Closed fjarri closed 3 years ago
Presumably, you've run it locally with a nucypher version containing #2574?
I did, but to be on the safe side someone else should too.
Ran it locally with relocked dependencies (pointed to nucypher
on main
), and it seemed to run fine. The scraping round took some time (9-10mins), but I usually find that it takes longer on my local machine, and ends up being much faster on the server (~ 1-2mins).
Current status: waiting for the new release with merged https://github.com/nucypher/nucypher/pull/2574 so that CI could run
With nucypher
updated to 4.8, tests pass now.
Merging #94 (79ebef3) into main (f8df51a) will decrease coverage by
1.00%
. The diff coverage is75.80%
.
@@ Coverage Diff @@
## main #94 +/- ##
==========================================
- Coverage 51.89% 50.89% -1.01%
==========================================
Files 11 11
Lines 1079 1059 -20
==========================================
- Hits 560 539 -21
- Misses 519 520 +1
Impacted Files | Coverage Δ | |
---|---|---|
monitor/cli/_utils.py | 40.00% <0.00%> (ø) |
|
monitor/crawler.py | 43.65% <75.00%> (-3.09%) |
:arrow_down: |
monitor/db.py | 95.69% <100.00%> (ø) |
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact)
,ø = not affected
,? = missing data
Powered by Codecov. Last update f8df51a...79ebef3. Read the comment docs.
Actually the dependency relock can probably be rolled back to minimize the changes. nucypher 4.8 will be installed anyway.
In https://github.com/nucypher/nucypher/pull/2352
FleetSensor
and node metadata API changed. This PR updates the calls in the monitor. Needs https://github.com/nucypher/nucypher/pull/2574 to work, so the CI will likely fail...There is also a change in how crawler storage is implemented. Before it was derived from
ForgetfulNodeStorage
, which presented several problems:ForgetfulNodeStorage
stores node metadata, so the hierarchical relation between them is questionableIn this PR,
CrawlerStorage
is its own class, and its calls are hooked toFleetSensor
updates (seehooked_tracker_class()
).Important: One consequence of this is that we now do not store nodes right away before they are even verified; now the nodes are stored (or removed) when the fleet state changes. This can be modified, but currently I am unclear as to what behavior was really expected from the crawler, so I went with the logical one.