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-13546 Lustre <p>This issue was created by maloo for S Buisson &lt;sbuisson@ddn.com&gt;</p> <p>This issue relates to the following test suite run: <a href="https://testing.whamcloud.com/test_sets/7e33be5d-a239-48ba-8938-17e0f334cb59" class="external-link" target="_blank" rel="nofollow noopener">https://testing.whamcloud.com/test_sets/7e33be5d-a239-48ba-8938-17e0f334cb59</a></p> <p>test_4 failed with the following error:</p> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre>mmap_sanity test failed </pre> </div></div> <p>This issue happens with client on CentOS 8. The mmap command fails with:</p> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre>mmap test8: SIGBUS for beyond file size (FAIL, 0.003286s) </pre> </div></div> <p>VVVVVVV DO NOT REMOVE LINES BELOW, Added by Maloo for auto-association VVVVVVV<br/> sanity-pcc test_4 - mmap_sanity test failed</p> LU-13546

sanity-pcc test_4: mmap_sanity test failed

Bug Major Open Unresolved Qian Yingjin Maloo always_except Mon, 11 May 2020 08:22:38 +0000 Mon, 10 Jan 2022 07:32:24 +0000 0 4 <p>Sebastien Buisson (sbuisson@ddn.com) uploaded a new patch: <a href="https://review.whamcloud.com/38564" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/38564</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-13546" title="sanity-pcc test_4: mmap_sanity test failed" class="issue-link" data-issue-key="LU-13546">LU-13546</a> tests: test sanity-pcc on CentOS 8<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: 1<br/> Commit: 27d41717cc961693b87068c626df1ad0b492b791</p> <p>Patch #38564 confirms that sanity-pcc test_4 fails on CentOS 8 with tip of master branch.<br/> I am increasing priority of this ticket. Should we push a patch to skip this test when client runs CentOS 8?</p> <p>I run the command using mmap_sanity as follows in local Ext4 filesystem, it also can not pass the mmap_sanity tst8 on CentOS 8. </p> <div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"> <pre class="code-java"> [root@bing tests]# ./mmap_sanity -d /root/work/debug/ -m /root/work/debug/ -e 1 -e 2 -e 3 -e 4 -e 5 -e 6 -e 7 mmap test1: basic mmap operation (SKIPPED, 0s) mmap test2: MAP_PRIVATE not write back (SKIPPED, 0s) mmap test3: concurrent mmap ops on two nodes (SKIPPED, 0s) mmap test4: c1 write to f1 from mmapped f2, c2 write to f1 from mmapped f1 (SKIPPED, 0s) mmap test5: read/write file to/from the buffer which mmapped to just <span class="code-keyword">this</span> file (SKIPPED, 0s) mmap test6: check mmap write/read content on two nodes (SKIPPED, 0s) mmap test7: file i/o with an unmapped buffer (SKIPPED, 0s) mmap test8: SIGBUS <span class="code-keyword">for</span> beyond file size (FAIL, 0.001586s) </pre> </div></div> <p>While the result on CentOS 7 with local Ext4 filesystem is as follows:</p> <div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"> <pre class="code-java"> [root@qian tests]# ./mmap_sanity -d /root/work/debug -m /root/work/debug -e 1 -e 2 -e 3 -e 4 -e 5 -e 6 -e 7 mmap test1: basic mmap operation (SKIPPED, 0s) mmap test2: MAP_PRIVATE not write back (SKIPPED, 0s) mmap test3: concurrent mmap ops on two nodes (SKIPPED, 0s) mmap test4: c1 write to f1 from mmapped f2, c2 write to f1 from mmapped f1 (SKIPPED, 0s) mmap test5: read/write file to/from the buffer which mmapped to just <span class="code-keyword">this</span> file (SKIPPED, 0s) mmap test6: check mmap write/read content on two nodes (SKIPPED, 0s) mmap test7: file i/o with an unmapped buffer (SKIPPED, 0s) mmap test8: SIGBUS <span class="code-keyword">for</span> beyond file size (PASS, 0.079921s) mmap test9: SIGBUS <span class="code-keyword">for</span> negative file offset (PASS, 