function does more than add a single file bucket. Pass an empty brigade
to recall_body(), and prepend the result to the existing output brigade,
the next bucket being eos.
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1021917 13f79535-47bb-0310-9956-ffa450edef68
link of the device/inode of the data file to the matching header
file, and to support the option of not writing a data file when
the data file is empty. Refactor the mod_disk_cache code so that
headers are written as late as possible (on commit), allowing the
device and inode of the body to be written to the header. At this
point, writes to the cache are now atomic, without locks.
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1002643 13f79535-47bb-0310-9956-ffa450edef68
mod_cache provider interface which is called to indicate to the
provider that caching is complete, giving the provider the opportunity
to commit temporary files permanently to the cache in an atomic
fashion. Move all "rename" functionality of temporary files to permanent
files within mod_disk_cache from ad hoc locations in the code to the
commit_entity() function. Instead of reusing the same variables for
temporary file handling in mod_disk_cache, introduce separate discrete
structures for each of the three cache file types, the headers file,
vary file and data file, so that the atomic rename of all three file
types within commit_entity() becomes possible. Replace the inconsistent
use of error cleanups with a formal set of pool cleanups attached to
a subpool, which is destroyed on error.
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@997545 13f79535-47bb-0310-9956-ffa450edef68
within the mod_cache provider interface to support an "in" brigade
and an "out" brigade instead of just a single input brigade. This
gives a cache provider the option to consume only part of the brigade
passed to it, rather than the whole brigade as was required before.
This fixes an out of memory and a request timeout condition that would
occur when the original document was a large file. Update the
mod_disk_cache provider implementation to take into account the new API.
Introduce CacheReadSize and CacheReadTime directives to mod_disk_cache
to control the amount of data to attempt to cache before sending the
data on to the client in the "out" brigade.
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@996395 13f79535-47bb-0310-9956-ffa450edef68
a 206 Partial Content. This stops a reverse proxied partial response
from becoming cached, and then being served in subsequent responses.
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@951222 13f79535-47bb-0310-9956-ffa450edef68
Instead move the check to mod_disk_cache. This leaves cache implementations
the flexibility to implement caching of incomplete responses.
PR: 15866
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@821763 13f79535-47bb-0310-9956-ffa450edef68
freshness tests which will attempt to revalidate the content.
* modules/cache/mod_disk_cache.c
(open_entity): If the root Vary index is stale, we don't need to throw away
potentially stale content - we can try to refresh it.
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@539054 13f79535-47bb-0310-9956-ffa450edef68
let us later throw away the renegotiated content due to a name mismatch.
We do not need to update the ->name field just before writing the headers as
the computed ->name on open and create is sufficient. Updating ->name here is
essentially a no-op - except in the case of Vary where the key has internally
changed (to account for the Vary prefix), but the original name has not.
However, writing the Vary'd-accounted key means that when we read the cache
next that we will error out and disregard the just renegotiated response and
fetch the content all over again - oops!
(This is largely a case where I think we may have overthought ourselves; hence
the explanation is here in the commit rather than in the file itself.)
* modules/cache/mod_disk_cache.c
(store_headers): The originally opened ->name is sufficient.
* CHANGES: Combined with r538992, note that renegotiation for Vary's seem to
work far better.
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@538997 13f79535-47bb-0310-9956-ffa450edef68
the refreshed .vary structure in the right directory. Otherwise, we'll create
a new .vary layout underneath the already existing one - oops!
* modules/cache/mod_disk_cache.c
(store_headers): Temporarily clear dobj->prefix if it is set to ensure that
we create the .vary structure underneath the cache root.
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@538992 13f79535-47bb-0310-9956-ffa450edef68
hook, giving each cache backend the ability to make a better
decision as to how it will allocate the tasks of writing to the
cache and writing to the network. Previously the write to the
cache task needed to be complete before the same brigade was
written to the network, and this caused timing and memory issues
on large cached files. This fix replaces the previous fix for
PR39380.
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@468373 13f79535-47bb-0310-9956-ffa450edef68
close/flush, remove some unneccessary code duplication instead
of calling the right helper in replace_brigade_with_cache().
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@462601 13f79535-47bb-0310-9956-ffa450edef68
for the CacheMaxFileSize and CacheMinFileSize parameters in the
config file. PR39380 [Niklas Edmundsson <nikke acc.umu.se>]
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@450042 13f79535-47bb-0310-9956-ffa450edef68
