Re: [tor-bugs] #19999 [Core Tor/Tor]: Maybe test-cases should complete without BUG warnings?

[Tor Bug Tracker & Wiki]

#1: Maybe test-cases should complete without BUG warnings?
-+-
 Reporter:  nickm|  Owner:  nickm
 Type:  defect   | Status:  closed
 Priority:  High |  Milestone:  Tor:
 |  0.3.0.x-final
Component:  Core Tor/Tor |Version:
 Severity:  Normal   | Resolution:
 Keywords:  testing, TorCoreTeam201609, review-  |  implemented
  group-10   |  Actual Points:  3
Parent ID:   | Points:  2
 Reviewer:   |Sponsor:
-+-
Changes (by nickm):

 * status:  new => closed
 * resolution:   => implemented


--
Ticket URL: 
Tor Bug Tracker & Wiki 
The Tor Project: anonymity online
___
tor-bugs mailing list
tor-bugs@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs

Re: [tor-bugs] #19999 [Core Tor/Tor]: Maybe test-cases should complete without BUG warnings?

[Tor Bug Tracker & Wiki]

#1: Maybe test-cases should complete without BUG warnings?
-+-
 Reporter:  nickm|  Owner:  nickm
 Type:  defect   | Status:  new
 Priority:  High |  Milestone:  Tor:
 |  0.3.0.x-final
Component:  Core Tor/Tor |Version:
 Severity:  Normal   | Resolution:
 Keywords:  testing, TorCoreTeam201609, review-  |  Actual Points:  3
  group-10   |
Parent ID:   | Points:  2
 Reviewer:   |Sponsor:
-+-
Changes (by nickm):

 * status:  needs_review => new
 * milestone:  Tor: 0.2.9.x-final => Tor: 0.3.0.x-final


Comment:

 rubiate: merged your patch as baa6be9fe7fb.  Thanks!

--
Ticket URL: 
Tor Bug Tracker & Wiki 
The Tor Project: anonymity online
___
tor-bugs mailing list
tor-bugs@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs

Re: [tor-bugs] #19999 [Core Tor/Tor]: Maybe test-cases should complete without BUG warnings?

[Tor Bug Tracker & Wiki]

#1: Maybe test-cases should complete without BUG warnings?
---+---
 Reporter:  nickm  |  Owner:  nickm
 Type:  defect | Status:  needs_review
 Priority:  High   |  Milestone:  Tor:
   |  0.2.9.x-final
Component:  Core Tor/Tor   |Version:
 Severity:  Normal | Resolution:
 Keywords:  testing TorCoreTeam201609  |  Actual Points:  3
Parent ID: | Points:  2
 Reviewer: |Sponsor:
---+---

Comment (by rubiate):

 Tangentially related; addr/ip6_helpers will usually crash if run with
 --verbose

 The loop which builds the string in src/test/test_addr.c:88 never gets run
 because it declares `ii_=0` but uses `i` as the condition, and `i` is
 already incremented to 16 on test_addr.c:278

 Possible fix: (using `j` here instead of `ii_` to avoid touching every
 line realigning the backslashes)

 {{{
 diff --git a/src/test/test_addr.c b/src/test/test_addr.c
 index c8a9e6d..1d91194 100644
 --- a/src/test/test_addr.c
 +++ b/src/test/test_addr.c
 @@ -85,10 +85,10 @@ test_addr_basic(void *arg)
  char *, "%s",\
  { char *cp;  \
cp = print_ = tor_malloc(64);  \
 -  for (int ii_=0;i<16;++i) { \
 -tor_snprintf(cp, 3,"%02x", (unsigned)value_->s6_addr[i]);\
 +  for (int j=0;j<16;++j) {   \
 +tor_snprintf(cp, 3,"%02x", (unsigned)value_->s6_addr[j]);\
  cp += 2; \
 -if (ii_ != 15) *cp++ = ':';  \
 +if (j != 15) *cp++ = ':';\
}  \
  },   \
  { tor_free(print_); },   \
 }}}

--
Ticket URL: 
Tor Bug Tracker & Wiki 
The Tor Project: anonymity online
___
tor-bugs mailing list
tor-bugs@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs

Re: [tor-bugs] #19999 [Core Tor/Tor]: Maybe test-cases should complete without BUG warnings?

