search for: iterate_dir

...r hangs up, the sys dir / lock is held and the node did not release it which cause the cluster hangs up. root at cvknode-21:~# ps -e -o pid,stat,comm,wchan=WIDE-WCHAN-COLUMN | grep D PID STAT COMMAND WIDE-WCHAN-COLUMN 7489 D jbd2/sdh-621 jbd2_journal_commit_transaction 16218 D ls iterate_dir 16533 D mkdir dlm_wait_for_lock_mastery 31195 D+ ls iterate_dir So the code reviewed, and I found the order of the lock may wrong. In the function dlm_master_request_handler, the resource lock is held and so after the lock of &dlm->master_lock is locked. But in the function dlm_get_...

...r hangs up, the sys dir / lock is held and the node did not release it which cause the cluster hangs up. root at cvknode-21:~# ps -e -o pid,stat,comm,wchan=WIDE-WCHAN-COLUMN | grep D PID STAT COMMAND WIDE-WCHAN-COLUMN 7489 D jbd2/sdh-621 jbd2_journal_commit_transaction 16218 D ls iterate_dir 16533 D mkdir dlm_wait_for_lock_mastery 31195 D+ ls iterate_dir So the code reviewed, and I found the order of the lock may wrong. In the function dlm_master_request_handler, the resource lock is held and so after the lock of &dlm->master_lock is locked. But in the function dlm_get_...

...8>] ? path_openat+0x158/0x680 [ 1080.060279] [<ffffffff8117abb5>] ? free_pages_and_swap_cache+0x95/0xb0 [ 1080.060285] [<ffffffffa0069efb>] ? ext4_readdir+0x17b/0x8d0 [ext4] [ 1080.060289] [<ffffffff811b884a>] ? do_filp_open+0x3a/0x90 [ 1080.060294] [<ffffffff811bafff>] ? iterate_dir+0x9f/0x120 [ 1080.060297] [<ffffffff811bb459>] ? SyS_getdents+0x89/0x100 [ 1080.060300] [<ffffffff811bb150>] ? fillonedir+0xd0/0xd0 [ 1080.060305] [<ffffffff81514a0d>] ? system_call_fast_compare_end+0x10/0x15 [ 1200.060128] INFO: task jbd2/xvda2-8:109 blocked for more than 120 sec...