mirrors/dragonfly
1
0
Fork 0
mirror of https://github.com/dragonflydb/dragonfly.git synced 2025-05-11 10:25:47 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions 5

chore: Fix test_rss_used_mem_gap for all types (#4254)

Browse source 
* chore: Fix `test_rss_used_mem_gap` for all types

The test fails when it checks the gap between `used_memory` and
`object_used_memory`, by assuming that all used memory is consumed by
the `type` it `DEBUG POPULATE`s with.

This assumption is wrong, because there are other overheads, for example
the dash table and string keys.

The test failed for types `STRING` and `LIST` because they used a larger
number of keys as part of the test parameters, which added a larger
overhead.

I fixed the parameters such that all types use the same number of keys,
and also the same number of elements, modifying only the element sizes
(except for `STRING` which doesn't have sub-elements) so that the
overall `min_rss` requirement of 3.5gb still passes.

Fixes #3723

* threshold

* list

* comments test assert

* previous numbers

* ???
This commit is contained in:
Shahar Mike 2024-12-08 13:34:59 +02:00 • committed by GitHub
parent 32ad00b09a
commit bafd8b3b8b
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 7 additions and 6 deletions
Download patch file Download diff file Expand all files Collapse all files

13
tests/dragonfly/memory_test.py
View file

@ -63,7 +63,7 @@ async def test_rss_used_mem_gap(df_server: DflyInstance, type, keys, val_size, e
client = df_server.client() client = df_server.client()
await asyncio.sleep(1) # Wait for another RSS heartbeat update in Dragonfly await asyncio.sleep(1) # Wait for another RSS heartbeat update in Dragonfly
cmd = f"DEBUG POPULATE {keys} {type} {val_size} RAND TYPE {type} ELEMENTS {elements}" cmd = f"DEBUG POPULATE {keys} k {val_size} RAND TYPE {type} ELEMENTS {elements}"
print(f"Running {cmd}") print(f"Running {cmd}")
await client.execute_command(cmd) await client.execute_command(cmd)
@ -76,11 +76,12 @@ async def test_rss_used_mem_gap(df_server: DflyInstance, type, keys, val_size, e
# It could be the case that the machine is configured to use swap if this assertion fails # It could be the case that the machine is configured to use swap if this assertion fails
assert delta > 0 assert delta > 0
assert delta < max_unaccounted assert delta < max_unaccounted
delta = info["used_memory_rss"] - info["object_used_memory"]
# TODO investigate why it fails on string if type != "STRING" and type != "JSON":
if type == "JSON" or type == "STREAM": # STRINGs keep some of the data inline, so not all of it is accounted in object_used_memory
assert delta > 0 # We have a very small over-accounting bug in JSON
assert delta < max_unaccounted assert info["object_used_memory"] > keys * elements * val_size
assert info["used_memory"] > info["object_used_memory"]
@pytest.mark.asyncio @pytest.mark.asyncio