[Tor Bug Tracker & Wiki]

#1: Maybe test-cases should complete without BUG warnings?
---+---
 Reporter:  nickm  |  Owner:  nickm
 Type:  defect | Status:  needs_review
 Priority:  High   |  Milestone:  Tor:
   |  0.2.9.x-final
Component:  Core Tor/Tor   |Version:
 Severity:  Normal | Resolution:
 Keywords:  testing TorCoreTeam201609  |  Actual Points:  3
Parent ID: | Points:  2
 Reviewer: |Sponsor:
---+---
Changes (by nickm):

 * status:  reopened => needs_review


--
Ticket URL: 
Tor Bug Tracker & Wiki 
The Tor Project: anonymity online
___
tor-bugs mailing list
tor-bugs@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs

Re: [tor-bugs] #19999 [Core Tor/Tor]: Maybe test-cases should complete without BUG warnings?

[Tor Bug Tracker & Wiki]

#1: Maybe test-cases should complete without BUG warnings?
---+---
 Reporter:  nickm  |  Owner:  nickm
 Type:  defect | Status:  reopened
 Priority:  High   |  Milestone:  Tor:
   |  0.2.9.x-final
Component:  Core Tor/Tor   |Version:
 Severity:  Normal | Resolution:
 Keywords:  testing TorCoreTeam201609  |  Actual Points:  3
Parent ID: | Points:  2
 Reviewer: |Sponsor:
---+---
Changes (by nickm):

 * status:  closed => reopened
 * resolution:  fixed =>


--
Ticket URL: 
Tor Bug Tracker & Wiki 
The Tor Project: anonymity online
___
tor-bugs mailing list
tor-bugs@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs

Re: [tor-bugs] #19999 [Core Tor/Tor]: Maybe test-cases should complete without BUG warnings?

[Tor Bug Tracker & Wiki]

#1: Maybe test-cases should complete without BUG warnings?
---+---
 Reporter:  nickm  |  Owner:  nickm
 Type:  defect | Status:  closed
 Priority:  High   |  Milestone:  Tor:
   |  0.2.9.x-final
Component:  Core Tor/Tor   |Version:
 Severity:  Normal | Resolution:  fixed
 Keywords:  testing TorCoreTeam201609  |  Actual Points:  3
Parent ID: | Points:  2
 Reviewer: |Sponsor:
---+---

Comment (by rubiate):

 rend_cache/store_v2_desc_as_client will occasionally cough up a warning:

 [warn] rend_cache_store_v2_desc_as_client(): Bug: Couldn't decode
 base32 [scrubbed] for descriptor id. (on Tor 0.2.9.3-alpha-dev
 153ff4f559d9b7ee)

 The problem is on test_rendcache.c:158, it tries to test an incorrect ID
 by incrementing the first character of the base32 ID, which can make it
 invalid base32 depending on what the first character is. The test still
 passes because rend_cache_store_v2_desc_as_client() still fails, just for
 a different reason.

