[CASSANDRA-12596] cosmetic issue with compaction-stress write

• Type: Bug
• Status: Open
• Priority: Low
• Resolution: Unresolved
• Fix Version/s: 3.11.x, 4.0.x
• Component/s: Tool/stress
• Severity: Low
• Assignee: Unassigned
• Reporter: Wei Deng
• Created: 01/Sep/16 23:55
• Updated: 24/May/23 22:12

Description

I got the following when I was trying to run `compaction-stress write` with a 10GB data set. Looks like a race condition with the lifecycle transaction log but it only affects the reporting of the progress temporarily.

Written 50.27GB of 100GB
ERROR 23:49:55,843 Error while getting /mnt/compaction/stresscql/blogposts-606839b5bd683964ad7f57f519eb6dae folder size. {}
java.nio.file.NoSuchFileException: /mnt/compaction/stresscql/blogposts-606839b5bd683964ad7f57f519eb6dae/mc_txn_write_c6877a90-709e-11e6-91a5-a1533b002bc1.log
        at sun.nio.fs.UnixException.translateToIOException(UnixException.java:86) ~[na:1.8.0_66]
        at sun.nio.fs.UnixException.rethrowAsIOException(UnixException.java:102) ~[na:1.8.0_66]
        at sun.nio.fs.UnixException.rethrowAsIOException(UnixException.java:107) ~[na:1.8.0_66]
        at sun.nio.fs.UnixFileAttributeViews$Basic.readAttributes(UnixFileAttributeViews.java:55) ~[na:1.8.0_66]
        at sun.nio.fs.UnixFileSystemProvider.readAttributes(UnixFileSystemProvider.java:144) ~[na:1.8.0_66]
        at sun.nio.fs.LinuxFileSystemProvider.readAttributes(LinuxFileSystemProvider.java:99) ~[na:1.8.0_66]
        at java.nio.file.Files.readAttributes(Files.java:1737) ~[na:1.8.0_66]
        at java.nio.file.FileTreeWalker.getAttributes(FileTreeWalker.java:219) ~[na:1.8.0_66]
        at java.nio.file.FileTreeWalker.visit(FileTreeWalker.java:276) ~[na:1.8.0_66]
        at java.nio.file.FileTreeWalker.next(FileTreeWalker.java:372) ~[na:1.8.0_66]
        at java.nio.file.Files.walkFileTree(Files.java:2706) ~[na:1.8.0_66]
        at java.nio.file.Files.walkFileTree(Files.java:2742) ~[na:1.8.0_66]
        at org.apache.cassandra.io.util.FileUtils.folderSize(FileUtils.java:478) ~[main/:na]
        at org.apache.cassandra.db.Directories.getRawDiretoriesSize(Directories.java:939) [main/:na]
        at org.apache.cassandra.stress.CompactionStress$DataWriter.run(CompactionStress.java:322) [stress/:na]
        at org.apache.cassandra.stress.CompactionStress.main(CompactionStress.java:353) [stress/:na]
Written 37.49GB of 100GB

You will see the next "Written xxxGB" number being inaccurate because it interrupted the directory accounting of the data files, but the subsequent "Written xxxGB" number will get back to normal.

Since it generates a ERROR log entry it's probably still worthwhile to fix.