sön 2012-12-02 klockan 22:27 -0700 skrev Alex Rousskov:
My understanding is that the bug is in Squid store rebuilding(?) code
that assumes that Squid will have time to do some I/O before the next
timed event is fired, but actually there is a lot of non-I/O work to do,
In such case the bug is
On 30.11.12 17:18, Alex Rousskov wrote:
but categorizing a response before it enters the cache does make a lot
of sense in many cases.
It does, except you don't necessarily know what categorisation is
required at that point. Ideally it would be nice to categorise a
request in
On 12/03/2012 01:44 AM, Henrik Nordström wrote:
The store rebuilding does
eventAdd(storeRebuild, RebuildStep, rb, 0.01, 1);
which says both
delay = 0.01, avoid saturating things too much
weight = 1, this is a heavy job, slow down event.cc processing when this
is seen. Only
On 11/24/2012 04:23 PM, Tsantilas Christos wrote:
A new patch.
This patch includes most (/all I hope) of Amos requested changes.
If there is no objection I will commit the latest patch
(cert_validator-v3.patch) to trunk.
On 11/22/2012 02:10 PM, Amos Jeffries wrote:
On 22/11/2012 9:14
mån 2012-12-03 klockan 09:07 -0700 skrev Alex Rousskov:
Yes. IIRC, it has been broken before my AsyncCall changes to the main
loop.
The sawActivity code is from the AsyncCall merge in rev 8812.
This will poll non-comm event loops until they are completely idle
having nothing to do.
On 12/03/2012 02:01 PM, Henrik Nordström wrote:
mån 2012-12-03 klockan 09:07 -0700 skrev Alex Rousskov:
I was going to agree with that wholeheartedly, but then I thought about
signals. If we visit each non-waiting engine once, then we will only
process signals once per main loop step. Is that
mån 2012-12-03 klockan 15:53 -0700 skrev Alex Rousskov:
True (although the current small maximum select(2) delay is a result of
bugs elsewhere in the code and should not really be there). If we can do
the right thing here easily, we should (instead of adding more future
problems). And I think
On 04.12.2012 05:07, Alex Rousskov wrote:
On 12/03/2012 01:44 AM, Henrik Nordström wrote:
The store rebuilding does
eventAdd(storeRebuild, RebuildStep, rb, 0.01, 1);
which says both
delay = 0.01, avoid saturating things too much
weight = 1, this is a heavy job, slow down event.cc
mån 2012-12-03 klockan 15:53 -0700 skrev Alex Rousskov:
We just need to make sure that heavy events (certain timed events and
signals) interrupt the sawActivity loop. This will fix the issue you are
facing without introducing new problems.
Another question on this.. how to fit the concept of
mån 2012-12-03 klockan 09:07 -0700 skrev Alex Rousskov:
Could you please fix this properly by adding
AsyncEngine::timeTillNextEvent() or similar API and removing loop_delay
manipulation from checkEvents() as discussed above?
The attached patch adds AsyncEngine::timeTillNextEvent() API. It
On 12/03/2012 04:16 PM, Henrik Nordström wrote:
mån 2012-12-03 klockan 15:53 -0700 skrev Alex Rousskov:
Ah, I see another problem with the each non-waiting engine runs once
approach. One engine may create work for other engines or for itself,
including async calls and run this now lightweight
On 12/03/2012 04:36 PM, Henrik Nordström wrote:
mån 2012-12-03 klockan 15:53 -0700 skrev Alex Rousskov:
We just need to make sure that heavy events (certain timed events and
signals) interrupt the sawActivity loop. This will fix the issue you are
facing without introducing new problems.
On 12/03/2012 04:18 PM, Amos Jeffries wrote:
Would it make sense to do one of:
* have the AsyncCall queue drain with a timeout. So that no sequence of
calls can lag any of the sensitive engines too much. That along with the
event engines scheduling their heavyweight actions as Calls instead
13 matches
Mail list logo