 Possible fix, go down if it can't go up:

 {{{
 diff --git a/src/test/test_rendcache.c b/src/test/test_rendcache.c
 index a5d3f35..baadf44 100644
 --- a/src/test/test_rendcache.c
 +++ b/src/test/test_rendcache.c
 @@ -87,7 +87,7 @@ test_rend_cache_lookup_entry(void *data)
  static void
  test_rend_cache_store_v2_desc_as_client(void *data)
  {
 -  int ret;
 +  int ret, mod;
rend_data_t *mock_rend_query;
char desc_id_base32[REND_DESC_ID_V2_LEN_BASE32 + 1];
rend_cache_entry_t *entry = NULL;
 @@ -155,12 +155,19 @@ test_rend_cache_store_v2_desc_as_client(void *data)
// Test incorrect descriptor ID
rend_cache_init();
mock_rend_query = mock_rend_data(service_id);
 -  desc_id_base32[0]++;
 +
 +  /* avoid incrementing beyond valid base32 characters */
 +  mod = (desc_id_base32[0] == 'z' ||
 + desc_id_base32[0] == 'Z' ||
 + desc_id_base32[0] == '7') ? -1 : 1;
 +
 +  desc_id_base32[0] += mod;
 +
ret = rend_cache_store_v2_desc_as_client(desc_holder->desc_str,
 desc_id_base32,
 mock_rend_query,
 NULL);
tt_int_op(ret, OP_EQ, -1);
 -  desc_id_base32[0]--;
 +  desc_id_base32[0] -= mod;
rend_cache_free_all();

// Test too old descriptor
 }}}

--
Ticket URL: 
Tor Bug Tracker & Wiki 
The Tor Project: anonymity online
___
tor-bugs mailing list
tor-bugs@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs

Re: [tor-bugs] #19999 [Core Tor/Tor]: Maybe test-cases should complete without BUG warnings?

[Tor Bug Tracker & Wiki]

#1: Maybe test-cases should complete without BUG warnings?
---+---
 Reporter:  nickm  |  Owner:  nickm
 Type:  defect | Status:  closed
 Priority:  High   |  Milestone:  Tor:
   |  0.2.9.x-final
Component:  Core Tor/Tor   |Version:
 Severity:  Normal | Resolution:  fixed
 Keywords:  testing TorCoreTeam201609  |  Actual Points:  3
Parent ID: | Points:  2
 Reviewer: |Sponsor:
---+---

Comment (by nickm):

 Thanks!  I thought I'd caught all of those...

--
Ticket URL: 
Tor Bug Tracker & Wiki 
The Tor Project: anonymity online
___
tor-bugs mailing list
tor-bugs@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs

Re: [tor-bugs] #19999 [Core Tor/Tor]: Maybe test-cases should complete without BUG warnings?

[Tor Bug Tracker & Wiki]

#1: Maybe test-cases should complete without BUG warnings?
---+---
 Reporter:  nickm  |  Owner:  nickm
 Type:  defect | Status:  closed
 Priority:  High   |  Milestone:  Tor:
   |  0.2.9.x-final
Component:  Core Tor/Tor   |Version:
 Severity:  Normal | Resolution:  fixed
 Keywords:  testing TorCoreTeam201609  |  Actual Points:  3
Parent ID: | Points:  2
 Reviewer: |Sponsor:
---+---

Comment (by rubiate):

 If util/time fails it doesn't call teardown_capture_of_logs()


 {{{
 diff --git a/src/test/test_util.c b/src/test/test_util.c
 index 5949eb9..224ec7b 100644
 --- a/src/test/test_util.c
 +++ b/src/test/test_util.c
 @@ -262,7 +262,6 @@ test_util_time(void *arg)
time_t t_res;
int i;
struct timeval tv;
 -  int old_log_level = 0;

/* Test tv_udiff and tv_mdiff */

 @@ -1112,8 +,7 @@ test_util_time(void *arg)
  #undef CHECK_TIMEGM_ARG_OUT_OF_RANGE

   done:
 -  if (old_log_level)
 -teardown_capture_of_logs();
 +  teardown_capture_of_logs();
  }

  static void
 }}}


 Similarly for tortls:

 {{{
 diff --git a/src/test/test_tortls.c b/src/test/test_tortls.c
 index 790c331..8502e8a 100644
 --- a/src/test/test_tortls.c
 +++ b/src/test/test_tortls.c
 @@ -2278,7 +2278,6 @@ test_tortls_finish_handshake(void *ignored)

