[ 
https://issues.apache.org/jira/browse/CASSANDRA-17473?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17511417#comment-17511417
 ] 

James Brown commented on CASSANDRA-17473:
-----------------------------------------

While {{--hard-dereference}} is definitely cargo-culted from some other script 
(and can be removed), I doubt it's at fault. It is, of course, impossible on a 
POSIX system to find out what the "target" of a hardlink is (without walking 
the entire filesystem's directory tree and building a reverse map of 
inode->dirent), since it's just a ddirentent pointing to the same inode as 
another dirent; as far as I know (and [this random Stack Overflow post 
confirms|https://unix.stackexchange.com/questions/43037/dereferencing-hard-links]),
 {{--hard-dereference}} merely prevents tar from deduplicating hardlinks that 
are in the backup set.

> sstables changing in snapshots
> ------------------------------
>
>                 Key: CASSANDRA-17473
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-17473
>             Project: Cassandra
>          Issue Type: Bug
>            Reporter: James Brown
>            Priority: Normal
>
> We use cassandra snapshots and tar to make full backups of our cassandra 
> clusters. Sometimes, tar fails with a message like
> {{tar: 
> data/addresses/addresses-eb0196100b7d11ec852b1541747d640a/snapshots/backup20220318183708/nb-167-big-Data.db:
>  file changed as we read it}}
> This is kind of strange, since we're reading from a snapshot.
> The (very simplified) relevant snippet looks roughly like
> {code:java}
> nice nodetool "${JMX_ARGS[@]}" snapshot -t "$TAG" "${KEYSPACES[@]}"
> tar --hard-dereference -czpf data///snapshots/"$TAG"/{code}
> This happens maybe 1% of the time when taking backups.
> There are no concurrent snapshots going on, but there are concurrent 
> compactions and repairs, of course. If it matters, this cluster _is_ running 
> incremental repairs.
> This is on Cassandra 4.0.3.
> It seems wrong to me that an sstable could ever be written to while it's in a 
> snapshot.



--
This message was sent by Atlassian Jira
(v8.20.1#820001)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to