4.1e-05s) </pre> </div></div> <p>Same results for Lustre with CentOS 8 and CentOS 7:</p> <div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"> <pre class="code-java"> <span class="code-comment">// CentOS8 on Luste </span>[root@bing tests]# ./mmap_sanity -d /mnt/lustre -m /mnt/lustre -e 1 -e 2 -e 3 -e 4 -e 5 -e 6 -e 7 mmap test1: basic mmap operation (SKIPPED, 0s) mmap test2: MAP_PRIVATE not write back (SKIPPED, 0s) mmap test3: concurrent mmap ops on two nodes (SKIPPED, 0s) mmap test4: c1 write to f1 from mmapped f2, c2 write to f1 from mmapped f1 (SKIPPED, 0s) mmap test5: read/write file to/from the buffer which mmapped to just <span class="code-keyword">this</span> file (SKIPPED, 0s) mmap test6: check mmap write/read content on two nodes (SKIPPED, 0s) mmap test7: file i/o with an unmapped buffer (SKIPPED, 0s) mmap test8: SIGBUS <span class="code-keyword">for</span> beyond file size (FAIL, 0.009503s) [root@bing tests]# df Filesystem                  1K-blocks      Used Available Use% Mounted on devtmpfs                      3975676         0   3975676   0% /dev tmpfs                         3988784         0   3988784   0% /dev/shm tmpfs                         3988784      9140   3979644   1% /run tmpfs                         3988784         0   3988784   0% /sys/fs/cgroup /dev/sda1                    46183252  10919180  32888392  25% / vmhgfs-fuse                 488245288 238522632 249722656  49% /mnt/hgfs tmpfs                          797756         0    797756   0% /run/user/0 /dev/mapper/mds1_flakey        125368      1956    112176   2% /mnt/lustre-mds1 /dev/mapper/ost1_flakey        325368     13508    284700   5% /mnt/lustre-ost1 /dev/mapper/ost2_flakey        325368     13508    284700   5% /mnt/lustre-ost2 192.168.150.131@tcp:/lustre    650736     27016    569400   5% /mnt/lustre [root@bing tests]# </pre> </div></div> <div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"> <pre class="code-java"> <span class="code-comment">// CentOS7 on Lustre </span>[root@qian tests]# ./mmap_sanity -d /mnt/lustre -m /mnt/lustre -e 1 -e 2 -e 3 -e 4 -e 5 -e 6 -e 7 mmap test1: basic mmap operation (SKIPPED, 0s) mmap test2: MAP_PRIVATE not write back (SKIPPED, 0s) mmap test3: concurrent mmap ops on two nodes (SKIPPED, 0s) mmap test4: c1 write to f1 from mmapped f2, c2 write to f1 from mmapped f1 (SKIPPED, 0s) mmap test5: read/write file to/from the buffer which mmapped to just <span class="code-keyword">this</span> file (SKIPPED, 0s) mmap test6: check mmap write/read content on two nodes (SKIPPED, 0s) mmap test7: file i/o with an unmapped buffer (SKIPPED, 0s) mmap test8: SIGBUS <span class="code-keyword">for</span> beyond file size (PASS, 0.026712s) mmap test9: SIGBUS <span class="code-keyword">for</span> negative file offset (PASS, 0.004452s) [root@qian tests]# df Filesystem                  1K-blocks      Used Available Use% Mounted on /dev/sda1                    52507040  18504952  31311836  38% / devtmpfs                      1920116         0   1920116   0% /dev tmpfs                         1930752         0   1930752   0% /dev/shm tmpfs                         1930752     11764   1918988   1% /run tmpfs                         1930752         0   1930752   0% /sys/fs/cgroup .host:/                     488245288 238528028 249717260  49% /mnt/hgfs tmpfs                          386152         0    386152   0% /run/user/0 /dev/mapper/mds1_flakey        125368      1956    112176   2% /mnt/lustre-mds1 /dev/mapper/ost1_flakey        325368     13508    284700   5% /mnt/lustre-ost1 /dev/mapper/ost2_flakey        325368     13508    284700   5% /mnt/lustre-ost2 192.168.150.128@tcp:/lustre    650736     27016    569400   5% /mnt/lustre </pre> </div></div> <p>It seems In CentOS8 kernel, if the file rang is unmapped, it will not trigger SIGBUS when reading beyond file size.