[
https://issues.apache.org/jira/browse/HTTPCLIENT-2068?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Philipp Nanz updated HTTPCLIENT-2068:
-------------------------------------
Description:
Hi there,
I've been playing around with the Caching HttpClient a bit. It works nicely,
but one thing really is of concern for me:
When dealing with lots of identical requests, the cacheDir quickly gets filled
up with copies of all the same files.
I understand that technically these responses have not been invalidated due to
the new request, but from my perspective it would make sense to invalidate them
because they are technically supersded by the new response. Or even better, use
the same file for all responses, as that would save IO on high-throughput
systems.
Thanks in advance for your time and keep up the good work!
was:
Hi there,
I've been playing around with the Caching HttpClient a bit. It works nicely,
but one thing really is of convcern for me:
When dealing with lots of identical requests, the cacheDir quickly gets filled
up with copies of all the same files.
I understand that technically these responses have not been invalidated due to
the new request, but from my perspective it would make sense to invalidate them
because they are technically supersded by the new response. Or even better, use
the same file for all responses, as that would save IO on high-throughput
systems.
Thanks in advance for your time and keep up the good work!
> Caching HtttpClient: Add ability to invalidate cached responses upon
> revalidation
> ---------------------------------------------------------------------------------
>
> Key: HTTPCLIENT-2068
> URL: https://issues.apache.org/jira/browse/HTTPCLIENT-2068
> Project: HttpComponents HttpClient
> Issue Type: Wish
> Components: HttpCache
> Affects Versions: 4.5.12
> Reporter: Philipp Nanz
> Priority: Major
>
> Hi there,
> I've been playing around with the Caching HttpClient a bit. It works nicely,
> but one thing really is of concern for me:
> When dealing with lots of identical requests, the cacheDir quickly gets
> filled up with copies of all the same files.
> I understand that technically these responses have not been invalidated due
> to the new request, but from my perspective it would make sense to invalidate
> them because they are technically supersded by the new response. Or even
> better, use the same file for all responses, as that would save IO on
> high-throughput systems.
> Thanks in advance for your time and keep up the good work!
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
