https://issues.apache.org/bugzilla/show_bug.cgi?id=44696
Ruediger Pluem <[EMAIL PROTECTED]> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |NEEDINFO
--- Comment #2 from Ruediger Pluem <[EMAIL PROTECTED]> 2008-08-12 04:36:12 PST
---
The attached patch does not fix all cases:
1. store_body is no atomic operation. The content of the entity may be split
across several brigades. So the race would be only fixed if the headers
would be stored *after* a brigade containing an EOS bucket was passed
to store_body. Keep in mind that store_body is called at 2 different
positions
in the CACHE_SAVE filter: One is used during the first pass of the filter
and
the other one is used during the following passes. But it is not guaranteed
that more than one pass ever happens (e.g. if the first brigade already
contains the EOS bucket).
2. There is still a second race: Lets assume the entity is still fresh if a
request does not contain additional conditions. Now we have a request that
requests a fresh response e.g. via max-age=0. Even if the patch works as
above
there is the possibility that a second client got the old headers decides
the entity is fresh enough for him and then tries to fetch the body which
is now the new one.
--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
