Commit Graph

25078 Commits

Author SHA1 Message Date
Nick Mathewson
0f6420e53a Changes file for dirreq refactoring branch 2017-04-24 12:15:25 -04:00
Nick Mathewson
d949589c09 Fix some leaks in the consdiffmgr tests 2017-04-24 11:45:13 -04:00
Nick Mathewson
e55c1412c1 Fix a signed/unsigned comparison warning on 32-bit 2017-04-24 11:41:11 -04:00
Nick Mathewson
b7567a6282 Merge branch 'consdiffmgr_squashed' 2017-04-24 11:02:22 -04:00
Nick Mathewson
eb14faa0c1 Remove a checklist item that was already tested
The item referred to the cdm_ht_set_status() case where the item was
not already in the hashtable.  But that already happens naturally
when we scan the directory on startup... and we already have a test
for that.
2017-04-24 11:01:40 -04:00
Nick Mathewson
af86895581 consdiffmgr test: do not launch a diff task that is already pending 2017-04-24 11:01:40 -04:00
Nick Mathewson
2e9e2d023b consdiffmgr: tests for consdiffmgr_validate() 2017-04-24 11:01:40 -04:00
Nick Mathewson
6cc21aa89c consdiffmgr: add tests for cdm_entry_get_sha3_value 2017-04-24 11:01:40 -04:00
Nick Mathewson
bb38657b77 consdiffmgr test: add a test for updating ht on clean/rescan.
This brings us back up to ~94% coverage
2017-04-24 11:01:40 -04:00
Nick Mathewson
de0142cd9d Expand diff-management test to cover reloading items from disk 2017-04-24 11:01:40 -04:00
Nick Mathewson
43d683e0ad Fix reference leak & handle leak in consensus_diff_worker_replyfn
Found by previous test.
2017-04-24 11:01:40 -04:00
Nick Mathewson
831e656baa consdiffmgr tests: add tests to validate diff lookup/application
This commit adds some helper functions to look up the diff from one
consensus and to make sure that applying it leads to another.  Then
we add them throughout the existing test cases.  Doing this turned
up a reference-leaking bug in consensus_diff_worker_replyfn.
2017-04-24 11:01:40 -04:00
Nick Mathewson
605bcfbf71 consdiffmgr: Enable in-progress test that was not previously working
Also, add a list of additional tests to write.
2017-04-24 11:01:40 -04:00
Nick Mathewson
7ca86b9cd6 Add a hashtable to consdiffmgr to keep track of diff status
In several places in the old code, we had problems that only an
in-memory index of diff status could solve, including:
   * Remembering which diffs were in-progress, so that we didn't
     re-launch them.
   * Remembering which diffs had failed, so that we didn't try to
     recompute them over and over.
   * Having a fast way to look up the diff from a given consensus to
     the latest consensus of a given flavor.

This patch adds a hashtable mapping from (flavor, source diff), to
solve the problem.  It maps to a cache entry handle, rather than to
a cache entry directly, so that it doesn't affect the reference
counts of the cache entries, and so that we don't otherwise need to
worry about lifetime management.
2017-04-24 11:01:40 -04:00
Nick Mathewson
69a212ff3d Consdiffmgr: extract "get a sha3 digest" function.
I'll be using this a lot in the hashtable tweaks here.
2017-04-24 11:01:40 -04:00
Nick Mathewson
655f1c8e01 consdiffmgr: function to re-validate stored sha3 digests at startup 2017-04-24 11:00:28 -04:00
Nick Mathewson
6c86e63029 Consdiffmgr: use aggressive-release flag on consensuses
This conscache flag tells conscache that it should munmap the
document as soon as reasonably possible, since its usage pattern is
expected to not have a lot of time-locality.
2017-04-24 10:59:25 -04:00
Nick Mathewson
82bb8afb60 Add handle support to consensus_cache_entry_t
This will allow us to have weak references to cache entries.
2017-04-24 10:59:25 -04:00
Nick Mathewson
5726fec9c2 Consdiffmgr test: Make sure that diffs are removable
A diff is removable as soon as it no longer takes you to the most
recent consensus of the appropriate flavor.
2017-04-24 10:59:25 -04:00
Nick Mathewson
35f6b678ab Test the easiest cases of consdiffmgr_cleanup.
One more to go: deleting the old diffs.
2017-04-24 10:59:25 -04:00
Nick Mathewson
1fade37287 consdiffmgr non-test: check for initialization failure
Unfortunately, this test doesn't work, so I've left it
defined-out. There is currently no way in our unit tests to catch a
fatal assertion failure.
2017-04-24 10:59:25 -04:00
Nick Mathewson
d418f28cb5 consdiffmgr test: survive failures to compute a diff. 2017-04-24 10:59:25 -04:00
Nick Mathewson
b9c2f135bd Another consdiffmgr test: only generate the diffs that are needed
This test makes sure that we only generate the diffs we actually
want, rather than regenerating all the diffs every time anything
changes.
2017-04-24 10:59:25 -04:00
Nick Mathewson
7fc37d41b4 Unit tests for consdiffmgr module
Initial tests. These just try adding a few consensuses, looking
them up, and making sure that consensus diffs are generated in a
more or less reasonable-looking way.  It's enough for 87% coverage,
but it leaves out a lot of functionality.
2017-04-24 10:59:25 -04:00
Nick Mathewson
fe584f4012 Make cpuworker_queue_work function mockable.
I'll be using this in the unit tests for consdiffmgr.
2017-04-24 10:59:25 -04:00
Nick Mathewson
bc91808c47 Add a "Consensus diff manager" module.
This module's job is to remember old consensus documents, to
calculate their diffs on demand, and to .

