Hi,
I've got a LAMP (Zend) application running on a load balanced cluster.
I am generating Lucene search indexes on one of the machines and rsyncing them to /dev/shm/ on the other machines in the cluster.
My problem is that the --delete function appears not to be working against the /dev/shm filesystem, probably because it's not posix compliant.
Obviously I can't keep rsyncing into /dev/shm if I can't remove the old files, otherwise my machines will run out of memory.
I'm just using rsync -a --delete :
Does anyone know a simple way of getting rsync to perform an equivalent of a --delete on /dev/shm?
(I had attempted removing the switch bit on /dev/shm (usually 1777) and that had made no difference)
Many thanks in advance for any ideas.
rsync --delete issues with /dev/shm
-
- Retired Moderator
- Posts: 18276
- Joined: 2006/12/13 20:15:34
- Location: Tidewater, Virginia, North America
- Contact:
rsync --delete issues with /dev/shm
Welcome to the CentOS fora. Reading [url=https://www.centos.org/modules/newbb/viewforum.php?forum=47]FAQ & Readme First[/url] is recommended for new users.
Perhaps I am missing something obvious due to my ignorance of the finer points of clusters, not to mention Lucene, but why not use shared storage for the search indices rather than rsync?
Perhaps I am missing something obvious due to my ignorance of the finer points of clusters, not to mention Lucene, but why not use shared storage for the search indices rather than rsync?
Re: rsync --delete issues with /dev/shm
In our environment, when we use lucene with a Linux cluster for financial data/news feed crunching, the indices are always stored on a SAN. That seems to be the "easy" way.