CASSANDRA-20149 Call JMX just once in nodetool listsnapshots #3750
Conversation
…call JMX twice in nodetool listsnapshots This patch also opportunistically adds a new MBean method for SnapshotManager to return true snapshot size for a particular snapshot. patch by Stefan Miklosovic; reviewed by TBD for CASSANDRA-20149
smiklosovic
force-pushed
the
CASSANDRA-20149
branch
from
a04978a to
2a0d19d
Compare
| @@ -74,14 +81,14 @@ public static void from(TableSnapshot details, TabularDataSupport result, Set<St | |||
| { | |||
| try | |||
| { | |||
| final String totalSize = FileUtils.stringifyFileSize(details.computeSizeOnDiskBytes()); | |||
There was a problem hiding this comment.
not sure why we are stressing so much that this has to be final ... doesnt matter at all.
| * @param snapshotName name of snapshot in given keyspace and table to get true size of | ||
| * @return true size of all snapshots in given keyspace and table | ||
| */ | ||
| long getTrueSnapshotsSize(String keyspace, String table, String snapshotName); |
There was a problem hiding this comment.
I noticed that a user does not have any way to get true disk space used for a particular snapshot except reading it from snapshot details (which is in human friendly format btw, not raw) which is overkill so I added this simple utility method for that.
| table.add(indexNames.subList(0, indexNames.size() - 1).toArray(new String[indexNames.size() - 1])); | ||
| else |
There was a problem hiding this comment.
the below logic deals with the fact that we do not want to have another column of "Raw true size" in the output of nodetool listsnapshots. All we are interested in are the numbers we sum up internally and display that in TrueDiskSpaceUsed at then end of the output.
We do have true sizes already returned, but they are in human friendly format and we can not sum these up.
src/java/org/apache/cassandra/service/snapshot/SnapshotDetailsTabularData.java
Show resolved
Hide resolved
Thanks for sending a pull request! Here are some tips if you're new here:
Commit messages should follow the following format:
The Cassandra Jira