There are some incomplete points in this code; I've marked them with
"XXXX". I intend to fix them in separate commits, since I believe
doing it in separate commits will make the branch easier to review.
2017-04-24 10:59:24 -04:00
Nick Mathewson
9ba10d714a Merge branch 'consdiff_numeric_squashed' 2017-04-24 09:33:55 -04:00
Sebastian Hahn
a16de7a7cf consdiff: Reject ranges with non-numeric chars
Fixes bug #21964
2017-04-24 09:33:48 -04:00
Nick Mathewson
8ff2a8c0e2 Merge remote-tracking branch 'dgoulet/bug22032_031_01' 2017-04-24 09:30:30 -04:00
Taylor Yu
7bc636fdc9 Add regression test for #22304 2017-04-24 09:20:59 -04:00
Nick Mathewson
8b89faf424 Merge branch 'maint-0.3.0' 2017-04-24 09:20:43 -04:00
Nick Mathewson
ae374e0a56 Merge branch 'maint-0.2.9' into maint-0.3.0 2017-04-24 09:20:26 -04:00
David Goulet
4a9f689430 control: Wrong check on base16_decode return value
The GETINFO extra-info/digest/<digest> broke in commit 568dc27a19 that
refactored the base16_decode() API to return the decoded length.
Unfortunately, that if() condition should have checked for the correct length
instead of an error which broke the command in tor-0.2.9.1-alpha.

Fixes #22034

Signed-off-by: David Goulet <dgoulet@torproject.org>
2017-04-24 09:20:13 -04:00
Nick Mathewson
91553417c7 Fix a comment that held the remaining reference to the old api. 2017-04-21 16:39:05 -04:00
Nick Mathewson
d8ff01b2b4 Refactor directory_command_should_use_begindir to use directory_request_t 2017-04-21 15:55:23 -04:00
Nick Mathewson
2a99bf008d move the definition of directory_command_should_use_begindir 2017-04-21 15:54:50 -04:00
Nick Mathewson
cc677e88d4 Refactor directory_send_command to take a directory_request_t 2017-04-21 15:49:10 -04:00
Nick Mathewson
900d36e3b1 Rename, revise, and document directory_request_is_dir_specified. 2017-04-21 15:42:09 -04:00
Nick Mathewson
d7fab36037 Make directory_request_set_guard_state static. 2017-04-21 15:42:09 -04:00
Nick Mathewson
81045e19eb Document the new directory_request_t API and add some assertions 2017-04-21 15:42:03 -04:00
Nick Mathewson
3327e675fd Remove old directory_initiate_command_*() functions. 2017-04-21 15:14:52 -04:00
Nick Mathewson
8aadd60eac Replace remaining directory_initiate_command_* instances 2017-04-21 15:08:28 -04:00
Nick Mathewson
4e393f5318 Make launch_direct_bridge_descriptor_fetch use dirreqs. 2017-04-21 14:47:14 -04:00
Nick Mathewson
4ae65bb5d3 Refactor directory_get_from_dirserver to use request API. 2017-04-21 14:40:48 -04:00
Nick Mathewson
eee733be33 Refactor directory_post_to_dirservers to use requests 2017-04-21 14:35:42 -04:00
Nick Mathewson
c300fd7e65 Have directory_get_from_all_authorities use requests. 2017-04-21 14:31:35 -04:00
Nick Mathewson
a55bd00b0f continue refactoring directory API: remove initiate_command_rend
This commit mainly moves the responsibility for directory request
construction one level higher.  It also allows a directory request
to contain a pointer to a routerstatus, which will get turned into
the correct contact information at the last minute.
2017-04-21 14:18:06 -04:00
David Goulet
74f7657e56 hs: Fix a spurious warn in rend_log_intro_limit
We do dump HS stats now at log info everytime the intro circuit creation retry
period limit has been reached. However, the log was upgraded to warning if we
actually were over the elapsed time (plus an extra slop).

It is actually something that will happen in tor in normal case. For instance,
if the network goes down for 10 minutes then back up again making
have_completed_a_circuit() return false which results in never updating that
retry period marker for a service.

Fixes #22032

Signed-off-by: David Goulet <dgoulet@torproject.org>
2017-04-21 11:09:34 -04:00
Nick Mathewson
b7dd3518b9 missing smartlist_free in new consdiff test 2017-04-18 22:56:34 -04:00
Nick Mathewson
afe1af7de6 Merge remote-tracking branch 'asn/bug21971' 2017-04-18 21:38:28 -04:00