X509 *c1 = read_cert_from(validCertString);
SESS_CERT_local *sess = NULL;
 -  int log_level = 0;

ctx = SSL_CTX_new(method);

 @@ -2298,7 +2297,6 @@ test_tortls_finish_handshake(void *ignored)
expect_single_log_msg_containing("For some reason, wasV2Handshake
 didn't "
 "get set.");
teardown_capture_of_logs();
 -  log_level = 0;

tls->wasV2Handshake = 1;
ret = tor_tls_finish_handshake(tls);
 @@ -2337,8 +2335,7 @@ test_tortls_finish_handshake(void *ignored)
tor_free(tls);
SSL_CTX_free(ctx);
tor_free(method);
 -  if (log_level)
 -teardown_capture_of_logs();
 +  teardown_capture_of_logs();
  }
  #endif
 }}}

--
Ticket URL: 
Tor Bug Tracker & Wiki 
The Tor Project: anonymity online
___
tor-bugs mailing list
tor-bugs@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs

Re: [tor-bugs] #19999 [Core Tor/Tor]: Maybe test-cases should complete without BUG warnings?

[Tor Bug Tracker & Wiki]

#1: Maybe test-cases should complete without BUG warnings?
---+---
 Reporter:  nickm  |  Owner:  nickm
 Type:  defect | Status:  closed
 Priority:  High   |  Milestone:  Tor:
   |  0.2.9.x-final
Component:  Core Tor/Tor   |Version:
 Severity:  Normal | Resolution:  fixed
 Keywords:  testing TorCoreTeam201609  |  Actual Points:  3
Parent ID: | Points:  2
 Reviewer: |Sponsor:
---+---
Changes (by nickm):

 * status:  accepted => closed
 * resolution:   => fixed


--
Ticket URL: 
Tor Bug Tracker & Wiki 
The Tor Project: anonymity online
___
tor-bugs mailing list
tor-bugs@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs

Re: [tor-bugs] #19999 [Core Tor/Tor]: Maybe test-cases should complete without BUG warnings?

[Tor Bug Tracker & Wiki]

#1: Maybe test-cases should complete without BUG warnings?
---+---
 Reporter:  nickm  |  Owner:  nickm
 Type:  defect | Status:  accepted
 Priority:  High   |  Milestone:  Tor:
   |  0.2.9.x-final
Component:  Core Tor/Tor   |Version:
 Severity:  Normal | Resolution:
 Keywords:  testing TorCoreTeam201609  |  Actual Points:  3
Parent ID: | Points:  2
 Reviewer: |Sponsor:
---+---
Changes (by nickm):

 * keywords:  testing => testing TorCoreTeam201609
 * owner:   => nickm
 * status:  new => accepted
 * actualpoints:   => 3


Comment:

 Down to 0

--
Ticket URL: 
Tor Bug Tracker & Wiki 
The Tor Project: anonymity online
___
tor-bugs mailing list
tor-bugs@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs

Re: [tor-bugs] #19999 [Core Tor/Tor]: Maybe test-cases should complete without BUG warnings?

[Tor Bug Tracker & Wiki]

#1: Maybe test-cases should complete without BUG warnings?
--+
 Reporter:  nickm |  Owner:
 Type:  defect| Status:  new
 Priority:  High  |  Milestone:  Tor: 0.2.9.x-final
Component:  Core Tor/Tor  |Version:
 Severity:  Normal| Resolution:
 Keywords:  testing   |  Actual Points:
Parent ID:| Points:  2
 Reviewer:|Sponsor:
--+

Comment (by nickm):

 Down to 17.

 Also, as of 3269307dafafa8c7c2c3e0be5d5c3cb5e7df3153 , it counts as a test
 failure any time that tor_assert_nonfatal() fails.