</p> <p> </p> <p>I will exclude sanity_mmap tst8 from the  testing list.</p> <p> </p> <p>It seems that the current Linux kernel does not follow POSIX syntax:</p> <p><a href="https://pubs.opengroup.org/onlinepubs/9699919799/" class="external-link" target="_blank" rel="nofollow noopener">https://pubs.opengroup.org/onlinepubs/9699919799/</a></p> <div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"> <pre class="code-java"> The mmap() function can be used to map a region of memory that is larger than the current size of the object. Memory access within the mapping but beyond the current end of the underlying objects may result in SIGBUS signals being sent to the process. The mmap() function can be used to map a region of memory that is larger than the current size of the object. Memory access within the mapping but beyond the current end of the underlying objects may result in SIGBUS signals being sent to the process. The reason <span class="code-keyword">for</span> <span class="code-keyword">this</span> is that the size of the object can be manipulated by other processes and can change at any moment. The implementation should tell the application that a memory reference is outside the object where <span class="code-keyword">this</span> can be detected; otherwise, written data may be lost and read data may not reflect actual data in the object.</pre> </div></div> <p>It also failed mmap_sanity tst 9 for the PCC backend formatted as Ext4 due to the too large offset.</p> <div class="code panel" style="border-width: 1px;"><div class="codeContent panelContent"> <pre class="code-java"> [root@bing tests]# ./mmap_sanity -d /root/work/debug -m /root/work/debug -e 7 -e 8 mmap test1: basic mmap operation (PASS, 9e-05s) mmap test2: MAP_PRIVATE not write back (PASS, 5.2e-05s) mmap test3: concurrent mmap ops on two nodes (PASS, 2.002s) mmap test4: c1 write to f1 from mmapped f2, c2 write to f1 from mmapped f1 (PASS, 2.0027s) mmap test5: read/write file to/from the buffer which mmapped to just <span class="code-keyword">this</span> file (PASS, 0.000589s) mmap test6: check mmap write/read content on two nodes (PASS, 0.00016s) mmap test7: file i/o with an unmapped buffer (SKIPPED, 0s) mmap test8: SIGBUS <span class="code-keyword">for</span> beyond file size (SKIPPED, 0s) mmap: Value too large <span class="code-keyword">for</span> defined data type mmap test9: SIGBUS <span class="code-keyword">for</span> negative file offset (FAIL, 4.9e-05s) </pre> </div></div> <p>Yingjin Qian (qian@ddn.com) uploaded a new patch: <a href="https://review.whamcloud.com/38598" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/38598</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-13546" title="sanity-pcc test_4: mmap_sanity test failed" class="issue-link" data-issue-key="LU-13546">LU-13546</a> pcc: exclude mmap_sanity tst8/tst9 from test list<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: 1<br/> Commit: 11f9c5e7d8454a6e4d47b6ebb3f9c01d97544b58</p> <p>Oleg Drokin (green@whamcloud.com) merged in patch <a href="https://review.whamcloud.com/38598/" class="external-link" target="_blank" rel="nofollow noopener">https://review.whamcloud.com/38598/</a><br/> Subject: <a href="https://jira.whamcloud.com/browse/LU-13546" title="sanity-pcc test_4: mmap_sanity test failed" class="issue-link" data-issue-key="LU-13546">LU-13546</a> pcc: exclude mmap_sanity tst8/tst9 from test list<br/> Project: fs/lustre-release<br/> Branch: master<br/> Current Patch Set: <br/> Commit: 9611b3c3f47a486adc41b2d5370561e94dcea753</p> Development Rank 1|i01027: Rank (Obsolete) 9223372036854775807 Severity