git-svn-id: http://www.varnish-cache.org/svn/trunk@4234 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/trunk@4233 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/trunk@4232 d4fa192b-c00b-0410-8231-f00ffab90ce4

progress.



git-svn-id: http://www.varnish-cache.org/svn/trunk@4231 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/trunk@4230 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/trunk@4229 d4fa192b-c00b-0410-8231-f00ffab90ce4

some significant changes to the policy.



git-svn-id: http://www.varnish-cache.org/svn/trunk@4228 d4fa192b-c00b-0410-8231-f00ffab90ce4

appropriate location for this code.



git-svn-id: http://www.varnish-cache.org/svn/trunk@4227 d4fa192b-c00b-0410-8231-f00ffab90ce4

to fix the discrepancy in the object/objcore count they would otherwise
represent.

XXX: Not quite sure if they should also be counted as a subset of
n_object to give people an easy way to access "total objects available
in cache".  Will decide this later.




git-svn-id: http://www.varnish-cache.org/svn/trunk@4226 d4fa192b-c00b-0410-8231-f00ffab90ce4

what is leaking.



git-svn-id: http://www.varnish-cache.org/svn/trunk@4225 d4fa192b-c00b-0410-8231-f00ffab90ce4

and instead (ab)used the obj.* set.

This allowed, amongst other things, the headers of cached objects
to be munged by subsequent cache-hits.

This was just as bad an idea as it sounds.

The downside was, that we had to reserve object workspace for the
potentially created headers, and this really bites, memory wise,
if your content is many small objects.

This commit allows, and sets obj_workspace to zero, allocating only
as much space as is necessary for the actual headers after vcl_fetch()
has returned.

The downside is, without extra workspace, you can no longer change
the content of obj.* string variables after the vcl_fetch() which
instantiated the object.

For now we still allow the obj_workspace to be set to reserve such
space, but in all likelyhood, this ability will disappear in the
future.





git-svn-id: http://www.varnish-cache.org/svn/trunk@4224 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/trunk@4223 d4fa192b-c00b-0410-8231-f00ffab90ce4

busy cache entry or to ditch it.

Ditch it as soon as we can, so we don't hold other clients on a busy objcore
longer than necessary.



git-svn-id: http://www.varnish-cache.org/svn/trunk@4222 d4fa192b-c00b-0410-8231-f00ffab90ce4

Fixes #549



git-svn-id: http://www.varnish-cache.org/svn/trunk@4221 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/trunk@4220 d4fa192b-c00b-0410-8231-f00ffab90ce4

replace a half-baked one in vtc.c with it.



git-svn-id: http://www.varnish-cache.org/svn/trunk@4219 d4fa192b-c00b-0410-8231-f00ffab90ce4

expiry thread from moving objcores from one list to another.



git-svn-id: http://www.varnish-cache.org/svn/trunk@4218 d4fa192b-c00b-0410-8231-f00ffab90ce4

I can tell FlexeLint to ignore them.



git-svn-id: http://www.varnish-cache.org/svn/trunk@4217 d4fa192b-c00b-0410-8231-f00ffab90ce4

Add a parameter, so the worker thread gets forced to dump stats
into the global pool for every N requests it serves.

This should solve the problem where a wget(1) client does not
show up in the stats, until the session ends.



git-svn-id: http://www.varnish-cache.org/svn/trunk@4216 d4fa192b-c00b-0410-8231-f00ffab90ce4

Move most of the stat fields used in cache_center.c into the workers
private stats, hopefully reducing the lock contention and stats
precision at the same time.



git-svn-id: http://www.varnish-cache.org/svn/trunk@4215 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/trunk@4214 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/trunk@4213 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/trunk@4212 d4fa192b-c00b-0410-8231-f00ffab90ce4

The cruical difference is that we can remove from a VLIST without
having the head of the VLIST, which comes handy when expiring a
non-instantiated object from a -spersistent stevedore.

We flip the LRU list around, so that the frontmost element has not
been accessed for the longest time, and simulate tail-inserts by
putting a senteniel on the tail of the list.

Presently the senteniel is a full objcore, which is a non-issue with
few machines having more than a handful of stevedores, but once the
persistent stevedore moves to per-segment LRU lists, this needs to
be addressed, probably by wrapping the LRU linkage in a separate
structure.


git-svn-id: http://www.varnish-cache.org/svn/trunk@4211 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/trunk@4210 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/trunk@4209 d4fa192b-c00b-0410-8231-f00ffab90ce4

This allows us to put items on a blacklist for a specific backend based on
the response it gave us. The basic syntax is (vcl_fetch):

if (beresp.status == 500) {
	set beresp.saintmode = 20s;
	restart;
}

Health checks are modified accordingly, which means graced objects are used
if no healthy backends are left to try.

git-svn-id: http://www.varnish-cache.org/svn/trunk@4208 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/trunk@4207 d4fa192b-c00b-0410-8231-f00ffab90ce4

and try to push objects away from the tail end so that the list
can be shortened.

It won't work if your bans contains any req.* tests, because the
lurker has no request to compare against.

By default the lurker is disabled, you enable it by setting the
'ban_lurker_sleep' parameter to the amount of time the lurker
should sleep between doing things.

If it cannot do anything, it always sleeps for a second.

NB: This feature is on probation, there is currently no guarantee
it will end up in a future varnish release, it depends on the
impact and usability.  Feedback much anticipated.


One of the quasi-nasty details of this feature, is the need to work
against the lock-order in cache_hash.c, see details in comments for
HSH_FindBan().




git-svn-id: http://www.varnish-cache.org/svn/trunk@4206 d4fa192b-c00b-0410-8231-f00ffab90ce4

the reference.



git-svn-id: http://www.varnish-cache.org/svn/trunk@4205 d4fa192b-c00b-0410-8231-f00ffab90ce4

it was possible to have infinite recursion between vcl_error returning restart and cnt_recv calling vcl_error when you had too many restarts

git-svn-id: http://www.varnish-cache.org/svn/trunk@4204 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/trunk@4203 d4fa192b-c00b-0410-8231-f00ffab90ce4

very first functions ever written in Varnish.   Since I knew
that the file stevedore would always return the size you asked for,
or panic, I didn't bother to actually check how big the storage
segment he allocated was, I just knew it would be the right size.

Guess what, segments are finite size in the -spersistent stevedore
so you may in fact _not_ get what you ask for.

(Cue music: Theme Music from The Keystone Kops)

DuH!


git-svn-id: http://www.varnish-cache.org/svn/trunk@4202 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/trunk@4201 d4fa192b-c00b-0410-8231-f00ffab90ce4

Don't explode manager if it gets overwritten.



git-svn-id: http://www.varnish-cache.org/svn/trunk@4200 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/trunk@4199 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/trunk@4198 d4fa192b-c00b-0410-8231-f00ffab90ce4

Add test-case for same.



git-svn-id: http://www.varnish-cache.org/svn/trunk@4197 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/trunk@4196 d4fa192b-c00b-0410-8231-f00ffab90ce4

git-svn-id: http://www.varnish-cache.org/svn/trunk@4195 d4fa192b-c00b-0410-8231-f00ffab90ce4