--
Ticket URL: 
Tor Bug Tracker & Wiki 
The Tor Project: anonymity online
___
tor-bugs mailing list
tor-bugs@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs

Re: [tor-bugs] #19999 [Core Tor/Tor]: Maybe test-cases should complete without BUG warnings?

[Tor Bug Tracker & Wiki]

#1: Maybe test-cases should complete without BUG warnings?
--+
 Reporter:  nickm |  Owner:
 Type:  defect| Status:  new
 Priority:  High  |  Milestone:  Tor: 0.2.9.x-final
Component:  Core Tor/Tor  |Version:
 Severity:  Normal| Resolution:
 Keywords:  testing   |  Actual Points:
Parent ID:| Points:  2
 Reviewer:|Sponsor:
--+

Comment (by nickm):

 Now we've got a huge pile of warnings to fix.  By my count, roughly 400
 lines with "Bug: " in them.

--
Ticket URL: 
Tor Bug Tracker & Wiki 
The Tor Project: anonymity online
___
tor-bugs mailing list
tor-bugs@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs

Re: [tor-bugs] #19999 [Core Tor/Tor]: Maybe test-cases should complete without BUG warnings?

[Tor Bug Tracker & Wiki]

#1: Maybe test-cases should complete without BUG warnings?
--+
 Reporter:  nickm |  Owner:
 Type:  defect| Status:  new
 Priority:  High  |  Milestone:  Tor: 0.2.9.x-final
Component:  Core Tor/Tor  |Version:
 Severity:  Normal| Resolution:
 Keywords:  testing   |  Actual Points:
Parent ID:| Points:  2
 Reviewer:|Sponsor:
--+

Comment (by nickm):

 My #20042 branch makes it harder to accidentally suppress log messages,
 and makes BUG messages on-by-default.  While doing that, I found #20041,
 where running the tests at --debug made one fail. (!)

--
Ticket URL: 
Tor Bug Tracker & Wiki 
The Tor Project: anonymity online
___
tor-bugs mailing list
tor-bugs@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs

Re: [tor-bugs] #19999 [Core Tor/Tor]: Maybe test-cases should complete without BUG warnings?

[Tor Bug Tracker & Wiki]

#1: Maybe test-cases should complete without BUG warnings?
--+
 Reporter:  nickm |  Owner:
 Type:  defect| Status:  new
 Priority:  High  |  Milestone:  Tor: 0.2.9.x-final
Component:  Core Tor/Tor  |Version:
 Severity:  Normal| Resolution:
 Keywords:  testing   |  Actual Points:
Parent ID:| Points:  2
 Reviewer:|Sponsor:
--+

Comment (by nickm):

 We could at least enforce that no BUG() or tor_assert_nonfatal()
 violations happen during unit tests.

--
Ticket URL: 
Tor Bug Tracker & Wiki 
The Tor Project: anonymity online
___
tor-bugs mailing list
tor-bugs@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs

[tor-bugs] #19999 [Core Tor/Tor]: Maybe test-cases should complete without BUG warnings?

[Tor Bug Tracker & Wiki]

#1: Maybe test-cases should complete without BUG warnings?
--+
 Reporter:  nickm |  Owner:
 Type:  defect| Status:  new
 Priority:  High  |  Milestone:  Tor: 0.2.9.x-final
Component:  Core Tor/Tor  |Version:
 Severity:  Normal|   Keywords:  testing
Actual Points:|  Parent ID:
   Points:  2 |   Reviewer:
  Sponsor:|
--+
 If you run the unit tests with the --warn option, you fill find a lot of
 warnings.  We deliberately test warning-paths, so that's not too bad...


 ... but you'll also see a lot of "Bug" warnings, and that's not so great.
 Some of them even cause stack traces.

 I think we should go over these before 0.2.9 finishes, too make sure they
 aren't causing actual bugs.

--
Ticket URL: 
Tor Bug Tracker & Wiki 
The Tor Project: anonymity online
___
tor-bugs mailing list
tor-bugs@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs