mirrors/git-annex
2
0
Fork 0
Code Issues Pull requests Projects Releases Packages 1 Wiki Activity Actions

fix testremote to not throw away annex state

Browse source 
aeca7c2207 exposed this problem, but it
was never a good idea to have a series of test cases, some of which depend on
prior ones, and throw away annex state after each.
This commit is contained in:
Joey Hess 2020-04-28 17:19:07 -04:00
parent e66f9d1c8c
commit fa98025de0
No known key found for this signature in database
GPG key ID: DB12DB0FF05F8F38
3 changed files with 41 additions and 8 deletions
Download patch file Download diff file Expand all files Collapse all files

22
doc/bugs/testremote_failures_starting_with_aeca7c220/comment_2_d53b98bca623d4d25bab58563e7577b3._comment Normal file
View file

@ -0,0 +1,22 @@
[[!comment format=mdwn
username="joey"
subject="""comment 2"""
date="2020-04-28T20:56:01Z"
content="""
Fairly sure it's something to do with the chunk log. The tests with no
chunks succeed, so the subsequent failures could be due to it failing to
see information about chunks used to store the key when it then tries to
retrieve it.
Aha: It's using Annex.eval, so it does stuff, and throws away the changed
state. So, after writing to the journal, when journalIgnorable = False,
it will then forget that, and will try to read from the branch, ignoring
the journal.
I checked all Annex.eval, and all the rest have the pattern
Annex.eval =<< Annex.new. This is the only one that *reuses*
an old state across multiple Annex.eval. And it's not like the commit that
triggered this made Annex.eval more dangerous; that were never a good thing
to do and probably only didn't break on some other state being lost before
due to luck.
"""]]