Lines Matching +full:system +full:- +full:trace

31 interface to the file system (via the ZFS POSIX layer), to when it completed.
32 This spans everything: block device I/O (disk I/O), file system CPU cycles,
33 file system locks, run queue latency, etc. This is a better measure of the
34 latency suffered by applications reading from the file system than measuring
37 Note that this only traces the common file system operations previously
38 listed: other file system operations (eg, inode operations including
42 A threshold of 0 will trace all operations. Warning: the output will be
43 verbose, as it will include all file system cache hits.
75 XFS file system for testing purposes. More debugging using other tools will
109 A -j option will print just the fields (parsable output, csv):
111 # ./zfsslower -j 1
139 # ./zfsslower -h
140 usage: zfsslower [-h] [-j] [-p PID] [min_ms]
142 Trace common ZFS file operations slower than a threshold
145   min_ms             minimum I/O duration to trace, in ms (default 10)
148   -h, --help         show this help message and exit
149   -j, --csv          just print fields: comma-separated values
150   -p PID, --pid PID  trace this PID only
153     ./zfsslower             # trace operations slower than 10 ms (default)
154     ./zfsslower 1           # trace operations slower than 1 ms
155     ./zfsslower -j 1        # ... 1 ms, parsable output (csv)
156     ./zfsslower 0           # trace all operations (warning: verbose)
157     ./zfsslower -p 185      # trace PID 185 only