test: Create cached blocks not in the future #18641

pull MarcoFalke wants to merge 1 commits into bitcoin:master from MarcoFalke:2004-qaNoCacheFuture changing 1 files +7 −3

MarcoFalke commented at 1:08 AM on April 15, 2020: member

This avoids test failures when tests assume blocks are not from the future, like in wallet_dump: https://cirrus-ci.com/task/6607130193035264?command=ci#L3306
test: Create cached blocks not in the future fa32097541
fanquake added the label Tests on Apr 15, 2020
MarcoFalke commented at 1:50 AM on April 15, 2020: member

Cirrus looks green now: https://cirrus-ci.com/task/4872526717255680
jonatack commented at 4:35 PM on April 17, 2020: member

What caused the failures to begin occurring?
MarcoFalke commented at 4:58 PM on April 17, 2020: member

A newly added test in wallet_dump.py
jonatack commented at 5:00 PM on April 17, 2020: member

Ok -- 555567ac. Looking, and running all the tests locally against this right now.

in test/functional/test_framework/test_framework.py:572 in fa32097541

 569 |                      nblocks=25 if i != 7 else 24,
 570 |                      address=TestNode.PRIV_KEYS[i % 4].address,
 571 |                  )
 572 |  
 573 | -            assert_equal(self.nodes[CACHE_NODE_ID].getblockchaininfo()["blocks"], 199)
 574 | +            assert_equal(cache_node.getblockchaininfo()["blocks"], 199)

jonatack commented at 5:20 PM on April 17, 2020:

while touching this, maybe simplify?

-            assert_equal(self.nodes[CACHE_NODE_ID].getblockchaininfo()["blocks"], 199)
+            assert_equal(self.nodes[CACHE_NODE_ID].getblockcount(), 199)

huanhd400 commented at 5:38 PM on April 17, 2020:

jonatack commented at 5:25 PM on April 17, 2020: member

I'm tentatively seeing failures in wallet_multiwallet with and without --usecli, not sure if related... rebuilding with make clean to be sure.
MarcoFalke commented at 5:29 PM on April 17, 2020: member

Did you run make distclean && make before executing the tests?
jonatack commented at 5:35 PM on April 17, 2020: member

ACK fa320975411af4f0e41771d89958a77fd7a2284b
jonatack commented at 5:38 PM on April 17, 2020: member

Yep, sorry, I needed to make clean.

I've been reusing my ./configure, then redoing with make clean if seeing issues, or starting straight away with make distclean when switching between gcc and clang. Do you generally begin with make distclean?

MarcoFalke commented at 5:45 PM on April 17, 2020: member

This is my bash alias:

./autogen.sh && ./configure && make distclean && ./configure CC=clang CXX=clang++ && make -j 9 check

jonatack commented at 5:54 PM on April 17, 2020: member

Thanks... ./configure twice? Mine are here: https://github.com/jonatack/dotfiles/blob/master/bash_aliases#L331
MarcoFalke commented at 5:57 PM on April 17, 2020: member

Jup, this way it works on any repo state, even after git clean -dfx

jonatack commented at 6:05 PM on April 17, 2020: member

Thanks!

+alias marco='./autogen.sh && ./configure && make distclean && ./configure CC=clang CXX=clang++ && make -j 5 check'

MarcoFalke commented at 6:06 PM on April 17, 2020: member

lol
MarcoFalke merged this on Apr 17, 2020
MarcoFalke closed this on Apr 17, 2020
MarcoFalke deleted the branch on Apr 17, 2020
sidhujag referenced this in commit ba32b232d3 on Apr 17, 2020
deadalnix referenced this in commit aff6784475 on May 14, 2020
DrahtBot locked this on Feb 15, 2022

Contributors

Labels