ff42682c41
PoC for bugsink-server-unified
2024-08-27 21:04:16 +02:00
d3029cbb70
Revert development.py stuff done for mysql poc
...
i.e. in 32f87457d0
2024-08-27 11:10:38 +02:00
24fcc7c2d8
Add version tags to Docker image
2024-08-27 10:14:09 +02:00
e4174d1e64
mysqlclient out of requirements.txt
...
just install it where we need it, namely in the Docker image
2024-08-27 09:50:31 +02:00
f20104770e
Docker from wheel, not sourcecode
...
this way we reuse all the thinking we did when getting the Python wheels in order.
* what to include/exclude
* setting up the scripts
2024-08-27 09:48:39 +02:00
200ea5eb84
Docker image: Ubuntu -> python-slim
...
* I'd rather harmonize with "the docker way" than with my own recommended setup
* slim: expectation is that this is smaller (not checked)
* this is a 2-stage build (slim has no gcc, needed for mysql)
* PEP668 does not apply here ("externally managed"), which leads to
* better ergonomics: no env/bin/some-command
2024-08-27 09:04:42 +02:00
d00d007306
Dockerfile(s): the afternoon's progress
2024-08-26 20:37:55 +02:00
32f87457d0
MySQL PoC
2024-08-26 11:22:58 +02:00
00af16b747
gitignore update for package name-change
2024-08-26 09:20:36 +02:00
3b5084edf6
pytoml project name lowercase
...
makes for wheel/sdist consistency
2024-08-22 10:41:05 +02:00
70fdbc9cbd
Sentry DSN setup: recommend what we do ourselves
2024-08-22 10:03:28 +02:00
b9292291b5
Remove auto_session_tracking=False from sentry-init
...
The default is True.
It was copy-pasted from earlier integrations, but its exact origin (and
whether it is strictly needed) is unclear. Perhaps installation instructions
from (some) Sentry or Glitchtip?
We don't have explicit support for sessions, but I'm not clear enough on this
one to recommend against it either. And for our own usage, this provides a
way to get samples of session data in the playground/dogfood.
2024-08-22 09:41:39 +02:00
ebd0a612d9
Flake8 and f-strings whitespace (E201, E202)
2024-08-21 10:01:38 +02:00
6caa57e0b0
Github workflow typo
2024-08-21 09:39:56 +02:00
6021af9ff3
Fix github workflow names
2024-08-21 09:39:27 +02:00
dd968d7d43
Flake8 Github Workflow
2024-08-21 09:37:01 +02:00
129a8db421
Fix various flake8 errors
2024-08-21 09:31:05 +02:00
7ac78fe922
Close files in test (-Wall warning fix)
2024-08-21 09:03:57 +02:00
f7972cbec0
use datetime.timezone.utc
...
RemovedInDjango50Warning: The django.utils.timezone.utc alias is deprecated. Please update your code to use datetime.timezone.utc instead.
2024-08-21 09:01:20 +02:00
5968e88c67
Bump github setup-python to v5
...
might solve the following:
> The following actions use a deprecated Node.js version and will be forced to run on node20: actions/setup-python@v3. For more info: https://github.blog/changelog/2024-03-07-github-actions-all-actions-will-run-on-node20-instead-of-node16-by-default/
2024-08-21 08:53:20 +02:00
1bfac5d8c6
assertEquals -> assertEual (Python 3.12)
...
<<insert remarks about fashion police>>
2024-08-21 08:49:49 +02:00
ee8ccdaeb4
Github Actions: try with strings to avoid '3.1' error
2024-08-21 08:41:29 +02:00
98a4973de0
Create Action tests.yml
...
Using Github's template for Django; adopting Python versions with currently supported ones
2024-08-21 08:39:23 +02:00
03f0a2f673
Add README
2024-08-21 08:30:46 +02:00
6560767764
Tailwind: Explicitly point to python file containing class
2024-08-19 22:12:20 +02:00
ff48ffb977
Query counts for DEBUG=False contexts (i.e. our own 'playground' etc)
2024-08-19 15:14:24 +02:00
9aaec10f95
Fix in_app display for 'playground' install
...
(also done on sparrow)
2024-08-19 14:49:56 +02:00
01839a887d
try to send more info when debugging my own install
...
(I've simultaneously updated 'playground' to do the same)
2024-08-19 14:27:12 +02:00
d6c61c25bb
When resolving, show history as 'resolved' (bugfix)
2024-08-19 11:43:14 +02:00
37b11e16ec
ALLOWED_HOSTS: use {{ host }}
...
as per what's documented (commented out, i.e. privately) in the website:
getting nginx to play nicely for 443/default_server is 'hardish'
2024-08-19 10:56:00 +02:00
63417d555f
Explain why we deal with SIGTERM as we do
...
(from memory, in response to the glib remarks in b09cfb21c3
2024-07-26 16:22:30 +02:00
22fdd8adae
Docs moved to website
2024-07-26 15:37:16 +02:00
8cee902db4
WHITENOISE_USE_FINDERS to avoid collectstatic
2024-07-26 15:28:33 +02:00
b76e474ef1
Navigation: fix for missing events
...
Now that we have eviction, events may disappear. Deal with it:
* event-specific 404 that still allows for navigation
* first/last buttons
* navigation to prev/next when prev/next is not just 1 step away
* don't use HttpRedirect for "lookup based" views
in principle, the thing you were looking for might go missing in-between
drawback: these URLs are not "stable"
2024-07-19 11:03:08 +02:00
aec78f6318
Clarify a docstring
2024-07-18 15:13:08 +02:00
c5dc3014ea
Migrations: don't limit queries in runtime
2024-07-18 15:07:14 +02:00
63cfbb2acd
Hide version info in the HTML source
...
so that we at least can quickly inspect it
2024-07-18 14:59:01 +02:00
3128392d9a
Distinguish ingested_at and digested_at
2024-07-18 14:45:59 +02:00
d23f1f0a3b
Remove load_performance_fixture
...
in the belief that it was only (mostly?) useful in the context of
the pc-registry performance tests. (We do our other tests end-to-end
using the stress tests)
2024-07-18 13:17:45 +02:00
717a632b7d
check_for_thresholds refactoring: 'metadata' is superfluous
...
because it was basically the input-tuple (in a different format)
2024-07-18 09:43:37 +02:00
d731dee4f6
Note on stress testing
2024-07-18 09:34:32 +02:00
e40e652722
Remove duplicate test-factory function
2024-07-18 09:33:50 +02:00
b211ba4c1e
Document possible way forward for counting all ingested events
2024-07-18 09:21:59 +02:00
f48c48f7e5
Implement 429 for the deprecated 'store' endpoint too
2024-07-18 09:19:37 +02:00
36f74acb2d
Document performance surprise
2024-07-18 09:18:11 +02:00
927587c132
Stress test interuptible, still show results
2024-07-17 17:33:30 +02:00
ec0877edb7
Document yet another problem with 'real streaming' and Nginx
2024-07-17 17:13:30 +02:00
65ea181f37
vbc-unmute: reduce calls to the expensive check
...
as done in the previous commit for project quota
2024-07-17 15:33:15 +02:00
51a53c09a4
quota: check as little as possible & check-on-digest
...
Also fix various off-by-one errors with the help of tests
2024-07-17 14:48:19 +02:00
8849a3e44b
Don't write to the DB on-ingest
...
In the previous commit I put the code for a small performance-experiment.
The results are (very) obvious: don't do this. Response times go through
the roof, and more importantly, the server becomes unreliable. Reason:
time-outs caused by waiting for the write-lock.
2024-07-16 16:39:12 +02:00
Klaas van Schelven