| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
| |
reference the decisions and factors involved.
BUG=77467
TEST=see bug
Review URL: http://codereview.chromium.org/6748012
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@79478 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Added net::IsLocalhost() function to net/base/net_utils.h
Unit tests for the above. Also fix flakiness in the ReceivedContentMalformed test that was caused by non-zero jitter.
Modify back-off policy to ignore first 4 errors to help avoid back-off from erroneously kicking in on flaky connections. Make maximum back-off period 15 minutes instead of 60. Added documentation of results of analyzing behavior this new policy will give.
Add a simple server for manual testing of the throttling feature.
BUG=66062
TEST=net_unittests
Review URL: http://codereview.chromium.org/6711046
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@79464 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
about:net-internals page:
- Add a new tab to the about:net-internals page.
- Add a preference item to persist user's choice.
- Remove the command line switch --disable-enforced-throttling.
- Make throttling feature off by default for now.
This is 95% based on yzshen@chromium.org's patch
http://codereview.chromium.org/6286001/ which was never landed. The
changes that have been made from that patch are mostly cosmetic, apart
from updating it to match the codebase (the patch was prepared ~2
months ago) and responding to eroman@'s initial round of code review
comments.
BUG=66062
TEST=Make sure the checkbox on about:net-internals > HTTP Throttling is checked.
(1) Type in the addresss bar a URL which is expected to return 5XX.
(2) Keep clicking the "reload" button.
(3) After a few clicks, Chrome displays an error page explaining why a net::ERR_TEMPORARILY_THROTTLED error occurs.
Make sure the checkbox on about:net-internals > HTTP Throttling is unchecked.
Following step (1) and (2) described above won't result in a net::ERR_TEMPORARILY_THROTTLED error page.
Review URL: http://codereview.chromium.org/6677085
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@79455 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
.cc file.
R=willchan@chromium.org
BUG=none
TEST=net_unittests
Review URL: http://codereview.chromium.org/6745014
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@79453 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
BUG=76666
TEST=RenderViewHostManagerTest.ClickLinkAfter204Error
Review URL: http://codereview.chromium.org/6724026
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@79440 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
| |
Also removes several empty and unused gtest test fixtures.
R=willchan@chromium.org
Review URL: http://codereview.chromium.org/6676134
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@79411 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
| |
Added a command-line switch: --enable-ip-pooling
BUG=42669
TEST=SpdySessionTest.IPPool*
Review URL: http://codereview.chromium.org/6594116
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@79372 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
| |
already in the net namespace.
Review URL: http://codereview.chromium.org/6730034
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@79340 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Make serial_number() private because it strips the leading 0 bytes
required for proper DER encoding of the ASN.1 INTEGER type.
Add info on the certificates on the blacklist so we know what hostnames
they can be used for and when they can be safely removed from the blacklist.
Use UMA_HISTOGRAM_ENUMERATION instead of UMA_HISTOGRAM_CUSTOM_COUNTS.
R=agl
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/6696059
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@79237 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
While this slightly degeneralizes some stats-collection code,
the svn history shows that this code has, over several years, only been used for SDCH,
and thus it's worth some degeneralization in order to greatly simplify URLRequestJob.
BUG=none
TEST=try bots
Review URL: http://codereview.chromium.org/6713019
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@79219 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
as not trusted. We should not allow the user to accept the certificate.
A simple way to accomplish that is to treat the certificate as revoked.
R=agl,rsleevi
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/6675030
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@79165 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
| |
BUG=none
TEST=n/a
Review URL: http://codereview.chromium.org/6708091
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@79124 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
| |
This makes it simpler for developers to deal with the release/set event pair generated by setting a cookie that already exists, and gives them more information about general cookie removal (e.g. that the cookie wasn't "removed" actively but expired).
BUG=70101
TEST=net_unittests
Review URL: http://codereview.chromium.org/6698023
Patch from Mike West <mkwst@chromium.org>.
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@79113 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
of StatsCounter, StatsCounterTimer and StatsRate variables to auto. This
change would fix possibel race condition during static's initialization
in multi-threaded environments.
BUG=76980
TEST=unit tests in net, browser and plugins
R=jar@chromium.org
Review URL: http://codereview.chromium.org/6719011
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@79065 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is part of the process to remove refcounting from URLRequestJob.
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/6717012
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@79057 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
| |
R=wtc@chromium.org
Review URL: http://codereview.chromium.org/6715007
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@79009 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
This is a second try, my first pass in Issue 6685030 had to be reverted due to windows try failures.
BUG=70957
TEST=about:histograms/Net.HttpTimeToFirstByte
Review URL: http://codereview.chromium.org/6708061
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78979 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
| |
BUG=None
TEST=None
TBR=rvargas@chromium.org
Review URL: http://codereview.chromium.org/6715024
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78977 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
| |
Also clean up net:: use in URLRequestJob while I'm there.
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/6709050
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78902 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Net.HttpTimeToFirstByte
BUG=70957
TEST=about:histograms/Net.HttpTimeToFirstByte
Review URL: http://codereview.chromium.org/6685030
TBR=gavinp@chromium.org
Review URL: http://codereview.chromium.org/6712060
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78886 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
BUG=70957
TEST=about:histograms/Net.HttpTimeToFirstByte
Review URL: http://codereview.chromium.org/6685030
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78883 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
| |
BUG=none
TEST=none
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78878 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
| |
BUG=none
TEST=none
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78877 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
(followup on r78478)
BUG=none
TEST=X509CertificateTest.SerialNumbers
Review URL: http://codereview.chromium.org/6708027
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78869 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
In theory we should be able to parse localized listings
in more languages now.
