm_cancel_resource_local+0x175/0x1e0 <span class="error">&#91;ptlrpc&#93;</span><br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.539995] <span class="error">&#91;&lt;ffffffffa07806d7&gt;&#93;</span> ? mdc_resource_get_unused+0xd7/0x170 <span class="error">&#91;mdc&#93;</span><br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.539998] <span class="error">&#91;&lt;ffffffff810d60c8&gt;&#93;</span> ? filemap_fault+0x88/0x550<br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.540003] <span class="error">&#91;&lt;ffffffffa078170d&gt;&#93;</span> ? mdc_unlink+0x9d/0x4d0 <span class="error">&#91;mdc&#93;</span><br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.540007] <span class="error">&#91;&lt;ffffffffa0944fba&gt;&#93;</span> ? lmv_unlink+0x1ba/0x500 <span class="error">&#91;lmv&#93;</span><br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.540015] <span class="error">&#91;&lt;ffffffffa08a2ff4&gt;&#93;</span> ? ll_unlink+0x164/0x410 <span class="error">&#91;lustre&#93;</span><br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.540018] <span class="error">&#91;&lt;ffffffff81144f1d&gt;&#93;</span> ? vfs_unlink+0x8d/0x100<br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.540022] <span class="error">&#91;&lt;ffffffff81145123&gt;&#93;</span> ? do_unlinkat+0x193/0x230<br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.540025] <span class="error">&#91;&lt;ffffffff81137ce4&gt;&#93;</span> ? vfs_read+0x124/0x170<br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.540029] <span class="error">&#91;&lt;ffffffff8114777d&gt;&#93;</span> ? SyS_unlinkat+0x1d/0x40<br/> Oct 23 15:49:05 beo-05 kernel: [ 1361.540032] <span class="error">&#91;&lt;ffffffff8150a226&gt;&#93;</span> ? system_call_fastpath+0x1a/0x1f</p> Vanilla kernel 3.12-rc7 client mounting a 2.4.1 ZFS server that works fine with a 2.4.1 client. LU-4157

Removing files hangs with 100%CPU on 3.12-rc7 client

Bug Critical Resolved Fixed WC Triage Roland Fehrenbacher Mon, 28 Oct 2013 18:07:01 +0000 Fri, 27 Dec 2013 17:02:38 +0000 Wed, 27 Nov 2013 19:02:54 +0000 Lustre 2.4.1 0 5 <p>It it the same hang I fixed before. The patch is queued by Greg KH in staging tree but not yet in Linus tree.</p> <p>The root cause is because of a generic layer change that makes truncate_inode_pages_range stop truncating the page at index ~0UL but Lustre always put the first page in a dir inode mapping at index ~0UL.</p> <p>See the patch for details <a href="https://git.kernel.org/cgit/linux/kernel/git/gregkh/staging.git/commit/?h=staging-next&amp;id=363090e74f3865c589f4026b40865596b0212f90" class="external-link" target="_blank" rel="nofollow noopener">https://git.kernel.org/cgit/linux/kernel/git/gregkh/staging.git/commit/?h=staging-next&amp;id=363090e74f3865c589f4026b40865596b0212f90</a></p> <p>Thanks Peng. I applied the patch and it fixes the problem indeed. Would be great if it could move to 3.12 final (together with the patch in <a href="https://jira.hpdd.intel.com/browse/LU-4127" class="external-link" target="_blank" rel="nofollow noopener">https://jira.hpdd.intel.com/browse/LU-4127</a> maybe?).</p> <p><a href="https://jira.whamcloud.com/browse/LU-4127" title="Loading fld module crashes kernel in mainline 3.12-rc.5" class="issue-link" data-issue-key="LU-4127"><del>LU-4127</del></a> was also fixed in upstream. See <a href="https://git.kernel.org/cgit/linux/kernel/git/gregkh/staging.git/commit/?h=staging-next&amp;id=86bac591def1a5b7060c0834828b1eaabfe7f0a7" class="external-link" target="_blank" rel="nofollow noopener">https://git.kernel.org/cgit/linux/kernel/git/gregkh/staging.git/commit/?h=staging-next&amp;id=86bac591def1a5b7060c0834828b1eaabfe7f0a7</a></p> <p>The two patches are both queue by Greg in his staging-next branch but I do not know when he will push them to Linus.</p> <p>OK, thanks for letting me know. I assume this will go into a 3.12.x then. Without these patches, the lustre client is absolutely useless.</p> <p>It's fixed in upstream.</p> Related LU-4416 Development Rank 1|hzw707: Rank (Obsolete) 11279 Severity