Skip to content

fix: gzip trace uploads at level 1 #1129

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
kcoopermiller merged 1 commit into from
Nov 2, 2026
Merged

fix: gzip trace uploads at level 1 #1129

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 4 additions & 3 deletions packages/prime-traces/src/prime_traces/core/async_client.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,6 +11,7 @@
from .client import (
AMBIGUOUS_TRANSPORT_ERRORS,
DEFAULT_TIMEOUT,
GZIP_LEVEL,
IDEMPOTENT_RETRY_ATTEMPTS,
BaseTracesAPIClient,
is_ambiguous_delete_failure,
Expand Down Expand Up @@ -95,9 +96,9 @@ async def upload_batch(
self._check_auth()

# mtime=0 keeps the compressed bytes reproducible across retries.
compressed = (
await asyncio.to_thread(gzip_module.compress, body, mtime=0) if compress else None
)
compressed = None
if compress:
compressed = await asyncio.to_thread(gzip_module.compress, body, GZIP_LEVEL, mtime=0)
files = self._multipart_files(
body, schema_version=schema_version, context=context, compressed=compressed
)
Expand Down
9 changes: 8 additions & 1 deletion packages/prime-traces/src/prime_traces/core/client.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -48,6 +48,11 @@
#: budget — see ``TracesClient._send_with_retry``.
IDEMPOTENT_RETRY_ATTEMPTS = 3

#: Gzip level for upload bodies. Level 1 compresses a multi-MB episode about 9x
#: faster than the default 9 for ~13% more bytes; the upload thread is otherwise
#: CPU-bound on compression and falls behind a fast producer.
GZIP_LEVEL = 1

#: Retryable statuses that leave it unknown whether the request reached the
#: service. 502/504 come from a gateway, which may already have forwarded the
#: request upstream and then lost or timed out the response.
Expand Down Expand Up @@ -379,7 +384,9 @@ def upload_batch(
self._check_auth()

# mtime=0 keeps the compressed bytes reproducible across retries.
compressed = gzip_module.compress(body, mtime=0) if compress else None
compressed = None
if compress:
compressed = gzip_module.compress(body, compresslevel=GZIP_LEVEL, mtime=0)
files = self._multipart_files(
body, schema_version=schema_version, context=context, compressed=compressed
)
Expand Down
Loading