You are viewing a plain text version of this content. The canonical link for it is here.
Posted to bugs@httpd.apache.org by bu...@apache.org on 2006/09/14 10:59:34 UTC
DO NOT REPLY [Bug 39380] - mod_disk_cache eats memory, has no LFS support, etc
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG�
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://issues.apache.org/bugzilla/show_bug.cgi?id=39380>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND�
INSERTED IN THE BUG DATABASE.
http://issues.apache.org/bugzilla/show_bug.cgi?id=39380
nikke@acc.umu.se changed:
What |Removed |Added
----------------------------------------------------------------------------
Attachment #18155|0 |1
is obsolete| |
Attachment #18157|0 |1
is obsolete| |
Attachment #18158|0 |1
is obsolete| |
Attachment #18159|0 |1
is obsolete| |
------- Additional Comments From nikke@acc.umu.se 2006-09-14 08:59 -------
Created an attachment (id=18860)
--> (http://issues.apache.org/bugzilla/attachment.cgi?id=18860&action=view)
httpd 2.2.3 - mod_disk_cache jumbo patch - lfs/diskformat/read-while-caching
etc.
The condensed version of what this patch does is the following:
* Implement Large File Support (LFS).
* Realise that files are files, by:
- copy files instead of reading them into memory or eat your mmap
space (depending on whether you have mmap enabled or not). This
caused segfaults on 32bit machines when caching LFS-files.
- cache files and URLs pointing to files separately, so if you have
a bunch of vhosts allowing access to a DVD image you'll only get
one copy in cache.
* Only initiate one caching session per file, originally all sessions
initiated separate caching sessions until the file was cached.
* When a file is being cached, other clients are served the data that has
been cached so far.
* Don't throw away a painfully cached file just because the connection
was aborted, since it's a file the result is valid anyway.
* Allow the caching of a file to happen in the background, the
original behaviour is to cache the whole file and then return data
to the client. This is bad for DVD-isos and other large files.
* Rerarrange the on-disk structure to allow the above items. The big
improvement here is to cache both header and body in the same file,
the original has separate files which are a hassle to do atomic
operations on without locks and also makes cleaning the cache
unneccesarily hard.
* As a side effect of the disk format change, %f in LogFormat now
works.
* Add option to not try to remove directories in the cache structure.
* Lots of code reorganisation and error handling to make it all
possible.
--
Configure bugmail: http://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org