This also allows us to get rid of hardcoded special cases
for German.
BUG=65917
TEST=net_unittests
Review URL: http://codereview.chromium.org/6697021
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78826 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
histogram initialization across threads.
BUG=76092
TEST=histogram unit tests. Sync tests
R=jar
Review URL: http://codereview.chromium.org/6712040
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78809 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/6709049
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78806 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is a step towards not refcounting URLRequestJobs.
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/6713054
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78803 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
BUG=None
TEST=Unittests.
Review URL: http://codereview.chromium.org/6676027
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78780 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
| |
BUG=none
TEST=none
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78751 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
I made a mistake in r78478: one of the serial numbers was listed twice.
BUG=none
TEST=none
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78748 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
They would hit the backup socket timer. This fixes that.
I've also removed the test exclusions added to work around the flakiness.
BUG=76592
TEST=net_unittests isn't flaky
Review URL: http://codereview.chromium.org/6714013
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78720 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
| |
TEST=none
BUG=none
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78690 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
| |
seeding the initial connection is always desirable.
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/6712021
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78670 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Create a wrapper for URLRequestHttpJob that subclasses FilterContext,
and pass an instance of this facade to Filter::Factory().
Reduce the scope of URLRequestJob's interface as much as possible, moving
methods into URLRequestHttpJob or its facade as appropriate, and making
methods that remain in URLRequestJob non-virtual where possible.
BUG=none
TEST=net_unittests,try bots
Review URL: http://codereview.chromium.org/6677104
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78576 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
| |
Memory and disk cache use the same set of events, with the same parameters (Though the disk cache has a couple events the memory cache does not). Most disk cache events were renamed so as to no longer imply a connection to the disk cache, and all disk cache-related NetLog parameter class definitions were moved to a new file, since they're shared by both entry type.
BUG=59382
TEST=none
Review URL: http://codereview.chromium.org/6613027
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78572 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
| |
BUG=76241
Review URL: http://codereview.chromium.org/6685064
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78570 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
Requested by: Paul Gibson
BUG=none
TEST=none
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78569 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
| |
BUG=none
TEST=none
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78568 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
in guard against null values.
We are no longer getting any new information from more minidumps for
this particular crash, so the aides are no longer necessary and I'm
leaving the guard in place to minimize the impact on our users.
BUG=71721
TEST=net_unittests
Review URL: http://codereview.chromium.org/6698033
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78552 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
BackoffEntry. Simplifying the logic slightly while I'm there,
removing special support for "after the fact" malformed
bodies (the error count doesn't need to be 100% accurate) and
removing a constant value added to back-off times (it was only
being added once anyway, had close to zero effect).
Modifying URLRequestThrottlerEntry and related tests to use the
new class instead of co-mingling sliding window logic and
exponential back-off logic.
Removing now-unnecessary StressTest and associated ugly
wart "SetEntryLifetimeMsForTest" method on the
URLRequestThrottlerEntryInterface class.
Fixing up a few minor things e.g. #pragma once while I'm in
there.
BUG=none
TEST=net_unittests.exe, unit_tests.exe
Review URL: http://codereview.chromium.org/6697001
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78548 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Tests for NE(OK..) rather than EQ(ERR_CERT_INVALID...).
Don't test for CERT_STATUS_INVALID.
Reason: openssl uses CERT_STATUS_AUTHORITY_INVALID for this cert.
(in fact, it now uses the same errors conditions as UnoSoftCertParsing).
BUG=none
TEST=X509CertificateTest.InvalidKeyUsage passes with use_openssl=1
Review URL: http://codereview.chromium.org/6665024
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78531 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
In order to allow this, provide a new factory method in Filter to return a
GZipFilter (used by URLRequestFileJob and URLRequestJobTrackerTest), and
invert control so that URLRequestJobs are responsible for constructing filters.
This is one step away from removing FilterContext as a base class of URLRequestJob,
which will be tackled in a followup change.
BUG=none
TEST=net_unittests
Review URL: http://codereview.chromium.org/6674042
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78489 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
| |
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/6670065
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78478 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
| |
It's apparently no longer used.
R=eroman@chromium.org
BUG=none
TEST=try bots
Review URL: http://codereview.chromium.org/6672036
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78417 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
where it's used.
FilterContext is only accessed by Filter's static Factory() and FixupEncodingTypes() methods,
and GZipFilter doesn't access it at all, so there's no need for the base class to hold a
reference to it. Instead, SdchFilter holds onto the reference itself, as it's the main consumer.
I argue that this makes it more clear which code has access to the FilterContext.
R=jar@chromium.org
BUG=none
TEST=net_unittests
Review URL: http://codereview.chromium.org/6670035
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78414 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
benchmark harness can force SPDY, but still report to the benchmark harness
itself in non-spdy mode.
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/6674031
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78400 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
| |
I don't have a reproduction case for hitting this.
BUG=75049
TEST=none
Review URL: http://codereview.chromium.org/6673083
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78398 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Removed most of the legacy XMPP notification code.
Simplified notification structs and calls a bit.
BUG=76130
TEST=
Review URL: http://codereview.chromium.org/6693007
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78385 0039d316-1c4b-4281-b951-d872f2087c98
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- Remove unused or unneeded methods.
- Make (almost) all data private.
R=eroman@chromium.org
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/6697035
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@78319 0039d316-1c4b-4281-b951-d872f2087c98
|