Re: Lower max size of HTTP cache?
Michal Novotny <[email protected]>
| Newsgroups | gmane.comp.mozilla.devel.netlib |
|---|---|
| Message-ID | <[email protected]> |
One more bug needs to be fixed: https://bugzilla.mozilla.org/show_bug.cgi?id=707436 I'd like to postpone evicting of the entries in case of startup and maybe spread the eviction in a longer timerange to avoid massive IO. This would mean that we wouldn't respect the cache size for a limited time. Is this acceptable? Michal On 12/08/2011 12:09 AM, Christian Biesinger wrote: > On Wed, Dec 7, 2011 at 2:57 PM, Jason Duell<[email protected]> wrote: >>> When the cache is found to be corrupt, could the whole directory be >>> renamed, a new cache created and >>> the old one be deleted in the background later? >> That's how it's been done for a long time. We hit a speedbump where the >> renaming of the directory itself takes a long time on NTFS for large >> directories, but that's hopefully completely fixed now. > > So it sounds like the slowdown issues are all fixed? Are there any > other reasons why we would want to decrease the size? > > -christian > _______________________________________________ > dev-tech-network mailing list > [email protected] > https://lists.mozilla.org/listinfo/dev-tech-network