-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dmitry Unkovsky wrote: > Probably if you compact view, and then query it with stale=ok and do > periodical queries in 5-10 minutes without stale to reindex updated > documents, things are not so bad. You may request compaction > periodically as well.
That is trying to address the symptoms and not the cause. Additionally because view data is not replicated, you have to address these symptoms on every server/db instance. In the short term my plans are to do the view generation on a separate machine and then use rsync to distribute the couchdb databases and views. I'd rather use CouchDB's builtin functionality for that though since I can then use any topology I want. Note that stale=ok is also not a sufficient solution. You do not know in advance if it will return you stale data, or if CouchDB will decide to update the view (and take ages to do so). In summary views are used as a performance optimization, but in the current implementation consumes far too much elapsed time to generate and far too much disk space. I'm hoping Paul has more "fun" with this fixing the underlying issue :-) Roger -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAktMshcACgkQmOOfHg372QSANQCgqy351uDy1aV3G4Su6yzsZS4B +SQAoLUDqKOjKD3F4SrrRVgBerB1ZCSf =TYrf -----END PGP SIGNATURE-----
