search for: j_commit_sequ

...Lustre: Failing over pfs1work-OST0026 [44882.674578] Lustre: *** setting obd pfs1work-OST0026 device ''unknown-block(9,7)'' read-only *** (gdb) l *(log_wait_commit+0xa3) 0xffffffff802e9f76 is in log_wait_commit (fs/jbd/journal.c:550). 545 while (tid_gt(tid, journal->j_commit_sequence)) { 546 jbd_debug(1, "JBD: want %d, j_commit_sequence=%d\n", 547 tid, journal->j_commit_sequence); 548 wake_up(&journal->j_wait_commit); 549 spin_unlock(&journal->j_state...

This version fixes three bugs in the 2nd patch of this series that caused kernel BUG when the system was under race. We weren't accounting with t_oustanding_credits correctly, and there were race conditions caused by the fact the I had overlooked the fact that __jbd2_log_wait_for_space() and jbd2_get_transaction() requires j_state_lock to be write locked. Theodore Ts'o (3): jbd2: Use

...quence=83, commit_request=84 Oct 7 20:31:46 fargo kernel: (journal.c, 205): kjournald: OK, requests differ Oct 7 20:31:46 fargo kernel: (commit.c, 78): journal_commit_transaction: JBD: starting commit of transaction 84 Oct 7 20:31:46 fargo kernel: (journal.c, 544): log_wait_commit: JBD: want 84, j_commit_sequence=83 Thanks David Gómez "The question of whether computers can think is just like the question of whether submarines can swim." -- Edsger W. Dijkstra